St. Johns Archaeological Field School 2007-10: Silver Glen Run (8LA1)

MISSING IMAGE

Material Information

Title:
St. Johns Archaeological Field School 2007-10: Silver Glen Run (8LA1)
Physical Description:
Technical report
Creator:
Sassaman, Kenneth E.
Gilmore, Zackary I.
Randall, Asa R.
Publisher:
Laboratory of Southeastern Archaeology, Department of Anthropology, University of Florida
Place of Publication:
Gainesville, Fla.
Publication Date:

Record Information

Source Institution:
University of Florida Institutional Repository
Holding Location:
University of Florida
Rights Management:
All rights reserved by the source institution and holding location.
System ID:
AA00009682:00001


This item is only available as the following downloads:


Full Text

PAGE 1

ST. JOHNS ARCHAEOLOGICAL FIELD SCHOOL 2007-2010 SILVER GLEN RUN (8LA1) Kenneth E. Sassaman, Zackary I. Gilmore, and Asa R. Randall Technical Report 12 Laboratory of Southeastern Archaeology Department of Anthropology University of Florida

PAGE 2

ST. JOHNS ARCHAEOLOGICAL FIELD SCHOOL 2007-2010: SILVER GLEN RUN (8LA1) Kenneth E. Sassaman Zackary I. Gilmore Asa R. Randall Technical Report 12 Laboratory of Southeastern Archaeology Department of Anthropology University of Florida Gainesville, FL 32611 December 2011

PAGE 3

2011 Department of Anthropology, University of Florida all rights reserved Cover map of the reconstructed landscape of Silver Glen Run produced by Asa R. Randall ii

PAGE 4

Management Summary Since 2007, faculty and students of the St. Johns Archaeological Field School, Department of Anthropology, University of Florida have been conducting field investigations at pre-Columbian sites along Silver Glen Run in Lake and Marion counties, Florida. The sites are located on property of the Juniper Club of Louisville, Kentucky, hosts of the field school and stewards of an archaeological record spanning thousands of years of intensive occupation. Among the si tes on the property are several shell mounds, at least one sand mound, and innumerable subsurface remains dating from the early Holocene. Field school investigations to date have involved reconn aissance survey of Juniper Club land fronting Silv er Glen Run and test excavati ons at several areas within the site on record for this location (8LA1/ 8MR3601). The eastern aspect of the site (8LA1-East) once housed a massive U-shap ed mound of shell and associated archaeological remains at the confluence of Silver Glen Run and Lake George. Although the mound was mined for shell in 1923, subsurf ace remnants preserve the footprint and basal strata of the deposit. A program of coring, ground-penetrating radar (GPR), and stratigraphic excavation has helped to documen t the footprint of the mounds south ridge, but its counterpart on the north side of the mound has been difficult to characterize due to advanced subsurface disturbances associated with mining. Observed through GPR survey along the south ridge are arcuate a rrays of subsurface shell deposits, possible evidence for circular settlements akin to co astal shell rings. The lack of definitive evidence for domestic architecture and associ ated deposits (i.e., ve rtebrate fauna, foodprocessing technology) among these arrays may be merely sample error, but perhaps the area was utilized for purposes other than da ily living. Elaborately decorated pottery of the Orange tradition (ca. 4600-3600 years ago) attests to specialized (most likely large social) activities along the north ridge of th e mound. Pottery from the south ridge is generally plain and infrequent, and appears to date a century or two later than the pottery from the north ridge. Irrespective of the prac tices responsible for the basal deposits of the south ridge, whole shell was deposited ov er the ground surface in large quantities, suggesting that the ridge was constructe d deliberately over a short timeframe. The western aspect of 8LA1 (8LA1-West) includes the post-mining remnants of a 200-m-long shell ridge dating to the middle pa rt of the Mount Taylor period (ca. 63005750 years ago). Surviving today are subsurfa ce deposits up to one meter deep, as well as mining escarpments averaging about two me ters high. At thre e locations along the ridge, field school students profiled escarpments to docum ent the above-ground layers and then excavated below the mining pit to expose basal deposits. Revealed in all exposures were complex sequences of ba sal midden capped by brown sand and then successive, relatively thin strata of crushed sh ell with artifacts, shal low pits, vertebrate fauna, charcoal and ash, paleofeces, and other indications of domestic living. A 6-m-long trench in one location enable d us to observe stratigraphic relationships between primary and secondary deposits, between presumed hou se platforms and associated refuse, and between emplaced sand and shell. Observatio ns to date suggest that the Mount Taylor ridge formed primarily through repeated o ccupation, although the emplacement of sand iii

PAGE 5

and clean shell, and occasional interment of subadult humans, point s to activities other than domestic living. A relatively small ridge nose to the west of Locus A contains evidence for intensive activity over the Mount Taylor and Orange periods. Reconnaissance survey in what is known as Locus B showed this portion of the site to be fully intact, with shellbearing deposits extending well below the su rface. Extensive tes ting, including block excavation, shows that Locus B contains the stratified remains of three successive but fundamentally different episodes of site use. At the base is a Mount Taylor component indicative of repeated occupations dating from ca. 5750 to 4600 cal B.P., followed by a period of intensive pit di gging ca. 4500-4000 cal B.P., and finally a capping event ca. 4000-3800 cal B.P. involving the emplacement of clean shell over the pit-pocked surface. The first use of pottery (Orange Plain) co incides with pit diggi ng. The shell capping event, however, was accompanied by the depositi on of incised Orange pottery of the Tick Island varietya rare curvilinear and z oned incised punctate type. Evidence for domestic activities dating to either Orange component has proved elusive. Results of GPR survey conducted in 2011 will be detailed in a later report, but of note is an arcuate pattern of subsurface features not unlike that observed at 8LA1-East. Field school investigations are ongoing. The results of work conducted over the first four years of th e project (2007-2010) are reported here in full with exception of work in Locus C, at the western end of 8LA1-West. Testing in this location did not begin in earnest until 2011. Sufficient data are ava ilable to hypothesize that Locus C was a St. Johns II period village (ca. 800-600 years ago) w ith an arcuate a rray of houses and associated features (hearths, pits) surrounding a small central plaza devoid of shell. Fronting this village along the spring run is a thick deposit of secondary refuse including ample pottery, vertebrate and invertebrate faunal remains, charcoal and ash, and other evidence of intensive hab itation overlooking the spring pool. Along with undocumented shell mounds and other deposits elsewhere on Juniper Club property, Locus C provides ample opportunity for continued investigations well into th e future. Recommendations for additional work are provided in the back of this report. iv

PAGE 6

Acknowledgments Since 2007, the St. Johns Archaeological Fi eld School has been enabled through the support and generosity of the Juniper Club of Louisville, Kentucky. I was introduced to the club by member Dr. James Gay, who reached out to the University of Florida in the Fall of 2005 for a speaker for the annual Main Camp, held every January at the clubs property along Silver Glen Run in Lake and Marion counties, Florida. By stroke of good fortune, the request landed on my desk, and I invited Dr. Gay and his wife for a visit to Gainesville. At that meeting Dr. Gay related his interest in providing club members with a lecture on the archaeology of the region, and he shared with me what he knew about sites on the club property. With an active program of archaeological research in the region, I agreed without hesitancy to visit Ma in Camp that January. Arriving early the day of the lecture, I was treated to a tour of some of the archaeological sites on the property, including remnants of a shell ri dge, surface shell deposits spanning several acres, and an intact sand mound. I also me t then-President Dr. Thomas Reichard, who agreed to take a proposal for a field sc hool to the clubs Board of Directors for consideration. The combination of archaeolo gical sites in need of testing and onsite facilities for housing 15 student s for five weeks was a godse nd for a field school program that had run its course at Hontoon Island after five good years. It was not until I got back to Gainesville and delved into the literature of Silver Glen that I fully understood the significance of the sites on Juniper Club property and their potential for research. Four years of field schools at the Juniper Club have substantiate d this potential manifold. Visits to Main Camp to report on the resu lts of the previous summer have become an annual tradition, as have visits to the summer field school by club members. In these and other opportunities to share what we have learned, my students and I have appreciated the sincere and enduring interest club member s have in its land and its history, and we respect that enormously. A great debt of thanks is owed to all Juniper Club members for their generosity, support, an d enthusiasm for what we do. The trust they bestow upon us in using their clubhouse and other resources, and in digging into their land is truly humbling. Special thanks go to Gene Nelson, Resident Manager, for not only tolerating the annual invasion of a hoard of students, but also teaching us about the land and its history, not to mention b ackfilling our excavation units with heavy equipment. Genes assistants over the years, most recently Tony, have helped in more ways that we can count. Our thanks to Richard Estabrook of th e Florida Public Archaeology Network (FPAN) for helping field school go high-tech. On three occasions Rich brought FPANs Ground Penetrating Radar (GPR) equipment to the site to survey for subsurface anomalies and to show the students how we can learn about the anatomy of sites without digging them up. Back in Gainesville the field school bene fited from the admini strative assistance of Karen Jones and her staff of the Department of Anthropology, and former chair Allan Burns. The departments blue van was indi spensable. Additional material support for v

PAGE 7

the field school has been pr ovided by the Hyatt and Cici Brown Endowment for Florida Archaeology. My deepest thanks go to Graduate Teach ing Assistants for the field school: Asa Randall, Zack Gilmore, Neill Wallis, Pa ulette McFadden, Jason ODonoughue, Meggan Blessing, and Elyse Anderson. Of course, th e dozens of field sc hool students who did nearly all the physical work deserve special recognition, so we include here photos and the names of all students in the three years of full-blown field school s (pp. vii-ix). In addition, the 2009 field season that invloved focused excavation at Locus B with a veteran squad benefited from th e capable help of Julie Byrd of Florida State University, Erik Johanson of the University of Tennessee, and Alisa French of the University of Florida. vi

PAGE 8

2007 Crew. Top row (left to right): David Ec heverry, Kira Beam, Neill Wallis, Ann Carvalho, Scott Major. Middle row (left to right): Stacie Sachs, Sheila Rojas, Elizabeth Olson, Josh Robinson, Raymond Wright. Bottom row (left to right): Asa Randall, Riefler Lee, Jeff Brzezinski, Jennifer Pietarila ( not pictured: Randi Wilson). vii

PAGE 9

2008 Crew. Top two rows (left to right): Alex Taylor, Matt McCarthy, Aleksei Moskvin, Natan Bastoky, Daniel Tobin, Patrick Donery, Asa Randall, Fernando Luque, Meggan Blessing, Zack Gilmore. Middle row (left to right): Brandon D eegan, Amanda Fisher, Christian Davis, Moriah Goldfarb. Bottom row (left to right): Mara nda Kles, Breann DeChellis, Heather Handegard, Catherine Aust, Jason Whitney (not pictured: Neill Wallis). viii

PAGE 10

2010 Crew. Back row (left to right): Asa Ra ndall, Paulette McFadde n, John Moran, Kevin Gaduski, Clayon Melhado, Zack Gilmore, Jason ODonoughue, Cody Davis, Matt Newton. Middle row (left to right): Rudy McIntyre, Lisa Wr ight, Summer Jupin, Erin Harris-Parks, Lori ONeal. Bottom row (left to right): Ed Zegarra Jessica Bartnick, Andrea Bartnick, Katie Cook, Allison Nick. ix

PAGE 11

x Contents Management Summary......................................................................................................iii Acknowledgments................................................................................................................v Chapter 1. Introduction and Research Orientation..............................................................1 Chapter 2. Environmental and Archaeological Background.............................................13 Chapter 3. Mouth of Silver Glen Run (8LA1-East)...........................................................35 Chapter 4. Reconnaissance Survey of 8LA1-West.........................................................101 Chapter 5. Silver Glen Run, Locus A (8LA1-West)........................................................121 Chapter 6. Silver Glen Run, Locus B (8LA1-West)........................................................171 Chapter 7. Conclusions and Recommendations...............................................................315 References Cited..............................................................................................................3 25 Appendix A: Catalog......................................................................................................339 Appendix B: Radiocarbon Data......................................................................................427

PAGE 12

CHAPTER 1 INTRODUCTION AND RE SEARCH ORIENTATION Kenneth E. Sassaman After a five-season stint on Hontoon Isla nd in Volusia County, the St. Johns Archaeological Field School moved to a local e on the western shore of Lake George known to contain some of the largest prehis toric shell deposits in northeast Florida (Figure 1-1). Owned by the Juniper Club of Louisville, Kentucky, the ca. 1250-hectare property contains the remnants of three or four ancient shell monu ments of mid-Holocene age, at least nine hectares of shell-bear ing deposits, two late-period sand mounds, and buried, shell-free strata dating to the early Holocene. Its major shellworks were noted repeatedly by naturalists since the mid-ni neteenth century (Bartram 1942:44; LeBaron 1884:774; Wyman 1875:38-39), an d the antiquarian C. B. Moore (1894:176-177) dug into one of the sand mounds in 1894. Despite this early interest, modern investigations have been lacking, perhaps because most of th e large shellworks were eliminated in the early twentieth century by shell-mining operations. Reported here are the results of the first f our seasons of field school investigations at Silver Glen Run (8LA1/8MR3601; hereafter 8L A1). As recorded in the Florida Master Site File (FMSF), 8LA1 is th e 7-hectare point of land formed by the confluence of Silver Glen Run with Lake George. This is the extreme northeas t corner of the Juniper Club property in Lake and Marion counties, known to club me mbers as Shell Point. The massive U-shaped shellworks that caught the attention of early observers was mined in 1923 and sold for the sum of $17,000 (J ohnson 1994:43). Notwithstanding this destruction, subsurface and subaqueous remnan ts of shell deposits remain undisturbed. As at Hontoon Island and scores of other sh ellworks across the region, mining operations at 8LA1 exposed the basal strata of mounded shell, as well as the underlying evidence of earlier human activities. The St. Johns Archaeological Field School continues its longstanding interest in exploring the orig ins of monumentality by targeting mined shellworks such as 8LA1. The remnants of shellworks at Shell Po int are but a small portion of diverse, expansive archaeological depos its along the south margin of Silver Glen Run. Field school investigations have incl uded subsurface testing across this locale, resulting in the tentative conclusion that arch aeological deposits are contin uous from Shell Point west along the run into Marion County. Lacking any clear break in archaeological deposits along this expanse, all areas tested since 2007 are describe d in this report under the rubric of 8LA1. To simplify discussion of this massi ve site, the Shell Point portion of 8LA1 is hereafter referred to as 8LA1-East, and its c ounterpart along the spri ng run is designated 8LA1-West. We anticipate the need to di vide 8LA1 into its Lake and Marion county components following additional subsurface testing. This report is divided into seven chapters. The balance of this chapter outlines the overarching research goals of the field school, the rationa le for establishing long-term investigations at the Juniper Cl ub, and a summary of findings from this first four years of 1

PAGE 13

2 St. Johns Archaeological Field School 2007-2010: Silver Glen Run Figure 1-1. Reconstruction of pre-mining landscape of 8LA1, showing U-shaped Orange-period monument (8LA1-East), the Mount Ta y lor rid g e ( 8LA1-West Locus A ), and villa g e sites to the west ( Loci B and C ) ( courtes y of Asa Randall )

PAGE 14

Introduction and Research Orientation 3 work. Chapter 2 summarizes th e natural and cultural historie s of the St. Johns region in general and the Silver Glen Run locale in particular. Results of 2007 and 2010 investigations at 8LA1-Eas t are reported in Chapter 3. The methods and results of reconnaissance survey at 8LA1-West are the su bject of Chapter 4, followed by the details of excavations at two loci in this portion of the site in Chapters 5-6. In the closing chapter we summarize the major findings of field schools from 2007-2010 and outline priorities for further investigations. PROBLEM ORIENTATION OF FIELD SCHOOL PROGRAM Good ethical practice in Amer ican archaeology dictates th at archaeological field school training be underwritten by bona fide research program s. That is, field schools cannot be conducted for the sole purpose of student field training. Accordingly, the St. Johns Archaeological Field School continues to be structured by a series of research questions with broad anthropological relevance. Field school interest in sh ellworks of the St. Johns River valley is guided by research on ancient monument construction in the greater Southeastern U.S. In recent years, archaeologists working in Louisiana (Saunders et al 1997), Florida (Russo 1991), and South Carolina (Saunders and Russo 2002) have documented cases of monument construction dating as early as 5500 years ago. These cases predate the more widely known mound-building traditions of the W oodland period by over 3000 years, and the onset of pottery making by at least 1000 years. Our own work at Hontoon Island and vicinity adds to this invent ory of early monuments with linear shell ridges that were constructed over habitation sites of the M ount Taylor period (ca. 7300-4600 calibrated radiocarbon years before present [hereafte r cal B.P.]) (Randall 2007; Randall and Sassaman 2005; Sassaman and Randall 2012). Becau se many of these ancient shellworks in the St. Johns drainage continued to be u tilized well into the last millennium, when pottery-making cultures of the St. Johns tr adition flourished, they have often been misidentified as late-period c onstructions. All of the four shell ridges we tested on or near Hontoon Island had basal components dati ng to the Mount Taylor period, most with definitive evidence for shell mounding. The anthropological significan ce of these recent findings lies in the contradictions they pose to longstanding perceptions about an cient hunter-gatherer societies. Whereas anthropologists acknowledge an advanced level of cultur al complexity among certain ethnographic and late-prehistoric hunter-g atherer societies (e .g., Chumash, Calusa, Northwest Coast groups), those of the ancient past are widely regarded as fundamentally simple people. In the 1960s and 1970s, the empirical benchmark for primitive society was constructed from modern observations of sm all-scale, mobile huntergatherers, notably those of the Kalahari Desert of southern Africa. Wrongfully conscripted as a baseli ne for cultural evolution, thes e icons of hunter-gatherer living became standard analogs for ancient hunter-gat herers. Presumed to be egalitarian in principle, flexible in practice, and never large or stable in numbers, these populations are envisioned as lacking the wherewithal to erec t large monuments, or, for that matter, any reason for doing so.

PAGE 15

4 St. Johns Archaeological Field School 2007-2010: Silver Glen Run The earthen mounds of Louisiana pose the cl earest threat to this line of reasoning. They are, without question, inte ntional constructions that were sited and erected in stagelike fashion with great precision (Clark 2004; Sassaman and Heckenberger 2004). Little is known about the people who erected Loui siana mounds over 5000 years ago, but they appear to have had a lineal, if somewhat indi rect, historical relati onship to Poverty Point culture, the regions most cel ebrated mound-building people of the preagricultural era (Gibson 2000). Archaeologists will forever debate the significance of these earthen mounds, but no one can deny that these were monumental events that required engineering skill, orchestrated labor, and material provisioning. In contrast, shellworks of the Atlantic and Gulf coasts, as well as those along the St. Johns River, are not so obviously monument al. There exists enormous variation in the size, internal structure, and content of shellworks. Because many contain the refuse of everyday living, there is a tendency to regard them as merely trash heaps or shell middens (e.g., Marquardt 2010). In contrast, th e enormous size and formality of certain shellworks bear witness to nondomestic, ritual practices that arguably were monumental in purpose. These latter work s commonly contain layers of shell, and sometimes sand or muck, that were deposited in massive loads. In some cases, episodes of mounding shell or sand coincide with buria l of the dead (e.g., Aten 1999), but in most cases purposes other than human interment were at play because they contain no skeletal remains. Above all, large shellworks almost always encapsulate a variety of deposits, making it difficult to generalize about the pr ocess and purpose of their accumulation. And, because shell itself was derived from a quatic species consumed by humans, its use in monument construction is a persistent source of ambiguity. The advanced level of variation among sh ellworks poses a great challenge to modern archaeology: How do we recognize monu mental acts in archaeological deposits that are, in many respects, similar to those le ft from routine living? And, if we can make this distinction, what do these acts of monume ntality tell us about th e social, cultural, and political life of these ancient hunter-gatherers? What we have learned thus far from inves tigations into St. Johns River shellworks is that they are best understood as historical (as opposed to evolutionary) phenomena. In regarding shellworks as histor ical phenomena we invoke two re lated themes: (1) acts of mound building were precipitated by specific ev ents, and (2) acts of mound building were routinized in ritual practice with reference to such events, either literally or figuratively. It follows that we regard monu ments as both products of hi storically situated persons, and as texts for interpreting th eir understandin g of the past. We have observed in several contexts th at the triggering events for mounding shell (and/or sand/muck) was apparently the abandonment of settlements (Randall 2010; Sassaman and Randall 2012). Insofar as this transformation from mundane to ritual space occurred at different times across locations in the region, the ultimate cause, if there is one, may have been lo cal ecological change, particularly a change in the spatial relationship between habitation spaces and the wetland habita t from which shellfish and

PAGE 16

Introduction and Research Orientation 5 other aquatic resources were obtained. Sh ellfish apparently were collected for subsistence purposes long before the first shell mound was erecte d, so its ecological limitations must have affected the sustainabi lity of any given settlement. Confounding this logic is the fact that abandoned sites were often capped by sh ell in quantities of unprecedented scale. In other locations across the globe, ab andoned settlements were covered with earth or rock in acts of appa rent commemoration. One of the better documented cases is the Neolithic frontier of northwest Europe where long houses became long barrows upon abandonment (Bradley 1998). Importantly, this particular tradition appears to have arisen in the context of cultural encounters be tween indigenous and foreign people. Abandonment, then, may have had little to do with the ecolo gical or economic sustainability of a settleme nt, and more with the tran sformations in culture and demographic alignments that encounters precipitate. Indeed, the covering of an abandoned settlement may in some cases mark efforts of certain persons to erase or hide the material evidence of former set tlement (i.e., someone elses history). Evidence for intercultural encounters in Florida coincident with the onset of monument construction has not been forthcoming. However, no one since the time of James Ford (1969) has been seeking such evidence. The widely held but seldom questioned assumption is that cultural developments in northeast Florida were strictly homegrown. In order to substantiate th is assumption, archaeologists must fine sound evidence for the cultural identities of so cieties in question. How people express themselves through material practices is th e chief means of archaeological inference about culture. This involves mostly the ar tifacts whose forms and uses take on the qualities of tradition, as in the traditional way to make a spear point or construct a house. Foodways are also relevant in this respect; however, like subsistence technology, they are constrained by ecological parameters. In this respect, th e orthodox perspective on culture history in northeast Florida is that the lo cal Mount Taylor pop ulations initiated a riverine lifestyle involving shellfish harv esting that continued una bated through the late prehistoric era (Miller 1998). The Orange period that followed Mount Taylor at ca. 4600 cal B.P., and the St. Johns period that came after that are widely regarded as local, in situ (evolutionary) developments (Milanich 1994). Th e addition of pottery at the onset of the Orange period, once thought to signal the in trusion of a foreign people (Ford 1969), has for the last 30+ years been treated as nothing more than the technological intensification one expects of a lifestyle that enabled populations to grow larger and more stationary through time. The chronology of fiber-tempered Ora nge pottery developed by Ripley Bullen (1972) exemplifies this in sit u, evolutionary development. In his scheme, Orange pottery appears on the scene at about 4600 years ago. In the earliest cen turies of its use, Orange pottery was plain and unassuming. By about 4100 years ago, incised Orange pottery became the dominate type, followed a few centu ries later by the appearance of spiculatepaste wares of the St. John series. Th e pottery sequence of preceramic Mount Taylor Orange Plain Orange Incised St. Johns is tacitly accepted by most Florida archaeologists as evidence of historic al continuity in northeast Florida.

PAGE 17

6 St. Johns Archaeological Field School 2007-2010: Silver Glen Run Recent investigations have brought Bu llens sequence into question. New radiometric dates on soot from Orange Incised pottery shows it to be as old as the oldest plainwares, roughly 4600 cal B.P. (Sassaman 2003a). Additional assays on soot from spiculate-paste wares (St. Johns Plain and In cised) include one as old as the oldest Orange wares (Jenks 2006). Moreover, petrographic analyses of Orange Incised sherds dating to at least 4400 cal B.P. contain sponge spicules (Cordell 2004). Taken together, these observations suppo rt the inference that pottery type s long-assumed to exist in serial fashion over a 1200-year period we re actually coeval over the first few centuries of the fourth millennium B.P. Now that we know that pottery types once thought to be sequen tial were actually coeval, we have to think about pottery in something other than chronological terms. Relevant in this regard is the spatial segr egation of the earliest forms. Plain fibertempered pottery (Orange Plain) is often found in stratified shell midden deposits overlying (prepottery) Mount Taylor depos its (e.g., Sassaman 2003b). The apparent continuity in subsistence, use of space, a nd overall lifestyle suggests that pottery was simply an addition to a longstanding and persis tent cultural tradition (i.e., Mount Taylor with pottery). But other locations with fiber-tempered pottery are dominated by Orange Incised pottery, occasionally in great abunda nce. Compared to locations dominated by Orange Plain, sites with abundant Orange Incised sherds tend to be larger: the biggest, most complex shellworks along the river, as well as on the coast (Saunders 2004a). The Mouth of Silver Glen Run (8LA 1-East) is one such location. The onset of pottery use in the middle St Johns River valley was far more than the addition of a durable ve ssel technology to an existing Mount Taylor inventory. Rather, it marked a major transformation in the cultural landscape. The segregation between locations of plainware and locations of decorated ware is a pattern that would persist for at least two millennia. The segr egation has been glosse d over the years as the distinction between sacred a nd secular (e.g., Sears 1973). Wher eas this simplistic notion outlived its analytical usefulness long ago, the pattern of spatial segregation is real. When Orange Incised was believed to be a late addition to an existing sequence, the appearance of massive shellworks with abund ant Orange Incised pottery was simply the consequence of culture evolution (i.e., societ ies gradually grow bi gger and more complex with time). Now that we know it is as old as the oldest Orange Plain, the onset of pottery in the context of massive monument construction was highly eventful. The term event is used deliberately in this context to suggest that the onset of pottery and the transformation in monument ality that took place about 4600 cal B.P. was the result of population realignments that incl uded the relocation of coastal groups to the middle St. Johns area. Given the pre-exis ting importation of marine shell by Mount Taylor groups in the middle St. Johns, th e immigration of coastal groups was likely predicated on longstanding exchange alliances No matter the degree of affinity, the influx of new personnel to the middle St Johns required negotiations with local communities that involved, among other things rituals at locations of massive shell monuments. Communities had to be reinvented out of two or more two distinct cultures, and ritual at sh ellworks was apparently in tegral to this process.

PAGE 18

Introduction and Research Orientation 7 Large Orange-period monuments are t ypically underlain by shell deposits of Mount Taylor age. Because water levels ha ve risen substantially since Mount Taylor times, much of these submound deposits are now submerged. More commonly, Mount Taylor locations of shell accu mulation, including the large, linear ridges not ed earlier, were not utilized by groups making and using Or ange pottery, plain or incised. Indeed, it appears as if pottery-using communities avoi ded many, maybe most of the Mount Taylor monuments (Randall and Sassama n 2010). It seems likely that many such locations were stranded from the wetland habitats that suppl ied all its shell, and thus incapable of supporting either long-term inhabitation or a dditional construction stages of shell after 4600 cal B.P. Irrespective of their reoccupation during the Orange period, Mount Taylor shell ridges register an earlier tran sformation in use, well before the era of pottery. As noted earlier, many locations of Mount Taylor oc cupation were abandoned and then capped, usually with shell, but sometimes with laye rs of sand or muck, and occasionally human interments (Aten 1999; Endonino 2010). As we will see in Chapters 5 and 6, shell deposits along Silver Glen Run (8LA1-West) encapsulate these sorts of transformations, dating to as early as ca. 5500 cal B.P. Potential causes for this transformation in Mount Taylor site use are many. It would appear that in some instances, cha nging water levels and fluctuations in food availability triggered settlement abandonm ent (Randall 2007, 2010). Still, as noted earlier, the enormous amount of shell that is often deposit ed over abandoned settlements causes us to question whether diminished she llfish productivity was a sufficient cause of change, and to consider instead historical factors similar to those hypothesized for the Orange period. Although evid ence for population movement s during the prepottery period eludes us, Mount Taylor communities of the middle St. Johns were involved in regional exchange networks that reached as far as northern Georgia and Mississippi (Endonino 2010). Such extralocal affiliations and interactions implicate a political economy that had the potential to influence local decision-making. That is, abandonment of Mount Taylor sites may have been precipitated not by lo cal conditions, but by processes operating at a regional scale. The deposits fronting the south bank of Silver Glen Run (8LA1-West) not only contain evidence for the Mount Taylor tran sformation, they hold some of the best analytical potential for identifying casual fact ors in the transformati on. Like other firstmagnitude springs in Florida, Silver Glen R un may be less vulnerable to local changes in precipitation and temperature compared to ot her water bodies of the St. Johns Basin (ODonoughue 2011). It follows th at locations along first-ma gnitude springs, while not necessarily the greatest habitat for shellfish, were among the most predictable locations for collecting aquatic resour ces and potable water. Identifying the environmental parameters that enabled the massive harvest of shellfish for monument construction is a core objective of field school research. The abundance of snail shell that ex ists in such varied strati graphic contexts, spanning many centuries, even millennia, is a woefully unders tudied record of paleoenvironment. The

PAGE 19

8 St. Johns Archaeological Field School 2007-2010: Silver Glen Run shells of snails and mussels, vertebrate fa una, plant remains, wood charcoal, and other organic remains have been collected in cont rolled sampling column s of shellworks at nearly all of the St. Johns field school sites. Ultimately, these numerous samples recovered from contexts of varying age, location, and formprovide a robust database for comparing the relationship between climate (as it affects aquatic habitat) and events of archaeological deposition. In pursuit of this objective, two additional datasets need to be collected: (1) modern control samples of species in question with accompanying data on temperature, water levels, and other rele vant microenvironmental data; and (2) proxy data on paleoclimate for the ar chaeological samples in question. As the most abundant species in shellworks, Viviparous georgianus (banded mystery snail) is a good candidate species for intensive study (Blessing 2009, 2010). Another aquatic snail species, Pomacea spp. (apple snail), is likewise an excel lent candidate because it occurs in discrete depositional lenses throughout shellwor ks. Analysis of the stable isotopes of these shellfish remains has th e potential to provide proxy da ta on climatic factors that influenced their abundance at any point in time and space. In sum, two distinct events punctua ted the long history of inhabitation and monument construction along the St. Johns during the Mount Ta ylor and Orange periods. The first, termed herein the Mount Taylor transfor mation, is the regionwide initiation of monument construction over abandoned se ttlements. The second, Orange-period transformation is marked by the construction of especially large, complex monuments and the spatial segregation between locations at which plain pottery was deposited and those at which incised pottery was deposited. For both events, we must determine the extent to which these transformations were co eval across the region and thus precipitated by the same triggers, be they climatic or cultural. This a matter of fine-grained chronology, which requires a great deal of stratigraphic observation and sampling, coupled with independent radiometric age estim ates. Of course, timing alone is only a piece of the puzzle, because explaining these events requires robust, multiscalar data on the environmental and cultural contexts in which these events transpired. The Juniper Club is among the best locat ions to investigate the onset and transformations of monument construction in northeast Florida. As noted earlier, 8LA1East was the location of one of the largest Orange Period monuments in the region. Although it was largely destroye d in 1923, its basal components remain largely intact. The antiquity of its Orange Incised pottery was established several years ago when 8LA1 sherds curated at the Florida Museum of Natu ral History were sampled for soot that was assayed by AMS technique (Sassaman 2003a). Complementing the Orange deposits are remnants of a Mount Taylor ridge (reported in Chapter 5 as Locus A) along Silver Glen Run with a stratigraphic sequen ce reflecting episodes of dw elling, abandonment, capping, and repeated reoccupation. Additional Mount Taylor, Orange, and late-period St. Johns II deposits extend across 9 ha of shell fields, to use Wymans term. On such location, reported in Chapter 6 as Locus B, has provided one of the best record s of change in the study area: a basal Mount Taylor compone nt overlain by an assemblage of large shellfish-steaming pits of Orange age, fo llowed by a capping event that effectively buried the evidence of earlier activity and cr eated a well-preserved stratified sequence of change. We have some preliminary data on a third locus of investigation along the run

PAGE 20

Introduction and Research Orientation 9 (Locus C) that encapsulates the record of aboriginal dwelling long after the Archaic era of shell mounding. In four seasons of work at the Junipe r Club, research objectives have revolved around fundamental archaeological documentation about the extent, internal configuration, and age of the shell deposits that survived mining operations 85 years ago. As the foregoing discussion anticipates, we ar e especially interested in documenting the archaeological evidence for transformative events in ancient history. The larger environmental and cultural contexts for these events will, of course, provide the inferential basis for knowing how and why they transpired. It follows that our research design is broad reaching, and involves data gathering at multiple scales of resolution. SUMMARY OF FINDINGS Chapters 3-6 of this report provide the fu ll details of archaeol ogical investigations at the Juniper Club from 2007-2010. The para graphs that follow below provide brief summaries of these respective chapters, which are organized by the spatial units noted earlier, started with 8LA1-East, and followed in turn by Loci A-C of 8LA1-West, along the south margin of the spring run. In addition, reconnaissance survey across most of the area shown in Figure 1-1 was conducted in 2007 and 2008 to provide baseline data on the distribution of subsurface deposits 8LA1-East Field investigations in 2007 at 8LA1-Eas t were divided among three tasks: (1) systematic coring to determine the extent of subsurface shell deposits; (2) limited subsurface testing on the largest of the three islands off of Shell Point; and (3) limited subsurface testing in the presumed location of the south ridge of the U-shaped shellworks Wyman reported in 1875. The results of cori ng corroborated the general size, shape, and orientation of the shellworks Wyman describe d in print, as well as sketched in an unpublished drawing located by Asa Randall in th e Medical Library archives of Harvard University. Subsurface testing the area of the south ridge was likewise successful in locating basal shell deposits and associated Orange period pottery, but too little was exposed to infer much about the activities or circumstances attending shell deposition. Testing on the largest island at the mouth of the spring run did not locate intact shell deposits but instead suggested that this entire island was redeposited fill from the 1923 shell-mining operation. Lacking evidence for intact shell strata along the north ridge of the U-shaped shellworks, investigations in 2008 were expand ed at Shell Point and at another of the islands at the mouth of the spring run. Neither location pr oved to contain intact shell deposits, although excavation at Sh ell Point was halted at th e top of the water table, below which intact strata appe ar to lie. Despite an abunda nce of large Orange Incised sherds in the test units of island, coring and a radiometric age estimate far too young substantiated the inference that this landform, like the island tested the year before, consisted of fill redeposited during mining operations.

PAGE 21

10 St. Johns Archaeological Field School 2007-2010: Silver Glen Run Field school efforts in 2010 refocused at tention on intact s ubsurface remains of the south ridge, first observed in 2007 and estim ated to date to the latter half of the Orange period, ca. 4000-3800 cal B.P. The application of Ground Penetrating Radar (GPR) by Richard Estabrook of the Florida Public Archaeology Network provided our first remote view of subsurface deposits over a relatively large area of the ridge. Coring and subsurface testing showed that high resist ance in the GPR signals was due largely to shell density, but we were unable to find de finitive evidence for the purpose of shell deposition or the circumstances under which shell was emplaced. An overall arcuate pattern to dense subsurface sh ell suggested the pr esence of a circular or semi-circular village beneath the shell ridge, but direct evidence for arch itecture and associated features was not observed in 16 m2 of controlled excavation. For now the U-shaped shellworks at the mouth of Silver Glen Run evade bett er definition and expl anation, although the results of GPR survey provide hope for improved results in the future. 8LA1-West, Locus A Originally consisting of a ~200-m-long shell ridge, Locus A of 8LA1-West was severely reduced by mining operations in the 1920s. Like others we have tested in the region, the Locus A ridge was not entirely de stroyed. Surviving today are subsurface deposits dating to the Mount Taylor peri od, as well as mining escarpments along the northern margin of the ridge and in select locat ions elsewhere. Our strategy in such cases has been to profile escarpments to expose th e above-ground layers and then continue the profiles below the present surf ace in controlled excavation units. In 2007 this process began at two locations in Locus A: at the eastern end of the ridge escarpment and some 90 m to the southwest, along the north edge of the ridge. Revealed in both exposures was complex sequence of basal midden capped by brown sand and then successive, relatively thin strata of crushed shell with artifacts, shallow pits, vertebrate fauna, charcoal and ash, paleofeces, and other indications of domestic living. A radiocarbon assay near the base of the 3-m-deep deposit returned a Mount Taylor age estimate of ca. 6200-5950 cal B.P. The following summer we opened a larger exposure in one of the only mining escarpments to run perpendicula r to the length of the ridge an d the spring run. This 6-mlong trench verified the ge neral sequence observed in 2007 and provided two additional radiometric assays: ca. 63006100 cal B.P. near the base of the deposit, and ca. 60005750 cal B.P. on charcoal approximately 60 cm higher up the profile. The larger exposure enabled us to observe stratigraphic facies not seen in the earlier profiles. Apparent were relationships between primar y and secondary deposits, between presumed house platforms and associated refuse, and be tween emplaced sand and shell. Expansion of the escarpment test at the east end of the ridge was halted after encountering the remains of subadult humans, which were retu rned to their original location, backfilled and thereafter avoided. Observations to date suggest that the Mount Taylor ridge formed primarily through repeated occupation, although the emplacement of sand and clean shell, and interment of at least subadults poi nts to activities other than domestic living.

PAGE 22

Introduction and Research Orientation 11 8LA1-West, Locus B A relatively small ridge nose overlooking Silver Glen Run was the locus of intensive activity over the Mount Taylor and Orange periods. Reconnaissance survey in 2007 showed this area, dubbed Locus B, to be fully intact, with stratified shell-bearing deposits extending down at least one meter below the surface. A single test unit excavated that same year revealed a sequenc e of stacked surfaces with abundant shell, vertebrate fauna, and plain fiber-tempered potte ry of the Orange tradition. The following summer Zackary Gilmore commenced with intens ive testing of Locus B, a project that carried forward through 2010 to form the basis for dissertation research at the University of Florida. As reported by Gilmore in Chapte r 6 of this report, Locus B houses deposits during three successive but fundame ntally different episodes of site use (see also Gilmore 2010). Revealed in a total of 45 m2 of excavation to date is a basal Mount Taylor component indicative of repeated occupa tions dating from ca. 5750 to 4600 cal B.P., followed by a period of intensive pit digging activity ca. 4500-4000 cal B.P., and finally a capping event ca. 4000-3800 cal B.P. involving the emplacement of clean shell over the pit-pocked surface. Coincident with the pit act ivity is the first use of pottery at Locus B, largely Orange Plain wares. The shell ca pping event, however, was accompanied by the deposition of incised Orange pottery of the Tick Island varietya rare curvilinear and zoned-punctated type that devi ates both technologically and st ylistically from the usual linear incised Orange pottery, such as that found in great abundance along the north ridge of the U-shaped shellworks to the east. Gilmore is investigating the circumstances surrounding the seemingly abrupt transformation of site use at Locu s B and will attempt to related these changes with coeval developments at the U-shaped shellworks beyond. Excavations in Locus shave proved to be highly productive and its results original. 8LA1-West, Locus C A second, larger ridge nose to the immedi ately west of Locus B and overlooking the spring pool of Silver Glen is home to a late-period component with abundant secondary refuse and promise for an intact vi llage site. Limited testing in 2008 revealed a thick and dense secondary midden on the northern slope of the landform, extending to the spring pool. Pottery of the St. Johns II pe riod (post-A.D. 750 or post-1200 cal B.P.) dominated an assemblage rich in vertebra te fauna and other evidence of intensive activities. Limited testing atop the ridge nose, on the nearest flat ground, revealed subsurface features and other indications of a possible village. About 150 m southeast of this location is an intact sand mound, one of two in the vicin ity that was not impacted by antiquarian digging or modern land use. Although we have very little additional information about Locus C, results of testing to date show great prom ise for an intact St. Johns II period village. Field investigations at Locus C were intensified in 2011. The results of this and the 2008 work are reserv ed for a follow-up report in the near future. CONCLUSION The first four years of archaeological investigations at the Juniper Club substantiate the claim that despite early 20th-century mining operations that removed

PAGE 23

12 St. Johns Archaeological Field School 2007-2010: Silver Glen Run large quantities of shell, both intact s ubsurface and above-ground deposits across a large tract of land hold enormous potential for improving our und erstanding of pre-Columbian life in the region since at least 6000 years ago. Even gr eater time depth awaits investigations of subsurface deposits beneath those of Mount Taylor and Orange age. Field schools have provided excellent opportunities for fledging archaeologists to learn the technical aspects of the trade, but they have likewise afforded professionals the opportunity to delve into research questions that are prec luded by the strictures of most contract archaeology. Our field school hostth e Juniper Club of L ouisville, Kentucky is owed a debt of thanks for its enduring support of this effort.

PAGE 24

CHAPTER 2 ENVIRONMENTAL AND ARCHAEOLOGICAL CONTEXTS Asa R. Randall This chapter situates field school investigations at the Juniper Club within regional environmental and archaeological contexts. Envi ronment is considered first, focusing in particular on physiography and hydrology. Th e archaeological contexts are then reviewed, with particular attention paid to the Middle and Late Archaic periods. In both cases a regional overview is provided, fo llowed by locality-speci fic discussions. ENVIRONMENTAL CONTEXT The Silver Glen Springs watershed is situ ated at the intersec tion of Marion, Lake, and Volusia counties, approximately 15 km north of Astor, Florida. The watershed is defined by the first magnitude Silver Glen Sp rings that issues forth along a 1-km-long run into Lake George (Figure 2-1). This lake is the second largest body of water in Florida. In addition to Silver Glen Springs, Lake George directly receives wa ter input from the St. Johns river, Juniper Springs, and Sweetwater Sp rings to the south, a nd Salt Springs to the north. The current configuration of these hyd rological features and the greater St. Johns basin resulted from a long history of fluctuat ing sea level and attendant progradations and regressions of surface waters, localized faultin g and solution of carbonate sediments, as well as more recent factors such as channel dredging for navigation. A number of syntheses and cogent discussions of the geology and geomorphology of Florida have been published (Randazzo and Jones 1997; Whit e 1970). Those aspects relevant to the Silver Glen Springs watershed are discussed here. Regional Physiography Like all of peninsular Florida, the regional physiography of the St. Johns River Valley ultimately owes its curre nt configuration to marine processes (Schmidt 1997). Currently, the dry land of peninsular Flor ida occupies approxima tely one-half of the Florida Platform. Extending out into the Gulf of Mexico and Atlantic, the Platform is characterized by low relief, and is com posed of Cenozoic carbonate sedimentary lithologies that lie unconformably upon a Paleozoic and metamorphic basement. The Florida Platform has been altern atively inundated by shallow seas and exposed as dry land during much of the Cenozoic era. The low elevation of the Platform (a maximum of 104 meters in the Panhandle) has made it particularly susceptible to relatively small changes in sea level. Sea level fluctuation has resulted in frequent progression and regression of marine, estuarine, and near-shore environments. This process has left the Florida coastal zone dominated by positive features including elevated relict upland ridges, barrier beaches, and sand dunes, and negative features representative of shallow seafloors (Schmidt 1997). Terraces that reflect long-term sea level stands have been identified. In the study area these include the Silver Bluff and 13

PAGE 25

14 St. Johns Archaeological Field School 2007-2010: Silver Glen Run Figure 2-1. Location of springs contributing to Lake George (USGS Daytona Beach Beach 100k quadrangle).

PAGE 26

Environmental and Archaeological Contexts 15 Palmlico Terraces (0-8 m amsl) and Penholoway and Talbot Terraces (8-21 m amsl). Additionally, the carbon ate composition of many of Floridas sedimentary deposits has been equally influential. Ca rbonate lithologies are particularly susceptible to dissolution, which results in karst topography and hydrogeol ogy. Typical features of karst topography are sinkholes, sinking rivers, disappearing lakes, and springs. Geomorphologists have recognized a number of physiographic regions defined by topography, surficial geology, and hydrology (Cooke 1939; Schmidt 1997; White 1970). The St. Johns River is located in the Atlantic Coastal Lowlan ds, a zone typified by coastparallel features. Most posit ive features in this region ar e relict beaches and marine terraces formed during the late Pleistocene a nd Holocene, and are composed of siliclastic marine sediments. The headwaters and mouth of the river are situated within the Eastern Valley, while the middle St. Johns occupies a position west of the Crescent City-Deland Ridge. The Crescent City-Deland Ridge is the only karst-dominated topography in the region, and is a major source of groundw ater via the Floridan Aquifer. Groundwater and channeled water hydrol ogy of the St. Johns is linked to precipitation and geology. Ultimately, all of Floridas freshwater is derived from precipitation (Miller 1997). Although much precipitation is lost due to evapotranspiration and runoff, a significant portion is returned for the recharge of aquifers. Water levels for most of Floridas streams and la kes are directly related to the aquifer levels. Florida has five principle aquifers, only two of which have output in the middle St. Johns. In general, the study area is typified by an undifferentiate d surficial aquifer. Water is typically unconfined in Pleistocene and Holocene sedime nts averaging 50 feet in thickness, and is present at or just below the ground surface. The Floridan Aquifer is the most extensive and productive of all of Floridas aquifers. It extends throughout the state, in addition to Georgia, Alabama, and South Carolina. Genera lly, the Floridan Aquifer is restricted to carbonate rocks of Tertiary Age, and remain s confined well below the ground surface. The aquifer is unconfined or outcrops in regi ons where these carbonate rocks are thin or have been penetrated by sinkho les. In the study region, the Floridan Aquifer discharges via several first-order magnitude (greater than 100 cubic feet per second or more) springs, including Silver Spring, Silver Gl en Springs, and Blue Spring. As Miller (1998:28) notes, the dominant f actor in the regions landscape is water, which is concentrated along the St. Johns Ri ver drainage. The St. Johns Rsiver, which has its headwaters in southern Brevard Coun ty and discharges into the Atlantic at Jacksonville, is the largest river in Florida, measuring 500 km l ong. It is also unique as it is one of few rivers in the northern hemisphe re to flow from south to north. Although it is extensive and broad, the St. Johns disc harges on average only 8,300 cubic feet per second. The discharge is related primarily to volume and less to velocity. This is due to a wide floodplain and a low gr adient (0.02 m per kilometer) (Miller 1998:28). For most of its length, the St. Johns is within five feet of mean sea level. The low gradient makes the river responsive to small changes in sea level, and even today the river is tidally influenced as far south as the Wekiva River.

PAGE 27

16 St. Johns Archaeological Field School 2007-2010: Silver Glen Run The St. Johns River is composed of th ree distinct segments whose different characteristics relate to a complex geomorphic history (Adamus et al. 1997; Schmidt 1997; White 1970). Like many of the large rive r systems in Florida, the St. Johns River is situated in a swale betw een elevated, upland ridges. Although this configuration was once thought to have formed during late Plei stocene times as a drowned lagoon, it is now believed to have been formed in part with in a beach-ridge plain (White 1970) during the early Pleistocene. With the exception of th e lower St. Johns, the ri ver is characterized by lakes arrayed in a linear fashion, oriented with the flow of the river. White (1970) suggests that these lakes are si nkholes which have been differentially filled with sediment and linked by channeled surface water. The upper segment of the river flows between southern Brevard County and Sanford, Florida. This segment is the he adwaters, and is characterized by poorly integrated braided streams and extensive wetlands. The middle St. Johns, between Sanford and Lake George, is often referred to as the St. Johns Offset. In a headwardconsequent course, the river would be expected to flow from the headwaters to Jacksonville in a relatively stra ight line following the late Pleistocene beach ridges of the Eastern Valley. However, at Sanford the St. Johns jogs to the west, flowing west of the Crescent City-Deland Ridge. North of Lake George, the river jogs back to the east. It is believed that this portion of the river formed during the early Pleist ocene, during a period of low sea level, when the offset portion of the river captured the headwaters south of Sanford. The river was eventually integrated when the basin was first inundated, creating an estuary. The drainage pattern of the middle St. Johns is dominated by an anastomosing pattern, characterized by numer ous parallel channel segments. The floodplain is composed of freshwater mars hes and swamps. The lower St. Johns is situated between the eastward jog north of La ke George to the mouth at Jacksonville. This course is parallel with Crescent Lake, a relict channel of the St. Johns abandoned when the middle St. Johns switched to its current location. This section of the river is essentially a drowned estuary, and is characterized by a broa d channel, averaging over 1 km in width, and by inshore marine habitats. Late Pleistocene and Holocene Environmental Trends The same processes that have aff ected the physiography and hydrology of Floridanamely fluctuating sea level and attendant shifts in climate and environmental regimeshave structured human settlement and their archaeological recognition in the study region. At the end of the Pleistocene s ea levels were significantly lower than today (upwards of 40 m), resulting in the extensi on of inhabitable land over 200 km into the Gulf of Mexico and to a lesser extent into the Atlantic (Faught 2004). Between roughly 12,000 and 10,000 years ago sea levels initially rose quickly, inundating large expanses of the Florida Platform and interior drainages. Although near -modern levels were gradually achieved by 6000 years ago (Faught 2004), sea level fluctu ated throughout the middle and late Holocene. The increase in sea level and surface water resulted in the inundation of many early sites. Although inundated sites are routinely discovered in lowenergy environments such as the Gulf of Mexi co and interior sinks and drainages, many

PAGE 28

Environmental and Archaeological Contexts 17 sites along the Atlantic Coast were likely destroyed or deeply buried by transgressing shorelines (Ste. Claire 1990). The reduction of river gradients in respons e to sea level change resulted in the initial alluviation and subse quent surface stabilization of in terior and coastal fluvial regimes, which in turn affected the flow and biotic characteristics of river channels and floodplains (Schulderein 1996). Peninsular Floridas arid late Pleistocene conditions, characterized by low surface water levels, gradua lly gave way to a wetter, modern regime ca. 7000 years ago (Watts et al. 1996). At 11,000 years ago oak scrub and prairies characterized peninsular Florida. Around 9500 years ago pine and swamp vegetation expanded from South Carolina throughout much of the Coastal Pl ain, becoming fully established by 5500 years ago in southern Florida (Watts et al. 1996: 37). Although the broad characteristics of th e middle St. Johns were in place well before humans entered the region, the late Pl eistocene and Holocene history of the valley has important consequences for settlement and archaeological recognition. Today, the floodplain is dominated by multiple channels, oxbow cutoffs, lakes, and lagoons. These features reflect a complicated history of ch annel switching, avulsion, and infilling. In part, this variation is related to the shallow gr adient of the river and sea level. Based on the distribution of archaeologica l sites, this hydrologic regime dates to at least 7000 years ago when the elevation of the river rose to within a meter of present-day levels. However, there were likely significant shifts in the course of the river that would have had effects on the distribution of swamps and wetlands. The presence of archaeological sites hundreds of meters from the main channe l, or outside of the range of productive shellfish beds, indicates changes have occurr ed (Wheeler et al. 2000). The picture is further complicated in shallow bodies of water, such as Lake George. Presumably, much of the lake bed was once exposed as land. Where the pre-inundation channel of the St. Johns was situated with respect to the SGS watershed is unknown. More data are necessary to understand the comp lexity of channel changes through time. Finally, while it is has long been assumed that springs along the St. Johns did not have established flow until well into the Holocene (Miller 1998), this hypothesis has yet to be tested across a wide range of locations. Indee d, recent investigations of Salt Springs indicate that water was flowing there as early as 9000 years a go (O'Donoughue et al. 2011). More recent changes in the flow character istics of the river have b een wrought during the last 200 years. In addition to the urba nization of the headwaters, the majority of the main channel of the St. Johns has been dre dged. Historic documents indi cate that the river was first dredged in portions as early as the 1880s, in cluding Volusia Bar at the south end of Lake George (207th House of Representatives, Document no. 1111). During the last century, the river has been fully channelized. Silver Glen Springs Watershed The Silver Glen Springs watershed is hydrologically defined by the firstmagnitude Silver Glen Springs. Water issues out from a main vent in the center of the spring pool, as well as a seconda ry vent to the west (Figure 2-2). From here it travels ca. 1 km along a channel of variable width, wher e it debouches into Lake George. Today the

PAGE 29

18 St. Johns Archaeological Field School 2007-2010: Silver Glen Run Figure 2-2. Topography of the Silver Glen Springs watershed (elevations derived from the Volusia County Department of Public Works LiDAR dataset). spring pool is surrounded by low-lying topography associated with a recreational area of the Ocala National Forest. This surface conf iguration is no doubt a product of ancient shell deposition and recent shell removal. The spring run flows between elevated landforms on either side, with wetlands of varying width separating the channel from the shoreline. The upland slope is generally quite steep, and rises rapidly in absolute elevation from 1.0 to 4.0 m. Although this slope no doubt reflects the natural geomorphology of the basin, partic ularly to the west, it should be kept in mind that anthropogenic deposits are present on both the no rth and south banks of the run. Like the spring pool, much of the topography here is a consequence of anci ent human deposition. A variety of soils are present within th e Silver Glen Springs watershed (Figure 23). The following descriptions are deri ved from USDA-NRCS (2011) definitions and soil surveys of Lake (USDA 1975) and Marion (USDA 1979) counties. Paola fine sand (0 to 8 percent slopes and 8 to 17 percent slopes) is present in the western aspect of the recreational area at the spring poo l. Subtypes of this soil vary by slope, and range from level sand hills to strongly sloping surfaces associated with sinks, ridges, and stream banks. This soil is excessively drained, and is associated with pine-scrub oak forest. The area also contains Pomello sand which is moderately well to somewhat poorly drained soil. Unmanaged vegetation is typified by scrub oak, dwarf live oak, saw palmetto, and various pines. Outside of the recreati onal area to the northeas t is poorly drained Immokalee sand which is host to longleaf and sl ash pines and undergrowth of saw palmetto, gallberry, wax myrtle, and pineland threeawn. Much of the spring run is

PAGE 30

Environmental and Archaeological Contexts 19 Figure 2-3. Soils present within the Silver Glen Springs watershed. bordered by soils of the Sellers and Pamlico series. These poorly drained soils are typical of depressional areas, poorly defined draina ge ways, and level floodplains. Native vegetation consists of pond pine, tupelo gum, sweetbay, bald cypress, gum trees, cypress, greenbrier, wax myrtle bushes, with underg rowth of gallberry or pickerelweed and perennial grasses. The one exception on the southern margin of the run is Made Land, a description that the USDA employs for locatio ns which have been heavily impacted by cutting or filling. Indeed, much of this area was mined for shell in 1923, which no doubt resulted in the reworking of deposits across much of the terrace edge. The forests comprising the margins and uplands of the Silver Glen Springs watershed are host to a wide array of fauna Those of economic importance to humans include white-tailed deer, black bear, r accoon, opossum, gopher tort oise, and turkey. Numerous species of birds, mammals, reptiles, amphibians, and gastropods also inhabit these zones. The wetlands associated with Silver Glen Run and Lake George provide habitat for a diverse arra y of aquatic fauna. Vertebrates such as alligator, turtle, otter, and upwards of 40 species of fish are present. In addition, the wetlands away from the main pool and within Lake George are potential habitat for numerous mollusks. Species of

PAGE 31

20 St. Johns Archaeological Field School 2007-2010: Silver Glen Run importance to the inhabitants of the region include the gastropods Viviparus georgianus (banded mystery snail) and Pomacea paludosa (Florida apple snai l), as well as the freshwater bivalve (Unionidae). Smaller gastropods such as Elimia sp. (rasp Elimia), and the rams horn and mesa-rams horn (Planorbella sp. ) can be found with these other species. Unfortunately, little detailed inform ation on the habitat pref erences, habit, and seasonal life histories of thes e species is currently availa ble. It is unknown in what frequencies these invertebrate species norma lly co-occur. Moreover, few data exist on whether there is predictable variation in their seasonal or spatial availability. In general, all species prefer shallow near-shore environm ents, such as grassy marshes and shallow lagoons (Quitmyer 2001). Viviparus prefer soft, muddy substrates with slack water, such as lagoons, creek edges, lakes, and springs (Clench and Turner 1956). Pomacea is known to prefer marshes with emergent vegeta tion, typically w ith at least 50 cm of water (Darby et al. 2002). Recent Landscape/Land Use Change (1923Present) Like the much of the St. Johns basin, the Silver Glen Springs watershed was radically transformed by mining and dredging operations. Mu ch of what we know about the pre-modification arrangement of the wate rshed is derived from early explorers and archaeologists whose observations will be rela ted in subsequent sections. Now is a good time, however, to outline changes that have occurred in the last century. Lake County probate documents record agreements regardi ng the sale of shell and permission to mine and dredge on both sides of the run. C. W. Perkins, later the Lake George Shell Corporation, was granted the right to mine shell on the south side of the run, including the mouth and portions to the southwest of th ere. He was also granted permission to mine shell on the north side, approximately half-way down the run. Furthermore, he was permitted to dredge the channel in order to f acilitate loading shell onto barges. In 1932, the agreement for the south side of the run was amended. Perkins apparently had excavated shell bearing deposits well below the water level in numerous places along the run, in violation of the agreement. Simila rly, the Juniper Club was faulted for removing shell for their own roads, and allowing Marion County workers to do the same. The Lake George Shell Corporation was given the right to mine remaining shell above the water level up to 500 feet from the spring run. We also know that beginning sometime in the early 1930s, Henry Henderson (then owner of the land surrounding the spring pool) began removing shell from around the spring, although it is unknown whether it was carted away or dredged (see below). He ev en constructed a house that overlooked the spring pool. Over the years, the northern side of the run has passed through several owners, who further developed the property. Many of these structures were recorded on the Juniper Springs USGS t opographic quadrangle. When the recreational area was acquired by the U.S. Forest Service in 1990, the majority of this infrastructure was removed.

PAGE 32

Environmental and Archaeological Contexts 21 Figure 2-4. A comparison of aerial photographs of the Silver Glen Springs watershed, taken between 1941 and 2004.

PAGE 33

22 St. Johns Archaeological Field School 2007-2010: Silver Glen Run There are no known photographs of Silver Glen Run prior to or during shell mining. However, aerial (Figur e 2-4) and terrestri al photographs record the effects of mining shell and developing the land for recr eational use. Aerial photographs captured by the USDA are available from 1941, 1958, a nd 1974, as is a recent image from 2004. There are several pattern s of note. In 1941, there are a number of areas that were cleared of vegetation, as evidenced by light colored pa tches. Importantly, th is includes the tract around the spring pool and extending into the upla nds, as well as the majority of southern border of the spring run, property of the J uniper Club. The southwest portion of the run is cleared back approximately 400 feet, consis tent with the probate court documents. At the mouth of the run is the other southerly cl earing. This is associated with the Juniper Clubs lodge and other infrastructure. Ac ross the watershed in successive years, vegetation filled in these cleared areas, leaving a patc hy network of open spaces and immature forests. ARCHAEOLOGICAL CONTEXTS A number of syntheses of Florida prehis toric archaeological contexts have been issued for the St. Johns Basin (Goggin 1952; Miller 1998; Russo 1990a) and for the state of Florida (Borremans 1990; Milanich 1994; Milanich and Fairbanks 1980; Russo 1990b). These and other locality -specific studies are drawn upon to review the culture history of the middle St. Johns River. Paleoindian (ca. 13,000,000 cal BP1) and Early Archaic (ca. 11,000 cal BP) The late Pleistocene Paleoindian trad itions include Clovis, Suwannee-Simpson, and Dalton, which are identified on the basis of diagnostic hafted bifaces. In addition to lanceolate hafted bifaces, the toolkits ar e characterized by a suite of formal unifaces (Daniel et al. 1986), bola stone s (Neill 1964), the Aucilla adze, and a variety of bone and ivory tools (Dunbar and We bb 1996). Early Holocene trad itions dating between ca. 12,000 and 10,000 years ago are iden tified by Side-Notched a nd Corner-Notched Bolen points (Bullen 1975). Aside from changes in hafted biface morphology and the addition of new tools, the toolkits of these horizons are consistent with Paleoindian forebears, particularly Dalton. Today the oldest sites are typically re stricted to inundated contexts such as drowned river segments (Dunbar et al. 1988 ; Faught 2004), sinkholes (Clausen et al. 1979), or perched basins and depressions (Daniel and Wisenbaker 1987; Neill 1964; Sassaman 2003c). A trend towards increa sed surface water ca. 11,000 years ago and subsequent settlement expansion is atte sted by Early Archaic diagnostics at Late Paleoindian sites, as well as small numbers of Early Arch aic diagnostics in previously uninhabited localities. In general, they are redundant and may represent frequent residential mobility (Milanich 1994). Noting the co-occurrence of Paleoindian artifacts and karst topography in northwest Florida, Dunbar and Waller (1983) posited the Oasis hypothesis, that in effect Pale oindian populations were tethered to karst regions, abundant in toolstone and reliable su rface water. Although this model matches the general 1 Calibrated years before present.

PAGE 34

Environmental and Archaeological Contexts 23 distribution of early component s, Paleoindian and Early Arch aic diagnostics have been recovered from the St. Johns Basin (see below). Between 10,000 and 7300 cal BP, Florid as Archaic traditions remain poorly defined (Austin 2004; Milanich 1994). Stemmed points, consistent with the Kirk Stemmed type and locally re ferred to as Kirk, Wacissa, Hamilton, and Arredondo (Bullen 1975) are distributed throughout the North, Central, and Gulf Central portions of the state, often in similar localities as early fo rms (Milanich 1994). Stratigraphic excavations at Harney Flats (Daniel and Wisenbaker 1987) West Williams (Austin 2004), and Trilisa Pond (Neill 1964) indicate an increase in the diversity of unifacial technology. This period also witnesses the establishm ent of a long-standing mortuary tradition involving the interment of individuals in shallow bodies of water such as ponds or sinkhole margins. Windover Pond (ca. 9500 cal BP) in Brevard County represents the earliest and is the most thoroughly investigated pond mort uary in the region (Doran 2002b). These sites are typified by large numbers of individual s, and appear to have been repeatedly used over extended periods. Fo r example, at least 168 individuals were interred at Windover Pond over th e course 1300 years. Outside of the middle St. Johns pond burials continue into th e Middle Archaic (Beriault et al. 1981; Doran 2002a). In general, Paleoindian and Early Arch aic sites are underrepresented within the study area (Sassaman et al. 2000) Several factors may account for this, including a lack of adequate toolstone as well as fewer surveys of submerged contexts. In the St. Johns Basin, early sites are expected to occur adjacen t to firstor second-magnitude springs fed by the Floridan Aquifer, including Silver Glen Springs, Salt Springs, Juniper Springs, Fern Hammock Springs, Green Cove Springs, Beecher Springs and Blue Spring (Miller 1998:84). The few known sites and isolated finds that have been documented seem to fit this overall pattern (Sassaman et al. 2000). A survey of Cres cent Lake demonstrated that there is great potential for reco vering early assemblages in the region (Sassaman 2003c). Crescent Lake is a perched water source that was well-watered throughout the late Pleistocene and early Holocene. Collector surveys and near-shore survey of submerged contexts revealed the presence of numerous early diagnostics. Paleoindian and Early Archaic diagnostics have been recovered from Silver Glen Springs, and have also been reported from the bottom of Lake George (Thulman 2009). Middle and Late Archaic (ca. 7500 cal BP) Several environmental and social trends define the Middle and Late Archaic periods. In broad terms the Mi ddle and Late Archaic periods are coeval with increasingly wetter conditions of the middle Holocene, w ith essentially modern conditions occurring by the end of the Late Archaic. Sites of this period are found throughout much of Florida, and for the first time are located in interior forests, along the St. Johns River and the Atlantic Coastal Lagoon (Milanich 1994:77). Lifeways predicated on intensive shellfishing are present in the St. Johns by 7300 cal BP and no later than 6000 cal BP on the northeast coast of Fl orida (Russo 1996). The distributi on of sites reflects an overall increase in available surface waters and the exploitation of new habitats, as well as a

PAGE 35

24 St. Johns Archaeological Field School 2007-2010: Silver Glen Run probable increase in population. By 5500 years ago regi onalization is evident across Florida. These new traditions, focused particularly on wetlands, presumably resulted in increasingly larger populations and more permanent settlements (Milanich 1994:87). Throughout Florida, changes in material culture, including projec tile point styles and the appearance of pottery, are used to de lineate subperiods and local traditions. In the middle St. Johns several subperiods ha ve been defined, including the Newnan Horizon, the Mount Taylor culture, and the Or ange period. Additiona lly, the preceramic Archaic is a generic term denoting Middle to Late Archaic traditions dated between 7000 and 4600 cal BP, which were without pottery technology. Archaeologists typically assign sites to the preceramic Archaic when Archaic-age assemblages lacking diagnostic artifacts are recovered. Newnan Horizon (7500 cal BP). Across much of Peninsular Florida researchers have recognized the Newnan Horizon, char acterized by short, narrow stemmed, broad bladed chipped stone hafted bifaces (Milanich 1994:76). A number of types have been defined, including Newna n, Marion, and Putnam (Bullen 1975). There is significant variation in the form of stemme d hafted bifaces from this period, leading to a less formal designation of the Florida Archaic Stemmed type, which includes any broad-bladed stemmed hafted biface. Lithic artifacts during this period were typically manufactured from thermally altered chert or silicified coral (Ste. Claire 1987). Dates place Newnan sites between 7500 and 4600 years ago (Milanich 1994:77), although similar forms were likely produced into the Late Archaic period. Settlement in interior Florida, which contains much of the available chert and silicified coral for the production of stone t ools, is characterized by a dichotomy between large, diverse assemblages and small lithic scatters. The large sites have been interpreted by Milanich (1994:79) as indicative of reduced seasonal mobility. Austin (2001) suggests, however, that the larger sites li kely represent more intensive short-term reduction episodes near raw material outcrops. Several quarries have been identified, including the Senator Edwards site in cen tral Florida (Purdy 1975). Newnan horizon hafted bifaces are routinely recovered in shell midden contexts along the middle St. Johns. The lack of toolstone in the middl e St. Johns precludes their local production. Lithic provenance studies indicate that chi pped stone tools were be ing imported into the region from West and Central Florida (Endonino 2007). Mount Taylor (ca. 7300 cal BP). The Mount Taylor culture (ca. 7300 cal BP) has been defined to describe the in tensive late Middle Ar chaic and early Late Archaic occupation centered on the extensive we tlands of the middle St. Johns River, the adjacent Ocklawaha and Wekiva rivers, and associated Atlantic Coastal Lagoon (Goggin 1952; Wheeler et al. 2000). This is an archaeolo gical construct, and it refers to a suite of site types and diagnostic artifacts. Recently, the period has been split into an early Mount Taylor phase (ca. 7300 cal BP) and a late Thornhill Lake phase (5600 cal BP), on the basis of changes in the frequency and style of exotic objects and mortuary traditions (Beasley 2008; Endonino 2010). Alt hough the broad details of lifeways are known for this period, the Mount Taylor cu lture remains poorly understood for several

PAGE 36

Environmental and Archaeological Contexts 25 reasons. Mount Taylor pe riod components are typicall y buried deeply under later components or submerged under alluvium or peat deposits. Moreover, many sites of this period have been destroyed or impacted by modern land-use practices. The majority of shell mounds were mined in part or whole for road fill during the middle of the 20th century (Milanich 1994). Settlement patterns during this period are not well known (Wheeler et al. 2000). Seasonality studies of late Mi ddle Archaic sites in the coastal Timucuan Preserve (Russo et al. 1993) suggest that Mount Taylor communities likely ha d well-established patterns of movement within each region. Although this does not preclude movement either within the middle St. Johns, or to the Atlantic coast or in terior, it does suggest that populations were relatively circumscribed. Based on botanical remains and hydrology, Groves Orange Midden has been interpreted as a multiseasonal occupation (Russo et al. 1992). Based largely on the assumption that shell mounds are villages, it is typically presumed that the large shell deposits throughout the middle St. Johns represent multiseasonal to permanent year-round base camp s that articulate with smaller task and season-specific localities (Wheeler et al 2000). This remains to be tested. Sites with Mount Taylor components ar e present throughout the middle St. Johns basin (Sassaman et al. 2000). Although many sites are located adjacent to the main channel of the St. Johns, many others are situated within low-lying swamps or marshes. Wheeler et al. (2000) suggest that there are several genera l configurations, including ovoid midden-mounds, ridges of shell, complexe s of shell fields, ri dges, and mounds in addition to small, diffuse middens. The conf iguration of Mount Taylor occupations is made less clear in multicomponent sites, wher e Mount Taylor assemblages are partially or completely obscured by later deposits, but are generally characterized by linear footprints (Randall 2010). It is unclear how these sites are inte rnally organized, and whether there are specific areas for habitation, refuse dispos al, or other tasks. Based primarily on stratigraphic inference and non-mounded she ll bearing sites, so me Mount Taylor occupations appear to be composed of households arranged in a linear fashion (Randall 2010). Few features are known from this tim e period. Aside from the occasional postmold, features that have been recorded at large sites such as th e Lake Monroe Outlet Midden (8VO53) (Archaeological Consulta nts and Janus Research 2001) and Fort Florida (8VO48) (Johnson 2005) tend to be large shell-filled basins. Further evidence comes from the Lake Monroe Outlet midden, where lithic reduction tasks were apparently segregated from domestic re fuse or processing ta sks (Scudder 2001). Similarly, at the Hontoon Island North site (8VO202) primary and secondary midden were separated in space suggesting the presen ce of discrete habitation and refuse areas (Sassaman et al. 2005). Stratigraphically, M ount Taylor shell mounds are characterized by shell lenses, typically com posed of whole and crushed Viviparus Pomacea and bivalve. Strata can be compos ed of a mixture of these taxa, or as concentrations of single taxa. The addition of sand within shell deposit s appears to be a later practice. In many cases individual strata are composed of a si ngle taxon, which may be burned, whole, or crushed. Another feature of Mount Taylor si tes is the presence of concreted shell, which

PAGE 37

26 St. Johns Archaeological Field School 2007-2010: Silver Glen Run can occur either as thick, extensive lenses or as localized conglomerates (Wheeler et al. 2000:145). It has been suggested that concreted shell deposits form by the interaction of ash, shell, and percolating water. In addition to basal deposits of conc reted shell, Mount Ta ylor sites typically contain saturated or submerged components that appear to have been inundated during or soon after deposition. Due to the cost and time involved in dewatering and excavating saturated deposits, these have only rarely b een investigated, but include Salt Springs (O'Donoughue et al. 2011) and Groves Oran ge Midden (8VO2601), a Mount Taylor and Orange period site on the eastern shore of Lake Monroe (McGee and Wheeler 1994). Groves Orange midden, for example, is a segment of the much larger multicomponent Old Enterprise mound and shell field comp lex (8VO55). Stratigraphic excavations yielded five discrete strata. The earliest primary deposition (Stratum IV) dates roughly between 7000 and 6000 cal BP and is characterized by dense Viviparus shell deposits. These early dates are supported by dates clus tering around 7300 cal BP from the base of Live Oak Mound and Hontoon Dead Creek Mound (Sassaman 2003b, 2005), indicating that the establishment of wetland habitat a nd its exploitation by residents of the middle St. Johns occurred by at least 7300 years ago, if not before. At Groves Orange Midden, this basal stratum underlies a thick peat de posit (Stratum III) which dates from 6000 to 4400 cal BP (McGee and Wheeler 1994). This p eat is thought to represent a seasonal marsh, which suggests a high water stand or an increase in the hydroperiod (Randall 2010). Rare artifacts within this stratum attest to shif ts in refuse disposal that likely relate to micro-environmental changes. Above this peat deposit is another dense Viviparus deposit dating to the end of the preceramic Ar chaic. These data not only demonstrate the variability in surface waters through time, but also demonstrate that much of the early record of the Preceramic Archaic lifeway s is likely submerged and covered along Floridas lakes and rivers. Ceremonialism was a widespread and prominent component of Mount Taylor lifeways, as evidenced by the construction of monumental shell mounds and mortuaryrelated sand mounds. Although traditionall y viewed as relatively late-period constructions or the result of mundane ac tivities, some Mount Taylor mounds were deliberately constructed as r itual and/or mortuary mounds as demonstrated by early observations by Jeffries Wyman (1875) a nd C.B. Moore (1999), and more recent excavations at Bluffton Burial Mound (8VO23) (Sears 1960), M ount Taylor (8VO19) mound (Wheeler et al. 2000), the Harris Creek site (8VO24) on Tick Island (Aten 1999), Live Oak Mound (8VO41) (Sassaman 2003b), Hontoon Dead Creek Mound (8VO214) (Sassaman 2005), and the Tomoka Mound co mplex (8VO81) (Piatek 1994) on the Tomoka River. Although Mount Taylor burials have been recorded in only a few cases, similarities in the form and internal struct ure of these mounds indicates that many if not all were mortuaries at one point in time (Endonino 2003). Although only a handful of mounds have been archaeological tested in modern times (Bluffton, Mount Taylor, Harris Cr eek, Live Oak, Tomoka, Hontoon Island North and Hontoon Dead Creek Mound, Thornhill Lake Silver Glen Run Locus A), many more no doubt existed prior to their destruction during the 20th century. That many of the

PAGE 38

Environmental and Archaeological Contexts 27 mounds contained preceramic deposits was well documented by Jeffries Wyman (1875). Wyman, then curator of Harvards Peabody Museum, made extensive collections and observations of shell-bearing sites thr oughout the middle St. Johns River between 1860 and 1873. Through pedestrian surveys and colle ctions, observations of cut-banks, and small excavations, Wyman recorded over 40 ridges, ridge complexes, and conical mounds throughout the basin. Later in the 19th century, C.B. Moore (1999) revisited many of these sites. His more intensive ex cavations provide both a confirmation of the preceramic origins of many mounds, as we ll as documentation of the stratigraphic sequences and mortuary natu re of these sites. Most ceremonial mounds share design el ements and internal sequences (Endonino 2003; Randall and Sassaman 2005; Wheeler et al. 2000). Many mounds are crescentshaped ridges, with steeply sloping sides and asymmetrical summits 5 to 11 m tall. Excavations at Hontoon Dead Creek Mound and Live Oak Mound determined that at least these two were erected rapidly, on the order of several hundred years at most, and composed primarily of shellfish remains (Randall 2010). The extent to which these mounds encase earlier mortuaries is unknown. Moreover, it is unlik ely that all mounds were constructed this rapidly, or for the same ceremonial purposes. The construction of mortuary mounds ma y be a practice as old as the Mount Taylor period. Early (Mount Taylor phase) mortuary mounds are best known from the Harris Creek site (8VO24). At Harris Creek, mortuary deposits have been dated between 7000 and 5900 cal BP. As related by Aten (1999), the Harris Cr eek mortuary was constructed by emplaced shell and white sand upon a preexisting shell deposit in small scale mortuary events. At least 140 individuals were interred here. Later Thornhill Phase mortuaries such as Bluffton and the T hornhill Lake mounds (8VO58) are round, truncated cones. Like the earlier mortuaries, these mounds tend to be erected on top of existing shell deposits. At Bluffton this la yer was intentionally burned (Sears 1960). Earthen mounds of sand (typically brown) or mu ck were then constructed on this midden. Burials were then placed into these depos its. Although grave goods are rare in early contexts (Aten 1999), Thornhill Lake phase burials (such as the type site) were interred with exotic artifacts. Subsequent to inte rment, these earthen mounds were frequently capped with shell, which in some cases wa s clearly excavated from preexisting midden deposits (Aten 1999; Piatek 1994). The importance of wetlands is evident not only in the placement of sites, but in the subsistence remains. Mount Taylor lif eways were characterized by a fishing and hunting subsistence economy. Faunal analysis at Groves Orange Midden (Russo et al. 1992; Wheeler and McGee 1994), Lake Monr oe Outlet Midden (Quitmyer 2001), Blue Spring Midden B (8VO43) (Sassaman 2003 b), and Salt Springs (Blessing 2011) demonstrate the dominance of aquatic specie s, which could have been acquired from marshes, slackwater lagoons, and sloughs. Although it was once thou ght that shellfish contributed a small percentage of the diet, recent studies indicate that between 33 and 98 percent of the dietary meat weight was derive d from freshwater shellfish. Studies have shown that shellfish diversity varies with s ite contexts, and may reflect local ecological variations (Quitmyer 2001). A diverse array of fish were collecte d, including catfishes,

PAGE 39

28 St. Johns Archaeological Field School 2007-2010: Silver Glen Run sunfish ( Lepomis sp.), gar (Lepisosteous sp. ), largemouth bass ( Micropterus salmoides ), and eel. Turtle was also collected, including such species as the soft shelled turtle ( Apalone ferox ), sliders, and mud/musk turtles. Where waterlogged conditions have enabled the preservation of plant matter, such as at Groves Orange Midden (Newsom 1994; Russo et al. 1992), Salt Springs (Talcott 2011), and Windover Pond (Newsom 2002) a st able pattern char acterized by high diversity is established by no la ter than 9500 years ago. Pulpy fruits such as black gum, prickly pear, saw palmetto, maypop, wild pl um, blackberry, persimmon, red mulberry, elderberry and grape appear to have been the most important (Newsom 2002). These fruits were supplemented with starchy seed s such as amaranth, pigweed, and knotweed, as well as the greens from these and other spec ies. Numerous tubers were potentially eaten. Cabbage palm hearts and shelf fungi have also been iden tified (Newsom 2002). Mount Taylor period assemblages are typified by mundane and decorative material culture manufactured from locally available bone fired clay, and wood, in addition to exotic materials (Wheeler et al. 2000). Bones from deer and other terrestrial animals were used to make a variety of tools including gouges, awls, needles, fids, projectile points, and decorati ve pins. Wooden tools have be en recovered from saturated deposits such as Groves Orange Midden (Wheeler and McGee 1994) and include tool handles and net floats. Fired clay objects of various shapes and sizes have also been recovered from numer ous contexts. Nonlocal materials used to manufacture to ols and items of adornment speak to the extensive trade networks in which Mount Ta ylor culture groups were engaged. Marine shell appears early in the Mount Taylor phase, and demonstrat es contact or movement to coastal regions. By the Thornhill Lake pha se, marine shell was a bundant, and in the case of Strombus gigas was apparently imported from southern Florida. Shell tool assemblages are dominated by woodworking tools, including Busycon sp. axes and adzes, as well as celts made from Strombus gigas shell. Marine shell was also used to make containers, which are often recovered with residue adhering to the interior surfaces. These are preferentially made from left-ope ning whelks, and may have been used for medicinal or ritual beverages (Sassaman et al. 2011b). Decorative shell artifacts are also typical, and include marine shell beads a nd plummets made from large whelk columella, as well as decorative shells such as Oliva sp. Shark teeth are often recovered. Many have been drilled to facilitate hafting for use as a tool or as personal adornment. Contact with the interior and west coast is demonstrated by the presence of lithic materials of nonlocal origin (Endonino 2007). There is no source for raw material for chipped stone tools in the St. Johns basin, and many artifacts appear to have been traded into the region as performs and finished forms. Hafted bifaces are consistent with those of the Newnan horizon. Aside from hafted bifaces, some Mount Taylor lithic assemblages contain microlithic tools that appear to have been us ed for the production of objects, potentially marine shell beads (Randall 2010). These appear to date to the Thornhill Lake phase based on excavations at Lake Monroe Outle t Midden (ACI and Ja nus Research 2001).

PAGE 40

Environmental and Archaeological Contexts 29 The presence of ground stone beads and bannerstones provides evidence for contacts far afield during the Thornhill Lake phase. Groundstone beads have been recovered from several mortuary and cache contexts, (Thornhill Lake mounds 1 and 2 and Coontie Island respectively) (Clausen 1964; Moore 1999). Although their origins are unknown, they are quite similar to tubular b eads produced in Mississippi and the Midsouth during the Middle Archaic. Bannerstone s have been recovered from several mound contexts, including Thornhill Lake, Tomoka Stone, and Coon tie Island. The forms are consistent in form and raw material with those manufactured in the middle Savannah River in Georgia and South Caroli na (Sassaman and Randall 2007). Orange (4600 cal BP) and Earl y St. Johns (3600 cal BP). The appearance of pottery in shel l middens of the St. Johns Ri ver and the Atlantic coast signals the end of the preceramic traditions and the beginning of the pottery-making traditions. Orange fiber-tempered pottery ha s been dated as early as 4800 years ago in the lower St. Johns, although po ttery does not appear in th e middle St. Johns until 200 years later (Sassaman 2003a). By 3600 years a go, fiber-tempered pottery ceases to be manufactured, signaling the end of the Or ange period, and is wholly replaced by spiculate-pasted wares. Once thought to be diagnostic of the St. Johns period, radiocarbon dates (Sassaman 2003a) and paste characterization studies (Cordell 2004) demonstrate that spiculate pottery was produc ed during the Orange period and continued through the end of the Late Archaic and into the St. Johns Period. Orange period lifeways have been portraye d as continuing the basic trends set in motion during the preceding Mount Taylor period (Milanich 1994:86). Excluding the production of pottery, and new hafted biface type s such as the Culbreath, Lafayette, Clay and Levy varieties, continuity is suggested by the continued use of ma rine shell and stone tools, although marine shell does appear in reduced frequency at some sites. As evidenced by subsistence data from Bl ue Spring Midden B (Sassaman 2003b) and Groves Orange Midden (Russo et al. 1992), communities continued to exploit aquatic habitats, routinely collecting from local she llfish beds and capturing fish and turtles. The economic importance of wetlands is demonstrated by the continued focus of settlement adjacent to the river. Orange sites are most likely to be found along productive wetlands and marshes, often in the same locales as earlier preceramic components, while there is a decrease in sites in the interior forests of northern Florida. The more numerous and larger Orange comp onents may very well reflect an overall increase in population. This observation, howev er, must be tempered by the fact that preceramic components may not be adequately recorded due to inundation, stratigraphic ambiguity, or a lack of diagnostic artifacts. Although there certainly is significant continuity, divergen ce in traditions within the St. Johns is evident duri ng Orange times (Sassaman 2004). The upper St. Johns is characterized by smaller sites that may, when taken as a w hole system, constitute yearlong settlement (Sigler-Eisenberg et al. 1985). In the lower St. Johns, large and presumably multi-seasonal settlements are surrounded by smaller, probable fishprocessing stations (Russo et al. 1993). In addition to these habitation areas, large shell

PAGE 41

30 St. Johns Archaeological Field School 2007-2010: Silver Glen Run rings have been identified both at the mout h of the St. Johns and along the coast (Russo and Heide 2001). These sites were likely acc retionally but intentionally constructed, and were the loci of communal feasting and ritu al activities (Russo 2004; Saunders 2004). Settlement in the middle St. Johns has been less well documented, but it appears to replicate Mount Taylor si te types, characterized by a dichotomy between extensive middens, mound complexes with abundant potter y, and small task sites. Because these sites have not been routinely investigated, da ta on their internal organization and function are scarce. Sassaman (2003c) has identified a possible Orange period semi-circular compound at Blue Spring Midden B. The co mpound was situated above a Mount Taylor midden and adjacent to a Mount Taylor moun d. At least three households and their associated refuse piles were inferred. Although seasonality data have not been forthcoming, the site was repeatedly occupied. Emerging new data, primarily from the Silver Glen locality, indicates that Orange communities in the middle St. Johns activel y mounded shell as their coastal neighbors did. In general, Orange pottery at mound s ites is rare. At Bl uffton the pottery was deposited adjacent to and not on top of the mound (Wheeler et al 2000). In excavations at Live Oak Mound, Sassaman (2003c) recovere d only a small number of sherds, all from near the surface. However, large quantities of decorated Orange pottery are present at several mounds, including the Mouth of Silver Glen Springs Run (8LA1), Harris Creek, Enterprise, and Orange Mound. In most cases it is unclear if the mounds were in the shape of a linear ridge or a U. Based on the observations of Wyman, however, the mound at the mouth of the SGS run was U-shap ed, although it remains to be determined when it attained that configuration (see Chapter 3). Orange fiber-tempered pottery has been treated primarily as a chronological marker. Bullen (1972) constructed five subperiods, based on changes in vessel construction and surface decora tion. The unilineal sequence consisted of a transition from Orange Plain to Incised (Orange Inci sed and Tick Island) wares, which were eventually replaced by spic ulate-tempered St. Johns In cised vessels. However, radiocarbon dates have shown th at variation in tempering agents, vessel form, and surface treatment likely reflect spatial variation in the production and use of pottery (Cordell 2004; Sassaman 2004), and not necessarily tem poral trends, as on ce thought. Sassaman (2004) suggests that village sites such as Blue Spring Midden B are dominated by plain pottery that was rarely used over fires, wh ile large and complex sites such as Harris Creek and Silver Glen Run are dominated by incised vessels that were routinely used over fire. He suggests, as Saunders (2004) does for the coasta l Orange shell rings, that the different distribution likely represents di fferent social contexts, where plain pottery was used in mundane contexts, and incised pottery was used prim arily during ceremony and communal feasts. Recent dates from 8LA1 on Tick Island Incised components indicate this variant post-dates classic Orange Incised vessels, a nd may be associated with domestic and ritual contexts (see Chapter 6). The recent upheaval in the chronol ogy and typology of fiberand spiculatetempered wares has left an approximately 1000-year gap between the Orange and St.

PAGE 42

Environmental and Archaeological Contexts 31 Johns I periods. A Transitional period was defined by Bullen as a bridge between primarily fiber-tempered assemblages and in cised spiculate-tempered wares (Milanich 1994:88). Isolating sites of th is period has remained proble matic (Miller 1998:76), likely because many of the wares thought to occur afte r the Orange period ar e actually coeval. Although the term Transitional should be disc arded, there is a need to document sites of this period. An early date on a spiculat e-tempered assemblage at the Joseph Reed Shell Ring (8MT13) in southern Florida i ndicates that this interval will likely be populated with components as more dates are acquired (Russo and Heide 2002). St. Johns (ca. 2800 cal BP) Although St. Johns pottery dates as early as 4400 cal BP, fully developed St. Johns lifeways began around 2800 cal BP, and c ontinued into European contact. The archaeological culture was defined by Goggin ( 1952), who used changes in pottery styles to identify subperiods. The St. Johns I (ca. 2800 cal BP), is typified by plain chalky spiculate-tempered wares, and th e St. Johns II (ca. 1300 cal BP), typified by plain and check-stamped varieties. These ceramic types are formally referred to as St. Johns Plain and St. Johns Check Stamped, respectively. Additional subperiods have been identified by the presence of foreign wares or local copies of them, as well as changes mortuary ritual (Milanich 1994:247). As Miller (1998:79) notes, however, these divisions are not easily traced because the diagnos tic artifacts or sites are rare. Although there are numerous changes in soci al organization, material culture, and ceremonialism that were incorpor ated from external contacts, the St. Johns period may be marked by conservatism (Miller 1998:78). In general it is assumed that along the St. Johns River, St. Johns I and to a certain extent St. Johns II lifeways continued seemingly unchanged from that of thei r late Archaic, Orange-per iod predecessors (Milanich 1994:254). This assumption is based primarily on the fact that St. Johns I components are likely to be found on sites with Orange components, and this tr end continues with a similar frequency of reoccupation for St. Johns II components (Miller 1998; Sassaman et al. 2000). However, there have been very few archaeological inve stigations of these post-Archaic components. The continuity made apparent by the reuse of locations may be superficial at best. In general, however villages, short-term task sites, and large ceremonial mounds are likely present along much of the St. Johns River and its tributaries, and along the coastal lagoons fr om Jacksonville into Brevard County. Although equally distributed on the coast and along the St. Johns, St Johns period sites are also located in interriverine localities. Increases in populati on from Orange to St. Johns II times are suggested by increases in sites per century. Unfortunately, village contexts have rarely been excavated, so it is unknown how large the residential populations of each these places ma y have been, nor how they may have been structured. Continuity with Orange period subsiste nce practices is a likely possibility. Coastal assemblages are dominated by oyster an d coquina, in addition to estuarine fishes (Milanich 1994: 257). Subsistence data from the St. Johns period wet-site deposits at 8VO202 on Hontoon Island indicate that populati ons continued to focus on the collection of aquatic resources, such as gar, catfish largemouth bass, allig ator, and turtle, in

PAGE 43

32 St. Johns Archaeological Field School 2007-2010: Silver Glen Run addition to Viviparus and bivalve (Wing and McKean 1987). A wide array of plants were also exploited, including many that were collected during the preceding Archaic (Newsom 1987). Cultigens that supported larg e populations and complex forms of social organization elsewhere in the Southeast occur in relatively limited frequencies. Bottle gourd ( Langeria siceria ) seeds and rind fragments and Cucurbita pepo gourd fragments were recovered in St. Johns II contexts, but these were likely used for containers or net floaters. Maize, a staple throughout much of the Southeast by St. Johns IIb times, was present in only historic contexts. Cultivation or encouraged gardening of cornmay have been practiced, but it does not appear to have been widespre ad or intensive in the middle St. Johns. Changes in material culture throughout St. Johns I and II times were primarily restricted to pottery decoration and hafted biface types (Milanich 1994:247, 263). Hafted bifaces were typically small and crude, and include the Jackson, Florida Copena, Bradford, Columbia, Broward, Taylor, West o, Florida Adena, Gadsen, Sarasota, and Ocala types (Bullen 1975). Plain St. Johns wares dominate St. Johns I components. Locally produced Dunns Creek Red vessels we re produced during Ia and Ib times, while during Ia copies of Deptford and Swift Creek and during Ib Weeden Island vessels were produced. These often were deposited in mort uary contexts. At ca. A.D. 750, potters began to apply check-stamped designs with wooden paddles. During IIa times, late Weeden Island pottery and copies were ma de, while elements of the Southeastern Ceremonial Complex are evident in IIb assemblages. During St. Johns IIa or IIb times, there is a shift to the use of small hafted bifaces such as Pi nellas, Ichetucknee, and Tampa Points. Other tools found throughout St. J ohns period assemblages were shell adzes, celts, picks and hammers. Bone tools include a variety of awls, pins, pendants, beads, and fishhooks. Ceremonial and political life appears to have been transformed in relation to external contacts (Goggin 1952, Milanich 1994 :260-262). Mounds of the St. Johns I period were low, truncated cones constr ucted of sand. Bundle burials, extended interments, and cremations we re placed into these mound. Many mounds were reused for multiple interments, which may indicate that interred individuals were members of the same lineage, as in Weeden Island mounds During the St. Johns Ia period, larger mounds were constructed, and exotic items such as galena and copper were interred, along with locally made St. Johns Plain and Dunns Creek Red pottery. Towards the end of Ia, Hopewell influences are evident in th e construction of log to mbs. Mounds of Ib age show evidence for Weeden Island influences. St. Johns IIa mortuary pr actices appear similar to earlier practices in that they con tinue to be used for multiple, likely kin-based burials (Milanich 1994:268). Beginning with the St. Johns IIb subperi od, the construction of mounds takes on a different character, and is clearly influenced by Mississippian cultures to the north and west. Although it is unknown precisely what leve l of social organiza tion was present at this time period, the symbolism and quantity of material cultu re is similar to chiefly societies elsewhere in the Southeast at this time. At least three large pyramidal mounds were present in the middle St. Johns basin, including Shields, Mt Royal, and the Thursby

PAGE 44

Environmental and Archaeological Contexts 33 Mound, located across the St. Johns channel fr om Hontoon Island. Thes e sites were large earthen works, likely constructe d in stages. C. B. Moore (1999) excavated all of these sites, and recovered caches of copper, galena, silver and gold, Busycon shells, greenstone celts, and clay vessels and effigies in add ition to scattered or poorly preserved human remains. CONCLUSION The configuration of the Silver Glen Sp rings watershed today reflects a complex and interwoven history of natural and cult ural processes reachi ng back nearly 12,000 years. The most visible components of this landscape are the four known shell and sand mounds that were constructed there. Yet th e preponderance of data indicates that most portions of the watershed were incorporated into the daily and ritual lives of its inhabitants. Although fragmente d by recent land use practice s, the area encompassed by the Juniper Club was among the more intensiv ely utilized areas since about 6000 years ago. It is to the archaeological investigat ion of these resources that we now turn.

PAGE 45

34 St. Johns Archaeological Field School 2007-2010: Silver Glen Run

PAGE 46

CHAPTER 3 MOUTH OF SILVER G LEN RUN (8LA1-EAST) Kenneth E. Sassaman The large shell deposit that was located at the mouth of Silver Glen Run is listed in the Florida Master Site File as 8LA1. It was first noted by William Bartram in his 1766 travels through northeast Florida (Bartram 1942:44), and more than a century later by Jeffries Wyman (1875:38-39) and J. Francis LeBaron (1884:774). Wymans account is the most detailed. The shellworks he desc ribed at the mouth of what was then called Silver Spring Creek were among the most gi gantic deposits of shells met with on the waters of the St. Johns (Wyman 1875:38). He goes on to describe a massive U-shaped construction on the south side of the run: The one last mentioned is much the larger and consists of three portions 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 hundred and fifty to two hundred feet in width, and the third extends inland at nearly right angles to this. Between th ese ridges is a deep valley, in which the shells are entirely wanting or are only sparingly found (Wyman 1875:39). In addition to this description, Wyman made a simple sketch plan of the shellworks during his visit (Figure 3-1), although it was never published. Asa Randall located the sketch in his review of Wyman s field notes, curated at the Countway Library of Medicine at Harvard University. Even though it does not provide much detail, and cannot be taken as a literal rendering, the sketch at least corroborates the general shape of the deposit given in Wymans description and thus provides a starting point for archaeological investigation. Having been mined for shell in 1923, the U-shaped shellworks at the mouth of Silver Glen are no longer visible on the surf ace. Still, subsurface contexts and adjacent waters hold clues to the pre-mining configuratio n of the deposit. In fact, the east end of the deposit, fronting the lake consists of exposed shel l both on shore and along a submerged ridge a few tens of meters into the lake. Wyman ( 1875:39) described this submerged feature as a beach wall, and righ tfully attributed it to wave action that eroded the shell mound. Shoreline erosion has no doubt altered much of the site since Wymans time, although most of its alteration can be attributed to mining activities. Despite massive alteration to the shoreline, we held out hope that three small islands at the mouth of the run were remnants of the orig inal north ridge. Be fore the shell deposit was mined, the run was narrow at the mouth, as it remains today at its midpoint just northwest of the club house. Thus, the mini ng operation not only removed virtually all of the above ground shell, it also reconfigured th e shoreline of the run and caused much of the basal portion of the north ridge to become submerged. Ou r testing on two of the three islands failed to locate intact deposits and s uggested instead that sh ell miners emplaced shell in these locations, perhaps as part of a reclamation effort to maintain fish habitat or to subdue erosion of the mainland shore. 35

PAGE 47

36 St. Johns Archaeological Field School 2007-2010: Silver Glen Run Figure 3-1. Digitized version of sketch map of U-shaped shellworks at 8LA1-East drawn by Jeffries Wyman, ca. 1875 (courtesy of Countway Li brary of Medicine, Harvard University, and Asa R. Randall). Little more is known about 8LA1 excep t that it contained an abundance of pottery, according to Wyman ( 1875:40). A large sample of sh erds from this location is curated at the Florida Museum of Natural History in Gainesville. Among the sherds are many examples of Orange Incised, a pottery t ype that is among the oldest in Florida. Soot samples scraped from the surface of three Orange Incised sherds were submitted for radiocarbon dating in 2002 and returned c onventional age estimates of 3680 60, 4020 60, and 4070 60 radiocarbon years before presen t (rcybp) (Sassaman 2003a). The latter two age estimates are especially notewort hy because they are among the oldest ever recorded for the type. These estimates also provide a minimum age for the shell ridges, although given recent work elsewhere in th e middle St. Johns regi on (e.g., Randall 2010; Sassaman 2003b; Randall and Sassaman 2005; Sassaman and Randall 2012), shell may have begun to accumulate as early as 7000 years ago. Nonetheless, 8LA1 is highly significant because it is one of only a fe w large shell deposits in the region with a sizeable Orange-period component. Wyman and other early visitors to Silver Glen Run did not make mention of the shell ridge to the immediate west of the U-shaped shellworks and only passing mention of additional shell deposits in what we now refer to as 8LA1-West. As discussed in Chapter 4, there is no discernable break in the distribution of subsurface archaeological remains between the east and west portions of 8LA1. To facilita te communication about

PAGE 48

Mouth of Silver Glen Run (8LA1-East) 37 different components of the site, we refer to the area of the U-shaped shellworks as 8LA1-East and treat it as a subunit of a larg er site for the purposes of this report. The initial goal of archaeological investig ations at 8LA1-East was to establish the distribution of subsurface remains across the entire landform through a program of augering. We were hopeful that enough of the base of the U-shaped deposit had survived mining to be detected simply by the presence of subsurface shell, and thus provide a means for inferring the original placement a nd orientation of the shellworks. In conjunction with augering, we began in 2007 a program of test unit excavation at the largest island in the run, as well as two locations in the presumed area of the south ridge. We doubled our efforts to locate remnants of the north ridge in 2008, but were largely unsuccessful. In 2010 we returned to 8LA1-East to continue testing in the area of the south ridge, this time in conjunction with Ground Penetrating Radar (GPR) and a program of close-interval coring. Although th e results of GPR and coring showed promise for locating portions of a possible circul ar village, controlled excavations yielded ambiguous results. Nonetheless, the combined investigations of 2007-2010 at 8LA1-East confirm the presence of a large, U-shaped shellworks as described by Wyman, and provided enough evidence to suggest that the south ridge was emplaced on a natural surface by users of Orange pottery. Efforts to locate intact portions of the north ridge at 8LA1-East generally failed, although we hasten to add that so much of this deposit now lies below the watertable and is thus inaccessible without dewatering. This chapter reports the methods and results of all archaeo logical investigations at 8LA1-East conducted by the St. Johns Ar chaeological Field Schools of 2007-2010, beginning with the establishm ent of a site-wide grid. SITE-WIDE GRID In 2007, the first year of investigation, a si te-wide grid was esta blished to provide horizontal and vertical spatial controls for al l aspects of fieldwork. An arbitrary datum was set about 20 m east of the southeast corner of the deck of the clubhouse. Designated Datum A, this point of reference was assi gned an arbitrary nort hing of 1000.00 m and an easting of 1000.00 m, with an arbitrary eleva tion of 10.00 m. A 4-ft long section of inch galvanized electrical condui t was driven into the ground at this location, eventually pushed in flush with the ground surface to prevent being dislodged. From this datum a cloth tape was pulled eastward across the la wn to a location near the bank of the easternmost pond and at 135.5 m a second section of conduit was driven into the ground to establish Datum B (N1000.00 E1135.50). A Nikon DTM-310 Total Station was used at Datum A to verify the taped distance to Datum B and to establish its elevation as 9.40 m. With this baseline established, the Total Station was used to collect data for topographic mapping and, over the course of multiyear investigations, to determine the coordinates of all subsurface te sts, surface finds, and various poi nts of reference. In due course, the grid system at 8LA1-East was extended via Total Station to 8LA1-West, where pairs of permanent data were establis hed at Loci A and B. The acquisition of

PAGE 49

38 St. Johns Archaeological Field School 2007-2010: Silver Glen Run LiDAR data in 2008 obviated the need to collect Total Station data for surface topography, and the use of high-resolution GPS data alleviated the need to locate all shovel tests, auger holes, and surface finds with the Total Station. However, all excavation units across all ar eas of 8LA1 were sited with the Total Station, which was likewise used to maintain three-dimensional controls for many of the point-plotted artifacts uncovered in the 2009 bl ock excavation at Locus B. AUGER SURVEY For the purpose of acquiring extensive subsurface data from the full extent of 8LA1-East, a series of augers were initiated in 2007 across the open terrain east and south of the clubhouse. Several augers were also sunk in the wooded area along the lakeshore, and on the largest of three isla nds (Island A) at the mouth the run. Two types of augers were used: a 6-cm diameter Dutch gouge au ger with a maximum reach of 1.2 m, and a 10-cm diameter bucket auger with extensions capable of reaching ~4 m. Subsurface shell deposits and related strata across the expansiv e lawn could be adequately characterized with the gouge auger, but the bucket auger was required along the lakeshore, at the confluence of the run and the lake, and on Island A. The dept h of shell deposits in these near-shore locations often exceeded 1.5 m in depth below the surface. Transects for auger sampling were orient ed parallel to the N1000 base line, spaced 20 m apart. Sampling along trans ects was conducted uniformly across all open terrain of 8LA1-East at an interval of 20 m (F igure 3-2). Sample points were determined by triangulating from baseline data with tw o cloth tapes. Afte r a sample point was augered, its location was established within the site grid with the To tal Station. All fill Figure 3-2. Field school students sinking a Dutch gouge auger into subsoil of area east of the clubhouse, July 2007.

PAGE 50

Mouth of Silver Glen Run (8LA1-East) 39 from augers was passed through -inch hardwa re cloth and any recovered artifacts or vertebrate fauna were bagged and labeled by transect and auger numbers. Recorded for each auger were observations on the presence/a bsence of shell, the depth and condition of shell (crushed, whole, burned), and the presence/absence of nonshell midden. The locations of 84 augers s unk in 2007 are displayed in Figure 3-3. As can be seen, sample coverage of 8LA1-East is bias ed toward the open, grassy portions of the site, and biased against its wooded and flooded portions to th e east. The latter area is very difficult to traverse due to an abundance of downed trees, mostly the victims of tornadic winds associated with one of three hurricanes in 2004. To minimize this bias in coverage, we surveyed the wooded area fo r tree throws in 2010 using a GPS unit to record locations. These data do not include estimates of the depth of shell or other midden below the surface, merely observations on the presence/absence of shell. We will review these observations following discussion of the auger results. Figure 3-3. Topographic map of the 8LA1-East area showing locations of augers and test units excavated in 2007 and 2008 (refer to Figure 34 for cross-sectional views of auger transects).

PAGE 51

40 St. Johns Archaeological Field School 2007-2010: Silver Glen Run Cross-sectional views of four transects of augers are provided in Figure 3-4. Starting with the northernmost cross-section (N1000 transect), we fi nd a surface that slopes downward gently to the east before turning back upward at the end of the landform, essentially the point of land that marks the confluence of Silver Glen Run and Lake George. This high point of land at the east end, as we discuss further below, appears to be a product of mining operations, most likely a loading platform for barges used to haul away shell. Irrespective of this surface modification, shell depth increases from west to east, reaching roughly 2 m below surface and about 1 m below the water table on July 11, 2007. Two of the augers of this transect encountered concreted shell ~25 cm below the surface. Patches of concrete d shell are evident at the surface just north of this transect, as well as in most of the augers that we re placed along the shoreline of the spring run (Transect N1020). Augers of the N1000 transect, as well as all augers north of this transect, penetrated shell matrix that ostensibly comprises the basal strata of the north ridge observed by Wyman. Because shell along th e spring run is often concreted and well below the water table to the east, we suspect that most of this basal shell was actually deposited during the Mount Taylor era, long before the U-shaped configuration took shape after ca. 4200 years ago. If so, Orange-period shell deposition along the north ridge would have been grafte d onto a ridge similar perhaps to Locus A of 8LA1-West (see Chapter 5). A second west-east transect (N880) shown in Figure 3-4 likewise dips to the east gently, but here the subsurface shell is relati vely thin (~50 cm BS) and its contact with underlying sand parallels the modern surface. We hasten to note that mining operations have altered all the surfaces shown in cros s-section, making it impossible to estimate the contours of emplaced shell before 1923. The N880 transect is located in the presumed area of the south ridge observed by Wyman. As we will see below, shell along this transect was emplaced directly on an old ground surface by people who also deposited Orange fiber-tempered pottery, mostly plai n, and dating from ca. 4050-3850 cal B.P. The lack of shell in augers at th e west end of this transect ma y signify the terminus of the south ridge, although additional shell is found in augers to the south, well beyond the expected width of the south ridge. Two north-south cross-sections in Figure 3-4 provide the best views of subsurface remains running perpendicular to the U-sh ape shellworks, showing clearly the area lacking shell in what should be the center of the deposit. Recall in the Wyman quote above that the center was a deep valley, in wh ich the shells are entirely wanting or are only sparingly found. This area is hardly a deep valley today, given that shell was removed on either side to form relatively flat terrain. Because subsur face strata were so variable in composition and structure in many of the shell-free augers in this central location, much of the valley may consist of redeposited fill. De spite possible infilling, the cross-sectional views show clearly that sh ell to the south (on what is presumably the south ridge) was emplaced on higher terrain than shell to the north, along the spring run. As alluded to earlier, we suspect that the south ridge was added during the Orange period to an existing landscape of shellworks that included a Mount Taylor ridge along the

PAGE 52

Mouth of Silver Glen Run (8LA1-East) 41 Figure 3-4. Cross-sectional views of two west -east auger transects (top) and two north-south transects showing surface topography, depth of shell, and presence of concreted shell. Vertical exaggeration x8. spring run. Incidentally, both no rth-south transects in Figure 3-4 show concreted shell in augers directly adjacent to the spring run, again an indirect measure of the greater antiquity of shell in this location. In sum, the results of augering enable the following conclusions: (1) shell deposits fronting Lake George are as much as 2.5 m in depth below the modern (mined) surface; (2) shell deposits fronting Silver Glen Run a nd Lake George contain numerous patches of concreted shell; (3) twenty-three augers l acking shell are concentrated in the south-

PAGE 53

42 St. Johns Archaeological Field School 2007-2010: Silver Glen Run central portion of open terrain; (4) additiona l shell deposits exist along the western margin of 8LA1-East, fronting a spring-fed swamp; (5) shell deposits in the purported location of the south ridge were placed on a low slope trending gently upward away from the run and lake; (6) significant archaeological deposits exist below the shell in several locations; (7) augers bearing shell in the pr esumptive center of th e shellworks are among the most variable of the sample, and likely refl ect considerable disturbance. On balance, the results of augering suggest that intact shell strata are deeply buried in the area of the north ridge and perhaps along the lakeshore, al though most, if not all intact shell strata may be below the modern water table. To the extent this is the case, the challenge will be to determine what shell, if any, was deposited on formerly dry land (and thus of Mount Taylor age) and what shell was deposited in nearshore waters (and presumably of Orange age). A second challenge is to determine the configuration and disposition of shell deposits in the location of the south ridge. Augering shows that shell was emplaced on a slightly elevated landform, apparently di rectly on a ground surface lacking older midden deposits. Test unit excavations in 2007 a nd 2008 were designed expressly to address these two challenges. TEST UNIT EXCAVATION: 2007-2008 Test unit excavations in 2007 and 2008 were designed to locate and sample intact subsurface shell deposits in the presumed locations of the north and south ridges of the U-shaped shellworks Wyman described in 1875. Test units in the ar ea of the north ridge were largely unsuccessful in this effort, wher eas those in the area of the south ridge were generally productive, albeit oc casionally ambiguous. Our report of this testing begins with units placed on islands at the mouth of the spring run. Islands at Mouth of Spring Run The effort to locate intact shell deposits on the islands at the mouth of Silver Glen Run began in 2007 with a single 1 x 2-m test un it on the largest of the three, Island A. The lack of success in this effort redirect ed our interest back towards the mainland, although a fallen tree at the west end of Island C offered hope that the smallest of the three islands and most proximate to the mainla nd, retained a bit of in tact shell mound. In 2008 we conducted limited testing on Island C to find that it too consisted of redeposited fill left by mining operations. Test Unit 2 A single 1 x 2-m excavation unit wa s placed in the center of Island A in an attempt to locate an intact portion of the north ridge. Island A, like its counterparts upstream in Silver Glen Run, was formed by the mining of shell in 1923. Presumably, before 1923, Island A was part of the northeast corner of the U-shap ed shellworks. The island today consists entirely of shell, w ith surface exposures of whole, unconsolidated Viviparus interspersed with patche s of crushed shell. Little soil development has taken place on the island due to the limited time since the island was formed 85 years ago. Test Unit 2 (hereafter TU2) was sited in the center of the island, at the topographic high of ca. 9.75 m, where a bucket auger placed one-half meter to the south

PAGE 54

Mouth of Silver Glen Run (8LA1-East) 43 revealed continuous shell deposits from immedi ately below the surface to at least 150 cm below surface. The water table was encountered in this auger at ca. 85-90 cm below the surface (8.79-8.74 m). Shoreline water level at the time the island was mapped measured approximately 8.75 m, consistent with th e observed water table in the auger. TU2 was excavated in 10-cm arbitrary levels using the ground surface at the southwest corner for a local datum (Figure 35). The upper three le vels were dominated by modern refuse, notably construction material s such as wall plaster, wire nails, and window glass. Island A today, and apparently since its formation, is the recipient of all sorts of modern refuse. Bottles and cans, fishing tackle, and miscellaneous trash are routinely deposited on the island today by wate r and passers-by, but earlier last century the island also received sizable dumps of debris from mainland activities. Augering before test excavations commenced suggested that shell deposits below about 30 cm were undisturbed and varied fr om whole, unconsolidated shell, mostly Viviparus to lenses of finely crushed shell. Shell matrix was removed in zones defined with successive levels, although after Level C, it became apparent that shell was laid down in cross-bedded strata, suggesting fluvial reworking of the deposits. Two other observations supported this conclusion. First, alternating whole and crushed shell strata were both thoroughly winnowed of sediment and sorted into discrete depositional units. Second, recovered sherds and vertebrate faunal remains showed an advanced degree of water erosion. Photographs and line drawings of all four profiles of TU are given in Figures 3-6 through 3-8, and Table 3-1 provides descriptions of each stratum. Figure 3-5. Excavation of Test Unit 2 in the center of Island A, 8LA1-East.

PAGE 55

44 St. Johns Archaeological Field School 2007-2010: Silver Glen Run Figure 3-6. Photograph and line drawing of south profile of Test Unit 2, 8LA1-East.

PAGE 56

Mouth of Silver Glen Run (8LA1-East) 45 Table 3-1. Stratigraphic Units of Test Unit 2, 8LA1-East Max. Depth Munsell Stratum (cm BS) Color Description I 15 10YR2/1 black, very fine sand surface humus with rootlets II 24 10YR4/2 dark grayish brown fine-medium sand with light grey (10YR7/1) fine sand mottles and traces of Viviparus shell III 41 10YR3/1 whole Viviparus shell in very dark gray fine sand with clayey texture (possibly degraded shell) IV 49 n/a whole and crushed Viviparus shell lacking clastic matrix (grades horizontally into Stratum III in east corner of south profile) V 63 n/a whole, unconsolidated Viviparus shell with trace of Unionid shell and occasional iron staining, lacking clastic matrix VI 71 n/a crushed Viviparus and Unionid shell lacking clastic matrix VII 79+ n/a whole and crushed Viviparus shell lacking clastic matrix VIII 86+ 10YR3/2 whole, unconsolidated Viviparus shell with trace of Unionid shell, lacking clastic matrix IX 49 10YR3/1 very dark gray me dium-coarse sand with historic-era refuse X 20 10YR2/0 black coarse sand XI 23 10YR6/2 light brownish gray fine sand with hist oric-era refuse (grades horizontally into Stratum II in west corner of north profile) XII 33 10YR2/2 very dark brown fine sand with whole Viviparus XIII 28 n/a crushed Viviparus and Unionid shell lacking clastic matrix XIV 53 n/a crushed Viviparus and Unionid shell lacking clastic matrix

PAGE 57

46 St. Johns Archaeological Field School 2007-2010: Silver Glen Run

PAGE 58

Mouth of Silver Glen Run (8LA1-East) 47 As the profiles show, TU2 penetrated what appears to be redeposited shell matrix, presumably the result of shell mining. The integrity of profiles was compromised by the collapse of unconsolidated, whole shell strata, especially af ter encountering the watertable at ca. 85 cm below surface. Excavat ion ceased at this point but deeper strata were sampled by two 2-inch percussion cores driven well past the basal shell deposits and into underlying peat (see below). There is no indication in these prof iles and the strata observed in cores below the watertable that Island A c ontains intact, undisturbed archaeological deposits. The distribution of artifacts recovered fr om TU2 corroborate the inference that shell matrix in this location is redeposited (Table 3-2). Historic refuse so prevalent across the surface of the island extended well in to shell strata of TU 2. Artifacts deeper than ~35 cm below the surface were exclusively pre-Columbian in age, but the relative order of artifacts was inconsiste nt with chronologies establis hed for the region. Notably, St. Johns sherds were more numerous than th e (presumably) older Orange-period sherds in levels greater than 50 cm below surface. Whereas the relative age of these two wares is generally well known, they a pparently overlapped for several centuries. However, a single sherd of St. Johns Check Stamped potte ry in Level G (58-68 cmbd) attests to the reverse nature of stratigraphy in TU2. Check stamped St. Johns pottery is believed to post-date A.D. 750 (Milanich 1994:247). Anothe r indication of distur bance to the strata in TU2 is that many of the sherds we re waterworn, irrespective of type. Test Units 17 and 18 During the 2007 field school the author and various members of the staff occasionally inspected exposures along the shoreline of the mainland and island by jonboat to collect artifac ts eroding from shell matrix. On one trip we inspected the root mass of a tree that had fallen at th e west end of Island C. Contained in the root mass and the water im mediately below were sherds of Orange Incised pottery, most of which are shown be low in Figure 3-9. Other exposures on the islands and along the south shor eline of Silver Glen Run al so produced Orange pottery, as well as sherds of the St Johns tradition, but none compar ed to the density of Orange pottery in this fortuit ous exposure. Hopeful that this re flected the existenc e of an intact portion on shell mound on Island C, we opened in 2008 two small test units (50 x 50-cm) in the narrow spine of land that constitutes th is island (Figure 3-10). Located but a few meters apart, TUs 17 and 18 both produced good examples of both Orange Plain and Incised sherds (Figure 3-11) in a charcoal -rich dark sandy loam with shells of Viviparus, other aquatic snails, and occasional Unionids. Water was encountered in both units at about 60 cm below the surface. Percussion core s sunk in the base of both units provided good profiles of subaqueous matrix. Unfortuna tely, a radiocarbon assay on charcoal from the TU 18 core indicates that the entirety of Island C, like Island A, consists of redeposited fill (see section on cores below). Shell Point A third attempt to locate intact portions of the north ridge was made in 2008 with the excavation of a 2 x 2-m test unit on the high ground of Shell Point, just to the east of the clearing shown in Figure 3-3. Test Unit 7 (TU7) was sited just to the south of the

PAGE 59

48 St. Johns Archaeological Field School 2007-2010: Silver Glen Run Figure 3-9. Examples of Orange Incised sherds recovered from a tree tip-up at the west end of Island C (Bag# 609). easternmost auger hole along the N1000 transect where subsurface shell extended well past the watertable, measured at ca. 110 cm below surface in July 2007. Before excavating TU7, a second sounding with a perc ussion core was inserted in the high ground of Shell Point, 0.5 m north of the aforementioned auger hole. Shell-rich matrix extended down nearly a meter below the watert able and rested on what appeared to be intact terrace sands. Located about 10 m south of the core locat ion at Shell Point, TU7 was excavated in the usual fashion of 10-cm arbitrary leve ls. After removing an upper level containing modern refuse, seemingly intact shell matrix wa s encountered in the south end of the unit. However, the next four leve ls produced a confusing array of matrices, some containing modern refuse (mostly metal fragments), as well as Orange plain pottery, chert flakes, and a limited amount of vertebrate fauna. Li ne drawings (Figure 3-12) and photographs (Figure 3-13) of the profiles show how disc ombobulated the matri ces were. Groundwater was relatively high when TU7 was excavated in July 2008, preventing excavation deeper than ca. 80 cm below surface. Incidentally, seemingly intact shell matrix was observed near the bottom of the unit, labeled Str atum XXXIX in Fi gure 3-12 (note that

PAGE 60

Mouth of Silver Glen Run (8LA1-East) 49 Figure 3-10. Extracting percussion core from bottom of Test Unit 17, Island C, 8LA1-East. stratigraphic descriptions of strata in TU7 ar e not included in the usual table format given the lack of integrity; all such descriptions are available at the Laboratory of Southeastern Archaeology). In an effort to extract materials from below the watertable, a 1 x 1-m plywood form was constructed to insert in the center of TU7. After repeated attempts to push the form into subaqueous matrix with heavy equi pment, the plan was abandoned. In lieu of this strategy and some means of dewatering, only two levels could be removed from a 1 x 1-m subunit (TU7A) before matrix collapse d. Although stratigra phic controls were severely compromised at this depth, the subun it appears to have penetrated intact shell midden. Further consideration of intact matrix at Shell Point is reserved for discussion of percussion cores below.

PAGE 61

50 St. Johns Archaeological Field School 2007-2010: Silver Glen Run Figure 3-11. Examples of Orange Plain (top row) and Orange Incised sherds recovered from two small test units (TU 17 and 18) on Island C.

PAGE 62

Mouth of Silver Glen Run (8LA1-East) 51 Found throughout the redeposited fill of TU7 and into what appears to be intact matrix in TU7A, were sherds of Orange pot tery (Table 3-3). Unlike those from the islands and along the spring run, however, sherds from TU we re plain with one exception (Figure 3-14). Although many such sherds did not come from intact shell strata, the dominance of Orange Plain pottery in this general area corroborates the pattern seen in surface collections of the lake shore and testin g along the south ridge. All three of these locations, in contrast to the spring run, have produced as semblages of almost exclusively Orange Plain pottery. Given what we know of the age of Orange Incised and Plain Figure 3-12. Photographs and line drawings of all profiles of Test Unit 7, 8LA1-East.

PAGE 63

52 St. Johns Archaeological Field School 2007-2010: Silver Glen Run

PAGE 64

Mouth of Silver Glen Run (8LA1-East) 53 Figure 3-14. Orange pottery sherds from Test Unit 7, 8LA1-East. All sherds in this sample are Orange Plain with exception of a single Tick Island Incised sherd (upper left). pottery from the greater Silver Glen area, the emplacement of shell along the lake shore and the south ridge appears to be relatively late in the Orange sequence (i.e., post 3800 rcybp). Percussion Cores at She ll Point and the Islands Sampling of subaqueous deposits at Shell Point and on islands at the mouth of Silver Glen Run was enabled by the placement of several percussion cores. Coring was done by simply driving a 2-inch PVC pipe w ith a beveled edge into substrate with a sledge hammer. After reaching maximum de pth, the PVC pipe was filled with water, capped with a rubber stopper, then extracted with a winch attached to a tripod. Cores were then split with a circular saw, photographed, mapped, and extracted for water processing through a #35 geological sieve. Annot ated profiles of four cores are provided in Figure 3-15 through 3-18.

PAGE 65

54 St. Johns Archaeological Field School 2007-2010: Silver Glen Run Figure 3-15. Split percussion core from Test Unit 2 on Island A, 8LA1-East.

PAGE 66

Mouth of Silver Glen Run (8LA1-East) 55 Figure 3-16. Split percussion core from Test Unit 17 on Island C, 8LA1-East.

PAGE 67

56 St. Johns Archaeological Field School 2007-2010: Silver Glen Run Figure 3-17. Split percussion core from shovel test near Auger 11-1, Shell Point, 8LA1-East.

PAGE 68

Mouth of Silver Glen Run (8LA1-East) 57 Figure 3-18. Split percussion core from Test Unit 7, Shell Point, 8LA1-East. Cores sunk on islands at the mouth of Silver Glen Run extended at least 2 m below the watertable. One core (Core 1) in the base of TU2 on Island A (Figure 3-15) penetrated about 1.5 m of shell matrix overlying a degraded silty peat with diffuse stringers of fine sand. One AMS assay on the peat returned an age estimate of 1360 40 B.P. (calibrated at two-sigma range to 13301260 B.P. or A.D. 620-690). It goes without saying that the Orange pottery recovered from shell-bearing levels overlying this peat must have been redeposited, presumably in the course of shell-mining operations, but certainly sometime after the th irteenth century B.P. We have no reason to doubt that the peat is autochthonous, but it is hard to reconc ile its depth with a near-shore environment given such a late age. With so much la ndscape modification atte nding shell mining, as well as subsequent channel dredging, it is certai nly possible that peat in this profile, like the overlying shell, was displaced and redeposited.

PAGE 69

58 St. Johns Archaeological Field School 2007-2010: Silver Glen Run Two percussion cores were driven into th e base of test units 17 and 18 on Island C. The one core profile shown in Figure 316 is representative of both cores. Unlike Core 1 on Island A, the Island C cores contained shell-bearing matrix to nearly the base, where shell-free black muck was encountered ca. 230 cm below the water level. Shell strata varied in composition and density but te nded to be contained within a silty or very fine sandy muck or degraded peat with abundant charcoal. A 12to 21-cm-thick stratum of sand stringers interrupted nearly continuous shell-bearing strata in both cores at ca. 170 cm below the water level. Orange Incised sherds were observed in the core of TU18 in shell strata at depths of 138 and 157 cm be low the water level. Charcoal taken from the TU18 core at 200-228 cm below the wate r returned an AMS assay of 2440 40 B.P. (calibrated at two-sigma range to 2710-2350 B.P. or B.C. 760-400). Given that this age estimate postdates the Orange pottery found above, the shell strata of this core evidently were redeposited much like the shell of Island A. However, it seems likely that below the sand stringers at ca. 170 cm she ll strata are intact and undisturbed. If so, the age of this charcoal some 30-40 cm deeper suggest th at the on-shore shel l mound continued to accrete outward, into the spring run, well past the Orange period. The abundance of St. Johns pottery in the spring run to the west attests to intensive activity in the general area through at least 1200 B.P. Cores driven into the onshore surface of Shell Point likewise penetrated thick shell strata well below the watertable. Core 3 at the east terminus of the N1000 auger transect also penetrated deeply into terr ace sands beneath the shell (Figure 3-17). The contact between shell strata and sands is roughly 90 cm belo w the top of groundwater and some 205 cm below the present surface. S ubaqueous shell in this core exists in alternating layers of sand and peat/muck with varying amounts of charcoal and vertebrate fauna. Like the subaerial shell strata of TU 7, this varied sequence is likely the result of displacement and redeposition during shell mi ning. In fact, it would appear that the elevated landform of Shell Point was cons tructed during the mining operations, perhaps as a loading dock for barges to cart off shell via water. Given the redeposited shell is found today nearly one meter below the water level, miners must have dug deeply into the shell mound, perhaps even flooding the landf orm before backfilling and then building up a platform. One final core placed in the bottom of TU7 likewise reached terrace sands at about 90 cm below the water, but here the sh ell strata above appear to be undisturbed (Figure 3-18). Lacking in ca. 90 cm of shell deposits ov erlying sand are any of the alternating strata observed in Core 3. Coupled w ith evidence for intact shell strata at the base of TU7, the subaqueous shell of th is core provides some hope that basal anthropogenic deposits remain intact in at le ast a portion of the ridge connecting the north and south ridges of the U-shaped shellworks. It remains to be determined at a later date if these basal deposits were emplaced by people of the Orange period, or by their Mount Taylor predecessors. If the latter, shell was likely laid down on dry ground, when water levels were at least one meter lower than today. If, however, the basal shell was emplaced during the Orange periodas they seem to be along the south ridge (see below)then they would have likely been de posited into the waters edge, meaning that

PAGE 70

Mouth of Silver Glen Run (8LA1-East) 59 the shell works would have prograded out to wards the lake some 20-30 m since Orange times. In sum, test excavations and coring at She ll Point and at two of the islands at the mouth of Silver Glen Run did not produce much evidence for intact archaeological deposits. Subaqueous deposits beneath TU7 app ear to be intact, but elsewhere all shell appears to be displaced and redeposited, mo st likely during the 1923 mining operation. Figure 3-19 provides a schematic cross-sect ion of the landform extending from Shell Point to Island A. The ancient land surf ace beneath shell is reasonably well documented at Shell Point some 2 m below the present su rface and nearly 1 m below the watertable. We repeat that the elevated surface in this location is likely the result of shell-mining activity, specifically the creation of a loading platform for barges. All shell in the water and on the islands today appears to be rede posited. Given the age of peat on Island A and subaqueous shell beneath Island C, the original shellworks of 8LA1-east must have prograded outward into the run, and perhaps the lake, well past the Orange period. Unfortunatey, evidence of later activity ha s been thoroughly erased by mining. Small portions of basal shell of eith er Mount Taylor or Orange age await dewatering. In the meantime, remnants of the south ridge of 8LA1-East, residing on higher, drier ground, provide opportunity for docum enting basal shell deposits. Figure 3-19. Schematic cross-section facing east of landform extending from Island A to Shell Point, 8LA1-East.

PAGE 71

60 St. Johns Archaeological Field School 2007-2010: Silver Glen Run South Ridge The results of augering reported earlier enabled an informed assessment of the likely location and orientation of the south ri dge at 8LA1-East. Test Unit 1 was a 2 x 2m unit placed in what was deemed to be the cen ter of the south ridge, close to the present woods line east of the orange grove (Figure 3-3). A seco nd 2 x 2-m unit, Test Unit 3, was sited on the far western edge of the cl earing, where augering revealed subsurface shell in an area that was ostensibly to the we st of the terminus of the south ridge (Figure 3-3). Both units were excavated in 10-cm arbitrary levels and all fill passed through inch hardware cloth. Test Unit 1. Excavated in 2007, Test Unit 1 (TU1 ) revealed an intact profile of emplaced shell on a buried A horizon (i.e., old ground surface) overlying sterile, subsurface sands. As expected, the profile was truncated at the top by shell-mining, but other than some minor intrusiv e features dating to the modern era, the profile was in good shape and appears to represent a basal remnan t of the south ridge. Figure 3-20 provides photographs and line drawings of all four walls, and Table 3-4 gives the descriptions of all strata recognized therein. Generally whole, unconsolidated shell is concentrated in a single stratum (Stratum X) extending from beneath the surf ace stratum (Stratum I; plowzone?) to as much as 66 cm below surface. The upper porti on of this shell was truncated by mining, so we have no basis for inferring its original thickness. We do, however, feel confident in the inference that this sh ell was placed directly on an an cient ground surface, one with a well-developed A horizon (Stratum III). Recall from earlier discussion of the results of augering that the south ridge appears to have been emplaced over a raised part of the landform, perhaps something akin to a rive r terrace. This apparent A horizon rests conformably on a mantle of white fine sand, th e sterile substrate in this portion of the landform. In July 2007, then TU1 was excavat ed, the water table was encountered just beneath the maximum depth show in Fi gure 3-20, roughly 80 cm below surface. Interrupting what is otherwise a simple profile are a few intrusive features, only one of which may be an intact aboriginal featur e. Feature 1, in the nor theast corner of the unit, is a large pit whose point of original was apparently truncated by mining (Figure 320). Recognized as a zone of alternating shell and organically-enriched sand, this ca. meter-wide pit was formally designated a featur e at the base of Level F, ca. 60 cm below surface. The bottom of the pit appeared to be relatively flat, where dark gray sand with minor shell produced charcoal that returned an AMS assay of 3600 40 B.P. (4060-4050 and 3990-3830 cal B.P.). Also recovered from the basal stratum of the pit were several sherds of Orange Plain pottery, examples of which are given in Figure 3-21. Otherwise, only trace amounts of vertebrate fauna were recovered from zones attributed to this feature. On balance, the evidence points to a bone fide Orange pe riod pit feature similar to those found in abundance at 8LA1-West Locus B (see Chapter 6). However, the stratified fill of Feature 1 and apparent di sturbance along its northern margins makes it difficult to judge the original form and functi on of this feature. At a minimum, we can

PAGE 72

Mouth of Silver Glen Run (8LA1-East) 61 Figure 3-20. Photographs and line drawings of all profiles of Test Unit 1, 8LA1-East. suggest that it was a relatively large pit that emanated from a surface well above the buried A horizon, and thus postdates the em placement of shell on this old surface. Feature 2, seen in the south profile of TU 1, is an infilled posthole of the modern era, most likely a fence post (Figure 3-20). Its point of origin is clearly at or very near the modern ground surface.

PAGE 73

62 St. Johns Archaeological Field School 2007-2010: Silver Glen Run Table 3-4. Stratigraphic Units of Test Unit 1, 8LA1-East Max. Depth Munsell Stratum (cm BS) Color Description I 16 10YR3/1 very dark gray fine sand with root mat II 58 10YR4/1 whole and crushed Viviparus shell in dark gray sand (Feature 1) III 62 10YR2/1 whole Viviparus shell in black fine ashy sand (Feature 1) IV 40 n/a dense crushed shell (Feature 1) V 67 n/a whole Viviparus shell (Feature 1) VI 76 n/a organic and iron-stained Viviparus shell (Feature 1) VII 79 7.5YR3/0 whole Viviparus shell in very dark gray medium-coarse sand (Feature 1; 3600 40 B.P.) VIII 88 10YR4/1 dark gray fine-medium sand with sparse whole Viviparus shell (buried A horizon) IX 90+ 7.5YR8/0 white fine sand with mottles throughout X 66 10YR5/2 whole Viviparus shell in grayish brown sand XI 32 10YR3/1 very dark gray fi ne sand with sparse whole and crushed Viviparus shell XII 65 10YR7/2 light gray mineralized root XIII 84 10YR4/1 dark gray clayey sand with whole Viviparus shell XIV 82+ 10YR6/1 gray fine sand with whole Viviparus shell (Feature 2; historic fence post) XV 30 10YR5/2 whole Viviparus shell in grayish brown sand (Feature 1) A third feature, Feature 3, wa s at first considered to a second large pit, but after completing the excavation, it seems more likely th at this is the outco me of an intrusive disturbance involving either a burrowing animal or tree roots. The mineralized or concreted zone seen in the east profile of Fi gure 3-20 (Stratum XIII) recurs in other units of the south ridge area, specifica lly in places beneath thick sh ell strata. The plan area in the northwest corner of TU1 showed root or burrow casts, as well as numerous palm roots. However, the basal porti ons of this feature also produc ed more bivalve than in the overlying shell stratum, which was dominated by Viviparus shell. Incidentally, the large pit features of 8LA1-West Locus B containe d a disproportionate frequency of bivalve

PAGE 74

Mouth of Silver Glen Run (8LA1-East) 63 Figure 3-21. Examples of Orange Plain sherds from Test Unit 1, 8LA1-East. shell compared to surface midde ns and other accumulations. It thus remains possible that Feature 3 is akin to Feature 1 and th e Locus B pits, only badly disturbed. The artifact inventory and associated verteb rate fauna from TU1 is rather meager (Table 3-5). A tapered stemmed hafted biface was recovered from the base of Level C in a zone in the northeast corner that is arguabl y in the upper portion of Feature 1. Virtually all of the fiber-tempered sherds that could identified to type (a ll Orange Plain) were also from various levels of the northeast corner, which, together with definitive Orange Plain sherds at the base of the Feature 1, puts them well within the fill of the pit. The same can be said for much of the vertebrate fauna, although the total assemblage is admittedly small. The only other notable class of rec overed materials is lithic flakes, the vast majority of which (37 of 42) came from the lig ht-colored sands beneath the shell. All of the historic era materials and three of four St. Johns peri od sherds came from the upper strata, near the surface. In sum, TU1 revealed a relatively simple profile of emplaced whole Viviparus shell on a buried A horizon that was intercepted by at least one large pit feature dug during the late Orange peri od, ca. 3600 B.P. Aside from the fill of Feature 1, TU1 produced little in the way of material culture or food remains to suggest that shell accumulated on the old surface in the course of routine, domestic activities.

PAGE 75

64 St. Johns Archaeological Field School 2007-2010: Silver Glen Run

PAGE 76

Mouth of Silver Glen Run (8LA1-East) 65 Test Unit 3 Located about 50 m west of the pr ojected location of the south ridge, Test Unit 3 (TU3) produced profiles resembling those of TU1 in some respects, but with notable differences. Like TU1, TU3 contained a near-surface stratum of Viviparus shell that appears to have been truncated by mi ning. Missing in TU3, however, was any clear indication that the shell was emplaced on an ol d surface (A horizon). The subshell strata of TU3 were instead dominated by concreted sand and shell matrix, perhaps a product of the lower elevation of this location and its fluctuating, near-surface watertable. Also, TU3 did noit produce much evidence of an Or ange period component, and instead has more evidence than TU1 for a prepottery (M ount Taylor) assemblage. Photographs and line drawings of all four profiles of TU3 can be found in Figure 3-22; descriptions of the strata observed in these profiles are provided in Table 3-6, and all artifacts and vertebrate fauna recovered are summarized in Table 3-7. Figure 3-22. Photographs and line drawings of all profiles of Test Unit 3, 8LA1-East.

PAGE 77

66 St. Johns Archaeological Field School 2007-2010: Silver Glen Run Table 3-6. Stratigraphic Units of Test Unit 3, 8LA1-East Max. Depth Munsell Stratum (cm BS) Color Description I 12 10YR3/1 very dark gray medium sand with root mat II 28 10YR2/2 sparse Viviparus shell in very dark brown fine-medium sand III 45 10YR3/3 whole Viviparus shell in minimal matrix (dark brown fine-medium sand) IV 56 7.5YR5/2 brown clayey sand with whole Viviparus shell V 62+ 7.5YR7/0 concreted light gray clayey sand with whole Viviparus shell and crushed Unionid shell VI 64+ 7.5YR4/0 dark gray fine sand (saturated) VII 53 7.5YR3/0 concreted light gray clayey sand with whole Viviparus shell and crushed Unionid shell VIII 21 7.5YR3/2 dark brown medi um-coarse sand with whole and crushed Viviparus shell IX 50 10YR3/1 very dark gray me dium sand with whole, iron-stained Viviparus X 47 10YR7/3 concreted brown sand with whole Viviparus shell and occasional Unionid shell XI 62+ 10YR3/1 very dark grayish brown clayey sand with whole and crushed Viviparus shell No definitive features were identified in the excavation of TU3, although some of the concreted strata arguably are thermal f eatures. Whether these are anthropogenic or simply natural is unclear. Unlike the mineralized roots s een in TU1 and throughout many of the units excavated in 2010 (see below), the concreted strata of TU3 are massive and generally larger in both horizon tal and vertical dimensions, pl us they contain substantial amounts of shell, including minor lenses of crushed Unionid sh ell. The best example is seen in Stratum V in the north and east prof iles. Well over a meter wide and at least 30 cm thick, this stratum has a basin-shaped cr oss-section. Vertebra te faunal remains may have been recovered at greater frequency in this stratum than elsewhere in the unit, but that cannot be substantiated with the relativ ely limited assemblage available. Also, the only substantial Orange Plain sherds from TU3 came from the northeast quadrant of Level E, in the general vicinity of this stratum, but a direct association cannot be substantiated. The single hafted biface from TU3 was recovered just to the south of the concreted stratum, in light-colored, fine sa nd, where most of the lith ic artifacts in TU1

PAGE 78

Mouth of Silver Glen Run (8LA1-East) 67

PAGE 79

68 St. Johns Archaeological Field School 2007-2010: Silver Glen Run were recovered. The tapered stem form of this biface signifies a probable Mount Taylor affiliation, as does the marine shell frag ments scattered throughout the unit. Basinshaped pits have been docume nted at Mount Taylor contexts at 8LA1-West Locus B, and concreted shell matrix in general tends to date to the Mount Tayl or period (presumably because such components are often close to or below the watertable). In sum, TU3 would appear to reflect the accumulation of shell and related materials earlier than and independent of th e emplacement of shell along the south ridge during the late Orange period. Presumably Mount Taylor in age, the anthropogenic deposits of TU3 probably represent the use of wetland habitat to the immediately west, namely the slough separating the east and west portions of 8LA1. Additional testing in this area will be needed to verify the age of the deposit and develop better information on the types of activitie s taking place here. 2010 INVESTIGATIONS OF THE SOUTH RIDGE Efforts to locate remnants of the South Ridge at 8LA1-East were intensified in 2010 with additional subsurface testing and th e application of Ground Penetrating Radar (GPR). The results of our GPR survey are reported first. Ground Penetrating Radar Survey A program of Ground Penetrating Radar (GPR) was undertaken in 2010 courtesy of Richard Estabrook of the Florida Public Archaeological Network. Efforts to locate and characterize the south ridge of 8LA1-Eas t to this point produced mixed results and the scope of these initial effo rts was simply too small to enable strong inference about the location and orientation of the ridge. The advantage of using GPR was twofold: (1) large tracts of open terrain in the projected vicinity of the south ridge coul d be surveyed relatively quickly, and (2) data on subsurface st rata could be collected with no additional impact to the site. Of course, ground truthi ng the results of GPR would require additional subsurface tests, but these could be targeted at specific anomalies of the GPR readings, rather than placed randomly or based on limited data, such as augers. GPR was deployed in two different ways. We first simply ran a series of 30-mlong transects spaced 10 m apart across the area of the site believed to contain subsurface evidence of the northern edge of the south ridge. Second, we ran the GPR unit across a series of five contiguous blocks ranging in plan from 20 x 30 m to 7 x 10 m, using a method of orthogonal transects in order to produ ce Z slices or time slices of data for display. Figure 3-23 shows the locations of th e GPR transects and grids, as well as the locations of test units added in 2010 and the results of a tree throw survey that is discussed later below. Richard Estabrook provided the following description of the GPR equipment, how it was deployed, and how the resulting data were processed: A Geophysical Survey Systems, Inc. (GSSI) SIR-3000 GPR system was used to collect the data. The configuration included a 400 MHz antenna mounted in a

PAGE 80

Mouth of Silver Glen Run (8LA1-East) 69 three-wheel cart with distance calibration pr ovided by an on-board survey wheel. The radargrams were collected along transect spaced 50 cm apart within predefined grids of varying sizes. The pe rimeter of the grids were staked at 1-m intervals and fiberglass surveys ropes used to establish and maintain the transect rows. Initially, information from a single 20 x 30-m grid and nine separate linear transects was acquired. Later, we returned to collect data from an additional four grids. Because of an expansive evergr een tree in one portion of the area of interest, these data were collected as a se ries of contiguous blocks around the tree. Within each grid, the radar data was collect ed in a zigzag pattern with transects oriented in one direction. Once the data were collected, the grid was flipped 90 degrees and a second dataset perpendicular to the first then collected. The port-processing employed GPR-Slice software (Version 7). The GPR data were converted from their GSSI file format, regained, and processed through a low-pass filter. These data are presented as individual time slices or as an animated sequence of time slices showing ho w the anomalies vary by depth. In the color ramp scheme selected, red indicates areas of greater reflectivity and blue shows areas of lower reflectivity. Yellow and green represent intermediate reflectivity grades. Red regions on the time slices represent locations that reflected more wave energy, and they ar e thought to be areas of higher density shell concentrations. The results of preliminary transect survey are shown in Figure 3-24. This area was selected for survey given the likelihood that it encompa ssed the northern edge of the south ridge of shell. A slight crease in the surface topography of this vicinity lent some support to this inference, although this feature was ne ither completely linear nor continuous. Nonetheless, we were hopeful th at the GPR unit would detect the boundary between subsurface shell and the natural, sandy substrate. The two easternmost transects (top two in Figure 3-24) were s horter than the other seven du e to the saturated ground of a wetland depression extending southward from the shoreline of one of the ponds. Survey began with these eastern transects after calibrating the GPR unit to the results of Test Unit 1, reported earlier. The results of this preliminary survey did not produce unambiguous evidence for the northern edge of the south ridge. The mo st prominent feature in the output of these transects is the high reflectivity of the wetland area. Another prominent feature is the waterline running east-west across the area surv eyed, evident in Figure 3-24 as a series of parabolas. The only suggestive evidence for an e dge to the south ridge is seen in the high reflectivity of the south end of the easternmost transects. In the first two transects, high reflectivity is seen in relatively thin bands of red tilted gently toward the north. A break in this pattern in the second transect is most likely the backfilled TU1. The next three transects (to the west, shown in Figure 3-24 as the third to fifth transects from the top) show variable levels of reflectivity but each shows a break or sorts at about 10 m from the southern line of origin. This is perhaps the best evidence we have for a break in subsurface shell, although as we progress fa rther west, this pattern dissipates and we know from the prior augering that subsurface sh ell extends north of this projected line by at least 15 m. On balance, the preliminary transect survey offers little evidence for a

PAGE 81

70 St. Johns Archaeological Field School 2007-2010: Silver Glen Run Figure 3-23. Topographic map of 8LA1-East, show ing locations of GPR transects and grids, as well as test units excavated in 2010. subsurface edge to the south ridge, alt hough with better sampling and time-slice processing, this feature may indeed be detectable. Grid survey with GPR began in an open ar ea directly south of the transects with a 20 x 30-m block (Grid 1). Following proced ures described above by Estabrook, field school students dragged the GPR unit at 50-c m intervals both east-west and north-south across the grid. Four subsequent grids were surveyed on a sec ond visit to the site late in the 2010 field season. Although final post-processi ng of data from the grid surveys took place after completion of fieldwork, Estabrook provided preliminary results from Grid 1 to guide out efforts in subsurface testing. All ground truthing of GPR results in 2010 was confined to the Grid 1 area. Figure 3-25 provides time sli ce output for all five grid s in the range of 47-55 cm below surface. Thirty time slices from the surface to 177 cm below surface. The depths values are not literal, but rather a relative measure of reflectivity by depth, with some values attenuated vertically due to vari ations in matrix composition, density and

PAGE 82

Mouth of Silver Glen Run (8LA1-East) 71 Figure 3-24. GPR results from nine north-south transects placed to detect north edge of south ridge, 8La1-East.

PAGE 83

72 St. Johns Archaeological Field School 2007-2010: Silver Glen Run Figure 3-25. GPR results in five contiguous gr ids for time slice 47-55 cm below surface, with projected arc of anomalies consisting of dense shell. moisture. In displaying slices in the 47-55 cm range, we are emphasizing those with the greatest clarity of anomalies; slices ranging from 35-78 cm below surface show the same general pattern as the out put provided in Figure 3-25. The output of GPR survey in all five grids offers some tantalizing patterning. As seen in Figure 3-25, major anomalies cluster in four areas some 5-10 m in diameter each, which together form an arc with a projected diameter of 80 m. The northern-most cluster shows a void in reflectivity that corresponds with the 2 x 2-m unit excavated in 2007, TU1. This alone provides strong indication th at anomalies correlat e with dense shell, although the occurrence of mineralized roots and pit features in this test unit may lend a bit more complexity to the output. In any event, if the arcuate array of anomalies comprises a complete, unbroken circle, some 80 m in diameter, we would expect a total of 16-17 clusters, each spaced about 15 m apar t on center. We hasten to add, however, that the clusters are of variable size and shape, and the middle two in Figure 3-25 may actually converge in the strip of unsurveye d land between two of the blocks. Also noteworthy is the lesser anomaly in the sout heast corner of the composite grid, where edge-matching of output form adjacent grids is problematic. Still, the area both outside and inside the projected arc of anomalous clusters is dominated by low reflectivity.

PAGE 84

Mouth of Silver Glen Run (8LA1-East) 73 Ground truthing the GPR results was accomp lished by a combination of coring and select controlled excavations, both in the area of Grid 1. Coring was accomplished with the use of an Oakfield soil tube, which is a foot-long, -inch diameter, chromeplated steel tube with an ope n face with a threaded fitting for extensions and a T-top handle. The entire 20 x 30-m area of Grid 1 was cored at 1-meter intervals along eastwest transects spaced 1 meter apart. Record ed for each of the 651 tubes inserted along these transects were observations on depth of shell, shell density, substrate beneath shell (if present), and maximum depth of core. Shell density was recorded as low, lowmedium, medium, medium-high, and high, if pres ent, and converted to a numerical scale of 1-5, with 1 for low and 5 for high density. Figure 3-26 provides interpolat ed output of shell dens ity and maximum depth of shell from the soil tube da ta using Surfer mapping softwa re (v. 6.01). Comparing these results to the time slice of Grid 1 used in Figure 3-25, we find relatively good conformity between GPR results and shell density, as we expected. The patterns are not a precise match, but there is general agreement between the arcuate shape of the GPR anomalies and shell density, event to the extent that shell density is low in areas to the inside and outside the projected arc. In contrast, depth of shell is more variable, with numerous occurrences in excess of 50 cm in the southern portion of th e grid. If such occurrences reflect the presence shell-filled pits, such as the one described in TU1 (see above), then pit features fall well outside the projected arc of GPR anomalies. In sum, the results of GPR show patterni ng in the distribution of anomalies that suggests the presence of a circ ular or arcuate arrangement of shell features. Although the pattern is far from clear, the results point to th e possibility of a circul ar village akin to the Late Archaic shell rings of the Gulf and A tlantic coasts (Russo and Heide 2001). Those known for Florida, such as Horrs Island near Fort Myers (Russo 1991), are nearly as large as the U-shaped configuration Wyma n observed at 8LA1-East. Others from Georgia and South Carolina tend to be smalle r and generally fully enclosed, perhaps a better model for the pattern suggested by the GPR data. Whereas we did not expect to find a circular village under th e shell of the south ridge, th e practice of capping old settlements with shell is not all that unusual for the region (Randall 2010). It will take a considerable subsurface testing to substantia te the existence of houses or households, let alone a complete village. That process got underway in 2010 with the excavation of several test units in various locations of the south ridge. Test Unit Excavation The locations of test units in and ar ound the area of GPR survey are shown in Figure 3-23. Two contiguous units (TUs 47 and 52) were placed to examine what was perceived to be the northern edge of the s outh ridge; four contiguous units (TUs 53-56) were excavated to investigate GPR anomalies in Grid 1; and two contiguous units (TUs 50-51) were placed to examine an area west of the GPR survey grids where incised Orange fiber-tempered pottery was found in an auger test. An account of the method and results of these tests follows below in the order just given.

PAGE 85

74 St. Johns Archaeological Field School 2007-2010: Silver Glen Run

PAGE 86

Mouth of Silver Glen Run (8LA1-East) 75 Test Units 47 and 52 As GPR survey got underw ay early in the 2010 field season, three contiguous 2 x 2-m units were laid out in the projected area of the south ridge where a slight crease in the surf ace topography hinted at the position and orientation of the north edge of the ridge. Ultimately, only two of the three units were excavated: Test Units 47 and 52 (TU47 an d TU52). Revealed in both units was a discontinuous stratum of shell beneath a well-defined plowzone, and underlain by lightcolored fine sands with a sparse but divers e assemblage of flaked stone artifacts. Definitive evidence for the north edge of the south ridge was not observed, although one of the long profiles of the contiguous units prov ided a hint of this feature. Photographs and line drawings of all pr ofiles of TU47 and TU52 are provided in Figure 3-27. Description of the strata mapped in these pr ofiles are given in Tabl e 3-8, and Table 3-9 gives an inventory of the artifact s and vertebrate fauna recovered. The upper stratum in both units was removed as a single level (Level A) to reveal well defined plow scars at the base of each unit, ranging up to 28 cm below the surface. Excavation thereafter proceeded in 10-cm arbitrary levels thr ough the shell stratum (Stratum II) and into the underlying light-color ed sands (Stratum III). All fill was passed through -inch hardware cloth. Planviews of levels in TU47 revealed matrix of varying composition, which was generally divided into zones as excavation proceeded. Subplowzone matrix in TU52 was compara tively simpler than in TU47, with the exception of the west profile, which revealed an attenuation of Stratu m II from TU 47. It was along this eastern profile in TU52 that we observed the only good indication of a terminus to emplaced shell. This can be seen in Figure 3-27 as a basin-shaped, shellfilled depression1 coterminus with dark gray fine sand (Stratum IV) intrusive to the surrounding gray sand substrate (Stratum III). A second zone of dark gray sand (Stratum IV) was observed a bit farther north of this cont act in the west profile of TU52, and a third in the north profile of TU52. The latt er zone was designated Feature 66, but it, like the two along the west profile, appear to be recent intrusive disturbances. One other notable aspect of the TU47 profile is the large, oval-shaped zone in the south profile designated Stratum V. Throughout excavation this zone was described as a clay or clayey sand, and was believed to be emplaced by either natural or human agents. After seeing several similar features in othe r test units of the south ridge, we came to understand this as the diagentic outcome of tree roots that we re covered in shell. Put another way, Stratum V is a mineralized ro ot mass whose source of mineral was the calcium carbonate that leached fr om overlying shell. In sup port of this supposition it is noteworthy that such features are not found outside of areas of overlying shell deposition. Given the size and shape of th e mass mapped as Stratum V, we suspect this particular example is the mineralized root ball of a palm tree. Additional examples are seen in the profiles of Test Units 55 and 56, discussed further below. The artifact inventory from level excavati on of TUs 47 and 52 is relatively sparse (Table 3-9). Materials recovered from the plowzone and shell stratum were limited to a few lithic flakes, some retouched, two crumb sherds, and a small assemblage of 1 recorded in the field as Feature 47, this shallow basin appears to be simply a low area of shell accumulation, as opposed to a pu rposefully dug and filled pit.

PAGE 87

76 St. Johns Archaeological Field School 2007-2010: Silver Glen Run

PAGE 88

Mouth of Silver Glen Run (8LA1-East) 77 Table 3-8. Stratigraphic Units of Test Units 47 and 52, 8LA1-East Max. Depth Munsell Stratum (cm BS) Color Description I 28 10YR4/2 dark grayish brown fine sand with minor Viviparus shell (plowzone) II 37 10YR5/3 whole and crushed Viviparus in brown fine sand III 71+ 10YR6/1 gray fine sand with no shell grading to 10YR5/3 brown fine sand with mottling due to fluctuating watertable IV 38+ 10YR4/1 dark gray fine sand with no shell V 46 10YR8/2 very pale brown fine clayey sand with 10YR5/2 grayish brown fine sand mottles and 10YR6/6 brownish yellow moist clay (mineralized root ball) Feat. 66 46 10YR4/2 dark grayish brown fine sand with no shell vertebrate fauna. Notably, most of the lith ic artifacts came from the light-colored sands beneath the shell. After heavy rains flooded th e units and caused the profile damage seen in Figure 3-27, TU47 was excavated a bit deeper to see how far lithic artifacts extending into the sandy substrate. Three additional 10-cm levels were removed before the receding water table was reached at 70 cm be low surface. Recovered in these additional levels were 54 chert flakes greater than -inch in size (students screening the sand from these levels noted an abundance of microflakes falling through the screen), and six chert tools, five of which are illustrated in Figure 3-28, along with tools from other units in the south ridge area. Many of the tools in this figure are unif acially modified flakes, some in forms archaeologists generally refer to as scrapers (e.g., Figure 3-28k-m). Another recurrent form is seen in the top row of Fi gure 3-28 (a-f). These small, pointed objects resemble the microliths of Mount Taylor affi nity (e.g., Randall et al. 2011; cf. Jaketown perforators ACI 2001:2-8), and were likely used as drills. The larger pieces in Figure 328 (o, p, r, s) are unifacially modified flak es, two from TU47; two examples of biface fragment (Figure 3-28n, q) are also illustrate d, including the haft element of a tapered stemmed point from TU52. The lithic assemblage recovered mostly from subshell sands reflects the broader distribution of Early and Middle Archaic artifacts across much of the 8LA1 site area. A similar pattern was detected on the north side of Silver Glen Run, where field school students assisted in a U.S. Forest Service project to assess the impacts of infrastructure repair to the recreational f acilities of Silver Glen Run (Randall et al. 2011). These artifacts reflect relatively intensive use of the greater Silver Glen area well before shell was deposited anywhere at the site, possible as early as the early Holocene, 9000 or more years ago. The Mount Taylor assemblage of microliths, as well as the tapered stemmed biface fragment is not unexpected of an ar chaeological complex that includes a massive

PAGE 89

78 St. Johns Archaeological Field School 2007-2010: Silver Glen Run

PAGE 90

Mouth of Silver Glen Run (8LA1-East) 79 Figure 3-28. Flaked stone artifacts from va rious test units in th e South Ridge area of 8LA1-East (a. 47D-3; b-c. 54I-2; d. 54H-3; e. 54 North G-2; f. 47C Zone D-3; g. 52C-2; h. 52A-4; i. 53F-2; j. 53D-3; k. 47D-3; l. 53G Zone D-2; m. 47F-3; n. 53G Zone B-2; o. 54J Zone A-3; p. 54K Zone A-3; q. 52A-3; r. 47E-1; s. 47E-2). Mount Taylor shell ridge in Locus A, and possibly along the spring run in the eastern aspect of 8LA1.

PAGE 91

80 St. Johns Archaeological Field School 2007-2010: Silver Glen Run In sum, TUs 47 and 52 provided a glimpse, in its west profiles, of the terminus of shell at the possible north edge of the sout h ridge. Because this observation was not duplicated in the east profiles, we are reluctant to put too fine a point on this inference. Irrespective of this ambiguity, TU47 provided two other useful observations: (1) mineralized roots in the sandy substrate are a good proxy for overlying shell, which is especially useful in contexts where shell ha ve been removed recently; and (2) the sandy substrate of 8LA1-East, like other locations in the greater Silver Glen area, encases a robust record of earlier site us e in the form of diverse cher t tools and the by-products of their manufacture and use. So much of this record now lies beneath the water table and will thus require dewatering to be adequately sampled. Test Units 53-56 A series of four contiguous 2 x 2-m units were excavated in the northeast corner of GPR Grid 1 to explore anomalies believed to be indicative of dense subsurface shell (Figure 3-25). The first unit, Test Unit 53 (TU53) was sited directly over a small anomaly, while three other units (TUs 54-56) were aligned offset to the south of TU53 to examine the edge of a larger anomaly. Together the four units provide a eight-meter-long profil e of the area, the largest verti cal exposure to date. With one exception, each of the units was excavated in 10-cm arbitrary levels and all fill passed through -inch hardware cloth. The ex ception was excavation of TU55, which was conducted within observed arch aeostratigraphy enab led by the leapfrog excavation of adjacent units (Figure 3-29). All four profiles of TU53 are illustrated in Figure 3-30. Prof iles that guided the excavation of TU55 are given in Figure 3-31, and the east profiles of TUs 54-56 are given in Figure 3-32. Descrip tion of the strata mapped in al l four test uni ts can be found in Table 3-10. An inventory of all artifacts and vertebrate fauna re covered for these test units is provided in Table 3-11. The profiles of TU53 are relatively simp le save for a few intrusive features (Figure 3-30). The south a nd west profiles of this uni t provide good perspective with minimal disturbance. Beneath the plowzone (Stratum I) is a stratum of grayish-brown sand with varying amoun ts of mostly whole Viviparus shell with limited vertebrate fauna. This shell-bearing stratum (Stratum II) is underlain by the light-col ored sands (Stratum III) seen elsewhere across the south ridge area. Unlike TU1, to the east, this shell stratum does not rest on a buried A horizon, but is inst ead directly atop the subsurface sands. Aboriginal artifacts recovered from TU53 c onsist of 47 chert flak es, 6 chert tools, 60 sherds, and a modest assemblage of vertebra te fauna. All but 6 of the 62 sherds came from the upper strata, 50 from the shell itse lf, most notably a cluster of Orange Incised sherds from the base of Level C (30 cm belo w surface) (Figure 3-33). These sherds were conjoined in the lab to form a relatively la rge upper rim portion of an open bowl with an estimated orifice diameter of 36 cm (Figure 3-34 ). The exterior surface of this vessel is badly eroded, although traces of its rectilinear incisions are visi ble in preserved patches. The admittedly cryptic incisions would appear to consist of nested triangles or diamonds, motifs not uncommon to Orange Incised pottery in the region. However, unlike the many Orange Incised sherds found at the mouth of S ilver Glen Run (see above), this particular

PAGE 92

Mouth of Silver Glen Run (8LA1-East) 81 Figure 3-29. Excavation of Test Units 53-56 in location of GPR Grid 1, 8LA1-East.

PAGE 93

82 St. Johns Archaeological Field School 2007-2010: Silver Glen Run Figure 3-30. Photographs and line drawings of all profiles of Test Unit 53, 8LA1-East.

PAGE 94

Mouth of Silver Glen Run (8LA1-East) 83 Figure 3-31. Photographs and line drawings of the south profile of Test Unit 56 (left), and the north profile of Test Unit 54 (right), 8LA1-East. vessel does not contain obvious traces of sponge spicules, making it more similar, technologically, to the Orange Plain pottery found elsewhere in the south ridge area (e.g., TU1), as well as Locus B (see Chapter 6). The intrusions evident in profiles of TU53 are instructive. The north profile (Figure 3-30) shows a large in -filled tunnel intruding throug h the shell stratum and into subshell sands at an angle of about 20 degrees A gopher tortoise is the most likely agent of intrusion in this case. Over the year s we have witnessed many excavations by gopher tortoises and they consistently enter the earth at about this angle. The depth and configuration of burrows varies depending on substrate and dept h of water table, but they are generally straight, as in th is example. A second intrusive feature is seen in the east profile of TU53. Consisting of two convergin g in-filled wedges this intrusion most likely comes from historic-era activity. Li ke the gopher tortoise intrusion, the contact between in-filled sediment and the surrounding matrix is sharp, indicating they were relatively recent. To the extent both were truncated by the plowz one, these intrusions must predate the last time the site was plow ed, but certainly postdate the mining of shell in 1923.

PAGE 95

84 St. Johns Archaeological Field School 2007-2010: Silver Glen Run

PAGE 96

Mouth of Silver Glen Run (8LA1-East) 85 Table 3-10. Stratigraphic Units of Test Units 53-56, 8LA1-East Max. Depth Munsell Stratum (cm BS) Color Description I 25 10YR3/1 very dark gray fine sand with minor Viviparus shell (plowzone) II 55 10YR4/2 whole and crushed Viviparus in dark grayish brown medium sand grading to 10YR5/1 gray medium sand towards bottom of stratum IIa 33 10YR5/3 mostly crushed Viviparus shell in brown medium sand (TU56 South and TU54 North only) IIa 90 whole and crushed Viviparus in matrix on varying color and texture; intrusive feature (TU54 and 55 East only) IIb 62 10YR5/1 fine-medium gray sand laminated with 10YR4/1 dark gray sand; gopher tortoise burrow (TU54 only) IIb 62 10YR4/2 whole and crushed Viviparus in dark grayish brown medium sand (TU56 South and TU54 North only) III 102+ 10YR5/1 gray fine-mediu m sand with no shell but abundant mineralized roots when shell dense in overlying stratum IIIa 62 10YR6/2 light brownish gr ay medium sand with no shell (TU53 only) IIIa 99+ 10YR4/2 dark grayish brow n medium sand with no shell (TU54 North only) IIIb 75+ 10YR4/1 dark gray sand mo ttled with 10YR6/2 light brownish gray medium sand; gopher tortoise burrow (TU54 only) IV 40 10YR6/1 gray medium sand with no shell (TU53 only) IV 57 10YR4/1 dark gray mediums and no shell (TU54 East only) V 90 10YR7/1 light gray medium sand mottled in upper half with 10YR5/2 grayish brown sand; no shell VI 79 10YR4/2 dark grayish brown medium sand with no shell VII 100+ 10YR5/1 gray medium sand with no shell VIII 100+ 10YR5/1-2 gray to grayish brown fine-medium sand with no shell IX 100+ 10YR7/2 light gray medium sand with no shell

PAGE 97

86 St. Johns Archaeological Field School 2007-2010: Silver Glen Run

PAGE 98

Mouth of Silver Glen Run (8LA1-East) 87

PAGE 99

88 St. Johns Archaeological Field School 2007-2010: Silver Glen Run

PAGE 100

Mouth of Silver Glen Run (8LA1-East) 89 Figure 3-33. Plan view of Test Unit 53 at the b ase of Level f, showing cluster of Orange Incised pottery, 8LA1-East. The profiles of TUs 54-56 are a bit more complicated than those of TU53, owing to an abundance of mineralized roots and at least one massive intrus ive feature (Figure 332). The general sequence of plowzone-shell-sand seen in TU53 characterizes the profiles of TUs 54-56, but in th e northern two units (TUs 55 and 56) the shell stratum is generally thicker and denser than in TU53, and the sand stra tum is dominated by an array of mineralized roots well below the shell. The southern unit in this trio, TU54, generally lacks the shell stratum and is instead dominated by an apparent pit filled with light gray sand. Each of these two deviations from the typical profile is disc ussed in turn below. Mineralized roots are especi ally abundant throughout th e sand stratum at depths of roughly 55-80 cm below surface. Figure 335 provides an example of what these look like in plan at the base of Level E (60 cm below surface) in TU56. As noted for the

PAGE 101

90 St. Johns Archaeological Field School 2007-2010: Silver Glen Run Figure 3-34. Reconstructed rim portion of Orange Incised vessel from Level C of Test Unit 53, 8LA1-East. mineralized root mass in the south profile of TU47 (Figure 3-27), the round masses in TU56 are most likely the root balls of palm tr ees. The linear masses see in this plan are instead the likely the mineralized consequence of hardwood tree roots. In either case, the degree of mineralization appears to be directly correlated with the density and thickness of overlying shell. Insofar as shell in these profiles was truncated by mining operations, the underlying roots or root balls must have been from trees that either died naturally before the shell was emplaced or were felled by those who emplaced the shell. When we consider the lack of an A horiz on beneath the shell, the possibi lity of deliberate felling of trees grows stronger. Th at is, the evidence for minerali zed roots and la ck of A horizon suggest strongly that the ground surface of th e south ridge area was prepared for the emplacement of shell.

PAGE 102

Mouth of Silver Glen Run (8LA1-East) 91 Figure 3-35. Mineralized root masses at the base of Level E, Test Unit 56, 8LA1-East. Turning now to the light gray sand seen in the east pr ofile of TU54 (F igure 3-32), it would appear that a large pit was excavated into the shell and underlying sands and then in-filled with sand lacki ng shell. Running under the north end of this in -filled pit is a stratum of displaced shell (Stratum IIa) that appears to have originated from the shell stratum (II) beneath the plowzone. On firs t inspection this deeper shell appeared intrusive; in fact, it had the hallmarks of an ancient tortoise burrow that was backfilled with shell. However, in other exposures afforded by the excavation of TU54, the relationship between displaced shell and the light gray sand goes well beyond what would be expected in a tortoise burrow. This is evident in the sectioned plan of the unit at the base of Level F (Figure 3-36). The upper view in Figure 3-36 shows the plan at 40 cm below surface (base of Level D), where the dark grayish brown matrix with shell (Startum II) stands in contrast to the light grey sand lacking shell. The bottom shows a sectioned TU54, with the southern half take n down an additional 20 cm (60 cm BS) to the base of Level F. Observed in both the pl an and profile of this cut is displaced shell following the basin-shaped outline of the light grey sand. It would thus appear that a large pit was excavated into the shell a nd underlying sand and before the pit was backfilled with light gray sand a good bit of the unconsolidated shall matrix dropped into the pit, forming an inverse talus slope of sorts. Based on these exposures, the pit measured at least 2 m in diameter at the t op and at least 1 m in diameter at the base.

PAGE 103

92 St. Johns Archaeological Field School 2007-2010: Silver Glen Run Figure 3-36. Plan view of Test Unit 54 at the base of Level D (top) and sectioned southern half at the base of Level F (bottom), 8LA1-East.

PAGE 104

Mouth of Silver Glen Run (8LA1-East) 93 Figure 3-37. Cross-section view of Feature 63, Test Unit 54, 8LA1-East. The cultural affiliation of this pit feature is difficult to infer based on stratigraphic principle alone. However, a small feature intrusive to the sand provides a good terminus ante quem for the backfilling of the pit. Seen in the TU54 plan as a circular dark stain in the north-central part of the light gray sa nd, this feature has the hallmarks of a burned post. Dubbed Feature 63, this apparent burned post has a well defined basal cross-section and diffuse margins (Figure 3-37). A sample of wood charcoal from this feature was submitted for an AMS assay and returned an age estimate of 670 40 B.P. (cal AD 12701330/ AD 1340-1400. This placed the burned post in the St. Johns IIb subperiod, coeval with a pit feature from 8LA1-West Locus C containing the diagnostic check-stamped pottery of St. Johns II times (see Chapter 7). We can infer from this age that the pit was dug and backfilled sometime prior to the placemen t and burning of this post. More than likely, the pit was dug, backfilled, and post emplaced at about the same time. If so, this portion of 8LA1-east was the locus of subs tantial landscape modifi cation long after the emplacement of shell along the south ridge. St. Johns Check-Stamped sherds are common in the water of the spring run, so a component of this age is not unexpected. One such sherd was recovered from the plowzone of TU56, four meters north of the in-filled pit. Bearing in mind that some portion of the sout h ridge was truncated by mining, a late date for the burned post suggests that the premining surface of the south ridge in the vicinity of TUs 53-56 was not all that high, perhaps well under a meter.

PAGE 105

94 St. Johns Archaeological Field School 2007-2010: Silver Glen Run The overall inventory of artifacts from TU s 54-56 is unremarkable (Table 3-11). Once again we see a tendency for vertebrate fa una, sparse as it is, to be concentrated in the shell stratum, and for chert flakes and tools to be concentrated in the subshell sands. Few diagnostic Orange period sherds were recovered in these units, although crumb sherds were pervasive, if not numerous, and te nded to be fiber tempered. On balance, the assemblage is consistent with others from th e south ridge area, with the addition of a St. Johns II component represented more so by the sand-filled feature, rather than numerous artifacts. Of course, even a minor amount of shell mining in this area would have removed the latest components of this site, so the lack of more St. Johns II material in the immediate vicinity is not surprising. One final note on the excavation of TUs 5356 speaks to the vulnerability of the profiles to slumping and collapse after heavy rains. Late in the field season we experienced a deluge that floode d all units in the s outh ridge area. Considerable damage was inflicted on TU54, whose pr ofiles collapsed soon after water receded (Figure 3-38). We were fortunate to have r ecorded much of the stratigr aphic information before the collapse, but did lose the opportunity to phot ograph and map the south and west walls of this unit. Figure 3-38. Collapsed profiles in Test Unit 54, 8LA1-East.

PAGE 106

Mouth of Silver Glen Run (8LA1-East) 95 Test Units 50-51. A third location of subsurface testing in 2010 was opportunistic. Augering in the orange grove to the west of the GPR grids produced an Orange Incised sherd in shell matrix. To investigate this occurrence a 1 x 1-m unit (TU50) was placed adjacent to the auger hole, and it was soon expanded to a 1 x 2-m unit with the addition of a second unit (TU51). A ll four profiles of TU s 50-51 are illustrated in Figure 3-39. Description of the strata ma pped in these profiles can be found in Table 3-12, and an inventory of all ar tifacts and vertebrate fauna rec overed for these test units is provided in Table 3-13. Figure 3-39. Photographs and line drawings of all profiles of Test Units 50 and 51, 8LA1-East.

PAGE 107

96 St. Johns Archaeological Field School 2007-2010: Silver Glen Run The profiles of TUs 50-51 deviate a bit fr om those found elsewhere in the south ridge area. Beneath a thin plowzone (Stratum I) are shell deposits, but they appear to be restricted to shallow basin-shaped features, the best example being Feature 46 in the west profile. Others are not so well defined, and in the east and south profiles of these units, several recent intrusions are apparent. It does not appear that shell was emplaced on a buried A horizon, as seen in TU 1, but instead in pits close to the present-day surface. It follows that shell may not have been mined fr om this area of the site, but that is not altogether clear. Despite the recovery of Orange Incised pottery in the nearby auger test, TUs 5051 did not produce much pottery. The recove red assemblage consists of only a small number of Orange Plain sherds, as well as cr umb sherds, mostly from the shell matrix. What little vertebrate fauna r ecovered was concentrated in the shell as well. Feature 46 contained little other than shell and a small am ount of vertebrate fauna As we have seen throughout the area to the south ridge, chert flakes and tools are found primarily in the subshell sands. Table 3-12. Stratigraphic Units of Test Units 50 and 51, 8LA1-East Max. Depth Munsell Stratum (cm BS) Color Description I 17 10YR3/2 very dark grayish brown fine to medium sand with moderate Viviparus shell (plowzone) Feat. 46 43 10YR3/2 whole Viviparus shell in very dark grayish brown fine to medium sand II 38 10YR4/3 whole Viviparus shell in brown fine to medium sand III 21 10YR3/3 dark yellowish brown fine to medium sand with trace of Viviparus shell IV 32 10YR5/4 yellowish brown fine to mediums and grading eastward to 10YR3/2 very dark grayish brown fine to medium sand with moderate whole and crushed Viviparus shell V 60 10YR6/3 pale brown fine to medium sand with no shell VII 21 10YR3/3 dark brown fine to medium sand with minor Viviparus shell (recent intrusion) VIII 52+ variable 10YR3/2 very dark grayish brown fine to medium sand, with 10YR5/3 brown and 10YR7/1 light gray sand stringers with trace of Viviparus shell (recent intrusion) IX 41 10YR3/3 whole and crushed Viviparus shell in dark brown fine to medium sand (recent intrusion)

PAGE 108

Mouth of Silver Glen Run (8LA1-East) 97

PAGE 109

98 St. Johns Archaeological Field School 2007-2010: Silver Glen Run In sum, the results of testing in TUs 50-51 reinforce the observation made in 2007 that the distribution of shell and shell-filled p its is not restricted to the projected location of the south ridge, but instead extends westward to the ed ge of the landform fronting wetlands. This does not mean that the south ridge, before mining, was ill-defined or simply graded into the natural contours of the landform, only that there is no abrupt termination to subsurface shel l in this portion of the site. CONCLUSION Field school investigations at 8LA1 -East in 2007, 2008, and 2010 produced mixed results. One the one hand, subsurface testing across much of the area believed to be occupied by a massive U-shaped shellworks revealed a great deal of disturbance owing to shell-mining operations in the 1920s, particularly along the shoreline of Silver Glen Run, the presumed location of the north ridge. C oupled with recent re view of Lake County probate records pertaining to the mining (Ra ndall et al. 2011), subsurface tests on two of the islands at the mouth of the run and at Shell Point suggest that more than shell was removed in the operation. Apparently, mini ng involved severe dredging of the spring run, as well as large-scale sculpting of the shoreline to accommodate barges and other equipment needed to remove the shell. Some of the county doc uments relate to a settlement among all parties involved over th e unauthorized excavation of the landform below the water table. The construction of a ramp at Shell Point appears to have enabled use of this portion of the landform for loadi ng of shell onto barges that apparently were brought into a slip cut well into the shoreline. The deposition of shell as islands at the mouth of the run may have been an atte mpt on the part of the mining company to ameliorate damage inflicted by the constructi on of this slip and ramp. A similar ramplike feature has been identified by Randall up th e north side on the north shore, land now under jurisdiction of the U.S. Fore st Service (Randall et al. 2011). Whereas the shoreline of the spring run and adjoining lake shore to the east appears to have been obliterated by mini ng, subsurface remnants of the north ridge appear to remain intact beneath the water ta ble of the mainland. Much of this resides below concreted shell and will thus require not only dewate ring to excavate, but also considerably energy to break up concreted sh ell. From experience elsewhere in the region, we suspect that the concreted shell and what lies beneath it will date to the Mount Taylor period. The subseque nt Orange period component an d St. Johns components that followed may be completely removed from th e north ridge. Abunda nt pottery of these periods has been recovered from beneath the wa ter of the run, but litt le has been observed on the adjacent land. The south ridge of 8LA1-East presents an altogether diffe rent challenge. Augering and limited controlled excavation in 2007 provided good evidence for the position and orientation of the south ridge, but it also showed that subsurface shell extended well beyond the projected western edge of the ridge. Subsurface shell observed in the profiles of TU1 suggest ed that shell along the south ridge was emplaced directly over an existing surface, with a well deve loped A horizon. However, subsurface shell elsewhere appears to have been placed over inorganic sands, the natural substrate of the

PAGE 110

Mouth of Silver Glen Run (8LA1-East) 99 landform. Throughout the area of the south ridge, shell-filled pits extended below the old surface, into the sands below, but it is never obv ious if these originated from the original ground surface, or from above, in the empl aced shell that was removed through mining. The application of Ground Penetrating Radar (GPR) in 2010 complicated the picture by offering evidence for a circular or ar cuate arrangement of shell features in the area of the south ridge. Circular villages of Orange age are not unexpected, as this is the configuration of shell rings on the coast, and we have circumstantial evidence for circular village sof this era at Blue Springs in Volusia County (Sassaman et al. 2003), and in the immediate upland area of Silv er Glen Run north of 8LA1-West (Randall et al. 2011). That being said, we continue to be disappo inted in subsurface te sting by the lack of obvious domestic features expected of a village occupation (e.g., heaths, house floors, post holes, etc.). To complicate matters, th e sand-field pit found in TU54 contains a St. Johns II period burned post. To the extent this re lates to domestic ac tivity, the circular arrangements of anomalies found by GPR ma y have more to do with late period dwellings than it does th e Orange period. However, it is equally possible that the entirety of 8LA1-East since Orange times was devoted to ritual activities that simply did not involve the sorts of domestic features and re fuse we expect from relatively permanent dwelling. On a positive note, the combined effort s of subsurface testing at 8LA1-East confirms the inference made since 2007 that the south ridge was added well after the formation of the north ridge and that this ac tivity resulted in a c oncentration of Orange Plain pottery in the former area and Ora nge Incised pottery in the latter area. Investigations of 8LA1-West Locus B by Zack Gilmore (see Chapter 6) addresses this pattern directly, lending creden ce to the hypothesis that the construction of the U-shaped shellworks was a multistage process involving several cultural constituencies, some perhaps nonlocal. We also learned through subsurface test ing of GPR anomalies that mineralized roots in the sandy substrate of the south ridge area offer a good proxy for overlying shell. This of course may be critical in the ongoing reconstruction of the pre-mined landscape because subsurface sands with mineralized ro ots likely escaped mining even in locations where mining was thorough in removing overlying sh ell. One such area in particular is seen in the wooded wetlands to the west of the orange grove. The survey of tree throws in 2010 showed that shell is absent across an area that is fully within the projected location of the south ridge, but which appa rently was mined aggr essively, leaving no trace of shell. Finally, because mineralized r oots occur in locations that ostensibly received large quantities of shell, we must consider further the possibility that the landscape was denuded of vegetation (and in some areas the A horizon) before shell as emplaced. This implies that the construction of the south ri dge was not only purpo seful but involved a greater amount of labor and effort than ev er imagined. No matter how badly damaged the U-shaped shellworks Wyman observed over 135 years ago may be, it remains a testament to the complexity and scale of Orange communities in the region.

PAGE 111

100 St. Johns Archaeological Field School 2007-2010: Silver Glen Run

PAGE 112

CHAPTER 4 RECONNAISSANCE SURVEY OF 8LA1-WEST Asa R. Randall During the 2007 and 2008 seasons, field school students conducted a shovel test reconnaissance survey of the western aspect of the Juniper Clubs spring-run fronting property. This area is designated 8LA1-West, and is entered into the FMSF as site 8LA1/8MR3601 (it spans both Lake and Marion c ounties). The western aspect of the site was once classified as site 8MR123. We have reassigned these por tions to 8MR3601 to reflect the sites position on th e south side of the run. This chapter describes the methods and results of this work. SURVEY SCOPE AND METHODS The primary goal of the field schools r econnaissance efforts was to document the extent, character, and culture-historical affiliation of cultural resources west of 8LA1East (Figure 4-1). This tract is positione d along a ca. 600-m long segment of Silver Glen Springs Run. It is bordered to the east by a linear wetland that a ppears to be a seep spring. This feature effectively separates 8LA1-East from the western landform (see Chapter 3). The western margin of the su rvey tract approaches the property boundary between the Juniper Club and the U.S. Forest Service (USFS). There is no natural border to the south. Interest in this tract was init ially piqued during a site visit in January 2007, when we observed extensive surf icial shell deposits up to 70 m from the run, in addition to pottery sherds scattered on the surface. Although it was ha rd to judge how deep most of the shell deposits extended, we observe d 2-m-tall vertical exposures of shell consistent with shell mining operations seen elsewhere along the St. Johns Rivernear the intersection of th e run and the linear wetland. The current configuration of this tract reflects a complex history of long-term geomorphic processes, ancient depositional practices, historic forest clearing, shell mining operations, and contemporary landscape maintenance (Figure 4-1). In general, the landform rises from the spring run (ca. 0 m) up to 9.5 m in absolute elevation1. The lower margin fronts open water (to the west) and wetlands characterized by cypress and other bottomland species (to the east). The slop e is relatively steep along the spring run margin, where the landform rises sharply 4 m in absolute elevation before attenuating into a gradual slope. There are several negative and positive surf ace features of note in the tract. Two depressions are present on this portion of the Juniper Club property. One is located to the far east, and is approximately 20-m in di ameter and 1-m deep. Another larger depression, ca. 60-m wide and 2-m deep, is situated to the nor thwest of this feature. These depressions presumably represent anci ent sink holes that ha ve undergone extensive infilling. This hypothesis needs to be thorough ly tested, however, as the depressions may 1 All absolute elevations in this chapter are derived from the 2006 Volusia County LiDAR survey, and are referenced to the North American Vertical Datum of 1988 (NAVD88). 101

PAGE 113

102 St. Johns Archaeological Field School 2007 Figure 4-1. Geography of the shovel test reconnaissance project area, highlighting the relationship of surface topography, cur rent trails, and forest clearings identified from aerial phot ographs in 1941 (yellow) and 2006 (green).

PAGE 114

Reconnaissance Survey 103 alternatively be related to Mount Taylor-era sediment removal (see Chapter 5). Positive surface features likely represent both geomorphic and anthropogenic processes. Several of the more distinctive subareas were given locus designations for ease of communication and description. Along the west ern aspect of the tract are two ridge noses, between 6 and 8 m in absolute elevati on, that overlook the spring run. This ridge complex is designated Locus C. To the eas t of these ridges, and some 75 m from the water, is an elongated dome that rises ca. 1 to 2 m above the surrounding terrain. This dome and surrounding terrain is designated Locus B. As is detailed in Chapter 6, this dome formed or was accentuated by the de position of shell and other materials in antiquity. Approximately 120 m southwest of the Locus B dome is a small conical sand mound, roughly 20 m in diameter and 1.5 m hi gh. In shape and scale this mound is consistent with other post-Archaic burial m ounds along the St. Johns. However, this temporal attribution has yet to be indepe ndently documented. This mound may also be the one opened by C. B. Moore in 1894, althoug h there is limited evidence on the surface for such an excavation. There is a slight depression around this mound that has the appearance of a borrow pit, and it may have resulted from the mounds construction. Finally, the northeast corner of the tract is characterized by variegated topography in an area roughly 200-m long (east-west ) and 100-m wide (north-south) It was in this area that we first observed deep shell escarpm ents and concreted surficial shell, both hallmarks of a mined shell mound. This area is designated Locus A (see Chapter 5). As discussed in Chapter 2, the north a nd south sides of the spring run were targeted for shell mining in the 1920s a nd 1930s. Aerial photographs from 1941 show how this process involved clear-c utting much of the terrace. As highlighted in Figure 41, evidence for a bare ground surface (light color) is present up to 120 m south of the run. Between 1941 and 2006, forest (mostly composed of juniper trees) was allowed to grow across much of the terrace. There are, howev er, two clearings that are maintained as bait fields. A tractor is used to disc-h arrow the clearings at least once a year, after which the fields are planted with grasses to encourage deer to fora ge. There is also a linear power line co rridor that courses through the surv ey tract from east (Locus A) to west (Locus C). Finally, distributed across the tract is a system of dirt and shell trails. These trails are approximately 3 to 5-m wi de, and are maintained primarily by chain dragging. A shovel test pit (STP) survey strategy was devised to provide coverage across the once-cleared terrace, from the eastern we tland to near the we stern property boundary (Figure 4-2). We first established an east-west baseline oriented relative to magnetic north. Starting at the small eastern depression, we tested 34 transects that were oriented north and south of this baseli ne. These transects were spaced at 20-m intervals. The majority of STPs within transects were also tested at 20-m intervals. The exception to this spacing is in portions of transects predominantly in the east (#2, 4, 6, 8, 9, 10, 11, 12, 13, 14, and 32), where the north-south spacing varied between 20 and 40 m. In addition, we tested around the sand mound in a crucifor m pattern. Transects were stopped to the north when either water or saturated deposits were encountered. To the south, transects were generally stopped when we ceased intercep ting shell-bearing deposits, or if artifact densities decreased substantially. Because our goal was to characterize near-water

PAGE 115

104 St. Johns Archaeological Field School 2007 Figure 4-2. Shovel test results.

PAGE 116

Reconnaissance Survey 105 deposits, we did not attempt to bound the sout hern border of the site. Although generally characterized by infrequent finds, this component will require further work to determine its full spatial extent and character. Shovel testing on the northern side of the run indicates that extensive shell-free deposits are present within the watershed (Randall et al. 2011). Shovel test pit nomenclature followed a standard system. North-south transects were each given a numeric designation. Within each transect, STPs were given a unique numeric designation. These two designations are combined in the final STP identifier (e.g., Transect 13, STP 1 = STP 13-1). STP excavation and data recording follo wed a standard protocol. Each STP measured 30 x 30-cm wide in plan. In genera l, they were excavated to a maximum depth of 1 m. In several cases STPs intercepted im penetrable concreted sh ell deposits, in others saturated deposits precluded further excavation. During excavation, all sediment was passed through 1/4-inch screen. All pre-Columbian cultural materials retained in the screen were bagged for subse quent analysis. A select sample of historic and modern materials (i.e. brick fragments, glass, plastic, and the like) was also kept. The stratigraphy in each STP was recorded, including the depth below surface for the top and bottom of each stratum and a description of the matrix When encountered, shell deposits were categorized using a subjective ordinal scale of low density (more non-shell matrix than shell), high density (more shell than non-she ll matrix), and concreted (shell and other matrix cemented together). After excavation and recording were completed, STPs were backfilled. During the survey, STPs were sited with a compass and distances between STPs were measured out by pacing. Precise STP spatial location was acquired in a variety of ways. The location of each STP was recorded on field maps. A subset of STPs also had their position located with a Magellan Meridian Platinum handheld GPS unit. The position of others was captured with a Nikon DTM-310 to tal station. All of these data were merged together in GIS. The resultant locations have an estimated +/5 m horizontal accuracy. SURVEY RESULTS During the reconnaissance surv ey the field crew tested an irregularly shaped area measuring 680-m along an east-west axis and 450-m on a north-south axis, and covering roughly 11.6 hectares. Within this survey tr act we excavated a total of 238 STPs (Figure 4-2). Of this total, 189 en countered pre-Columbian artifact -bearing strata and 36 yielded historic or modern materials. Summaries of objects recovered are presented in Table 4-1, while an enumeration of objects recovered from each STP is presented in Table 4-2. Several object classes were routinely recovered. The lithic assemblage is dominated by debitage (n = 394). A significantly smaller number of stage bifaces, hafted bifaces, modified flakes, and unifaces were also en countered. The pottery assemblage included all varieties typically found within the St. J ohns basin. The Archaic pottery assemblage included both Orange Plain (n = 65) and Incise d (n = 17) varieties. The Post-Archaic assemblage is dominated by St. Johns Plain (n = 530), but also incl udes St. Johns Check Stamped (n = 71), and other minority types su ch as sand tempered plain (n = 31). The majority of pottery sherds were classified as crumb sherds, those that are less than 1/2inch in minimum dimension (n = 1197). The zooarchaeological assemblage is composed predominantly of unmodified vertebrate faunal bone. In addition, we recovered three

PAGE 117

106 St. Johns Archaeological Field School 2007 Table 4-1. Summary of Artifacts Recovered from Shovel Tests. Category Count Weight (g) Lithics Biface 2 5.9 Hafted Biface 4 24.5 Modified Flake 4 10.2 Uniface 4 63.0 Debitage 394 421.0 Pottery Orange Incised 17 50.6 Orange Plain 65 191.6 St. Johns Plain 530 2006.1 St. Johns Check Stamped 71 396.1 Misc. Pottery 31 83.9 Crumb Sherd 1197 624.1 Marine Shell 21 98.8 Vertebrate Fauna 3650 2153.9 Modified Bone 3 8.3 Historic 89 618.1 pieces of modified bone and 21 marine shell fragments. Fi nally, 89 modern or historic objects were also retrie ved during testing. Shell Deposit Distribution The coverage provided by the STP survey allows us to consider the distribution and variety of cultural deposits across the landform. As the most visible evidence of ancient depositional practices, the presence and character of shell deposits provides an entry point into discussing the spatial distribution of anthropogenic deposition. Shellbearing deposits were recorded in a total of 113 STPs. The density of shell recorded during testing was used to generate the dist ribution of shell deposits presented in Figure 4-3. It is important to note that density is a relative measure of the frequency of shell within a particular stratum. As such, it doe s not equate with shell depth below surface. Using these data, several large-scale patterns are evident. Shell is principally restricted to the northern half of the survey tract, a nd is typically found within 140 m of the terrace/wetland interface. The trail system asso ciated with the bait fields serves as an approximate boundary between shell-free and shell-bearing deposits. Moreover, dense shell tends to be found closes t to the water. As shown in Figure 4-4, the distribution of vertebrate fauna is principally correlated w ith the appearance, if not density, of shell within an STP. This is an expected result, as shell tends to neutralize Floridas naturally acidic soils that would normally destroy animal bone. Given the complex history of the landscape, it is no surprise that there is considerable variation in the stratigraphic profiles encounter ed during the survey. A few examples serve to show the range between de posits. The compositi on of shell deposits

PAGE 118

Reconnaissance Survey 107

PAGE 119

108 St. Johns Archaeological Field School 2007

PAGE 120

Reconnaissance Survey 109

PAGE 121

110 St. Johns Archaeological Field School 2007

PAGE 122

Reconnaissance Survey 111

PAGE 123

112 St. Johns Archaeological Field School 2007

PAGE 124

Reconnaissance Survey 113 Figure 4-3. Distribution of shell identified during shovel testing, 8LA1-West. Figure 4-4. Distribution of vertebrate fauna rec overed during shovel testing, classified by weight, 8LA1-West.

PAGE 125

114 St. Johns Archaeological Field School 2007 differs by density and species composition, but tends to include mystery snail, bivalve, and apple snail. A typical STP with dense shell, such as 28-6, yielded the following profile: 08 cm below surface (cmbs) dark brown sand with low density whole and crushed shell; 18 cmbs, high de nsity whole and crushed shell with some brown sand; 32 cmbs, brown fine sand with low density sh ell. Excavation at high elevations to the south, excluding shell-bearing deposits, en countered profiles similar to STP 24-11: 0 8 cmbs, dark gray/brown sand; 0 cmbs, very light gray sand; 85 cmbs, dark yellow/brown sand. Slightly different profiles were encountered in the large depression, no doubt due to the different pedogenic proc esses at work during soil formation, as indicated by the profile w ithin STP 9-4: 0 cm cmbs white sand; 65 cmbs light brown sand; 75 dark gray/brown sand. At a smaller scale, there are deviations from the larger patt ern of shell and nonshell distributions. Concreted shell was noted only in nine STPs. These tend to cluster in several locations. Four STPs encountered co ncreted shell at low elevations near the terrace/wetland interface. Concreted shell in this location is expected, given that it is thought to form through percolating water. Similarly, two STPs in Locus A intercepted concreted shell, which is frequently enc ountered in basal depos its of shell mounds (Wheeler et al. 2000). Concreted shell was noted in two STPs within Locus B, both on the northern edge of the dome. Surface cuts there indicate that some shell was removed in this locality, but the presence of concre tion may have made extraction too difficult for extensive operations. Finally, one STP with in Locus C (19-3) intercepted concreted shell between 85 and 100 cmbs. This deposit is unusual in the area, and may represent the base of a pit. In this same vein, severa l STPs with shell were found farther south than expected given the broader la ndscape patterns. For example, STP 20-4 is situated south of the elevated portion of Locus B. The STP intercepted deposits with the following profile: 0 cmbs, light gray brown sand; 17 cmbs, dark yellow brown sand; 80 cmbs, dark gray/brown sand with dense mystery snail, apple snail, and bivalve; 99+ cmbs, brown fine sand. Field notes suggest that the basal shell depos it had the appearance of a feature. Although no pottery was recovere d, this STP may have intercepted another example of a deep Orange period pit that ha s been documented in Locus B (Chapter 6) and across the run at 8MR123 (Randall et al. 2011). Other southerly occurrences of shell may have resulted from road construction a nd maintenance, particularly STPs 5-11, 7-1, and 11-2 in the east and 29-4 and 33-4 in the west. Within the principal zone of shell, there are several shell-free voids. To the west of Locus A are three shell-free STPs (122, 13-3, 14-3). Whether the lack of shell reflects ancient depositional pract ices or recent shell mining is hard to discern from the STP results alone. Some evidence for di sturbance in STP 12-2 was noted, including mottled sediment down to 45 cmbs, a coin (penny), and a metal pipe fragment. However, no such disturbances were noted in the other cases: STP 13-3 yielded a homogenous profile of light brown sand (0 cmbs), a nd STP 14-3 yielded a pr ofile consisting of dark brown/gray sand 0 cmbs) and yellow/brown sand (60 cmbs). Based on topography alone, these STPs would have be en on the backside of the mined shell mound. Elsewhere along the St. Johns River, off-mound testing frequently finds that shell deposits are circumscribed, and so the pattern within Locus A may reflect ancient

PAGE 126

Reconnaissance Survey 115 practices. Another shell-free void, measuring approximately 20 x 40 m in plan, is located on the apex of the ridge nose segment of Locus C (STP 27-2, 27-3, 28-1, 28-2). Although this area was once cleared of trees, presumably as part of the mining operations, there is no other su rficial evidence that would suggest the area was mined. As such, this feature may represent a purposef ully maintained shell-free zone. Finally, shell is found along most, but interestingly, not all of the terra ce/wetland interface. Indeed, shell was rarely encountered west of the Locus C ridge nose, particularly in Transects 31 through 34. Artifact Distribution The distribution of artifacts across th e landform provides further evidence for differences in land-use practices across the survey area. In general, l ithic flakes and tools were the most widely distri buted artifact class (Figure 4-5). Flakes were found in quantities ranging between 1 and 32 per STP, while no more than one lithic tool was found in any STP. The presence of lithic obj ects is independent of shell deposits, and lithics are just as likely to occur in near-water deposits as in the uplands. Indeed, a lithic flake and a biface fragment were found ~290 m south of the spring run, and several lithics flakes were even recovered from the large depression. There are no apparent clusters of lithics across the la ndform. STPs that yielded larg e numbers of flakes (greater than 13) were widely distribute d, and there is no clear gradati on of flake or tool density. The patterning on this side of the run is significantly different than that recently documented north of the Silver Glen Springs main vent (Randall et al. 2011). Lithics were preferentially clustered away from the shell deposits and were found in much higher densities. Moreover, the lithic assemblage was characterized by a wide array of chipped stone tools, including numerous stage bifaces, hafted bifaces, microlit hs, modified flakes, and sandstone abraders. In contrast to the lithic assemblage, the pottery assemblage shows spatial patterning that likely reflects changing land-use practices thr ough time (Figure 4-6). At the largest scale, potte ry of the Orange and St. Johns series is clustered in the western aspect of the survey tract, and was infrequently recovered in the vicin ity of Locus A. As detailed in Chapter 5, stratigraphic excavati ons within the remnant Locus A shell mound revealed intact preceramic deposits up to 3-m thick. Thus, the lack of pottery in this locus is not necessarily unexpected. However, it is notable that St. Johns pottery is found on the northern, swamp-facing edge of the remnant shell mound, albeit in small quantities. The presence of sherds on the southern edge of the escarpment could be dismissed as resulting from mining and the subsequent movement of material. That could be the case on the north side as well, but it is much less likely. The implication is that pottery was emplaced upon portions of the mound after the Mount Taylor (preceramic) period, but those deposits were subsequently removed during the mining process. A closer inspection of the distribution of pottery by type elucidates further chronological and spatial trends. Sherds of the Orange seri es were recovered in only 29 STPs, in frequencies ranging between 1 and 12 sherds per STP (Figure 4-7). Of this

PAGE 127

116 St. Johns Archaeological Field School 2007 Figure 4-5. Distribution of lithic tools and wast e flakes recovered during the shovel test survey. Figure 4-6. Distribution of pottery recovered during the shovel test survey, classified by series.

PAGE 128

Reconnaissance Survey 117 Figure 4-7. Distribution of Orange Plain and Incised sherds recovered during the shovel test survey. total, three STPs yielded Ora nge Incised sherds (n = 17), while the remaining 26 yielded only Orange Plain sherds (n = 65). The dist ribution of Orange Incised and Plain sherds overlaps. Excluding two plai n fiber-tempered sherds in STP 7-7, Orange sherds are restricted to the west in th e survey tract. Al though typically found in small numbers, plain and incised sherds were relatively abundant in STP 22-1, while the nearby STP 236 also contained a large numb er of plain sherds. Aside from the tendency for Orange sherds to be found away from Locus A, this a ssemblage tends to be restricted to either terrace-edge deposits or the upland dome in Locus B. This pattern is most evident in the vicinity of the Locus C ridge nose, where Orange pottery is not found above 6 m in absolute elevation. This is not to say that fiber-tempered pottery is restricted to low elevations, as the highest density (by STP) of sherds occurs in the vicinity of Locus B. However, the presence of po ttery at low elevations around, but not on top of, Locus C would suggest that pottery was preferentially being deposited downslope. The dominant pottery type recovered during the survey was St. Johns Plain (n = 530) which was encountered in 94 STPs widely distributed across the survey tract (Figure 4-8). St. Johns Plain was not only found in a large number of STPs, but the density of sherds per STP was quite high in some cases (a maximum of 50 sh erds was recovered from STP 31-1). Like the Orange series, St. Johns Plain sherds were mostly clustered in the west, although several STPs around Locus A did produce sherds as noted above. More interestingly, however, is that high-de nsity STPs were clustered along the terrace

PAGE 129

118 St. Johns Archaeological Field School 2007 Figure 4-8. Distribution of St. Johns Plain sherds recovered during the shovel test survey. edge. It would seem that during the St. Johns period there was a preference for depositing sherds downslope, although small numbers of St. Johns Plain sherds were found at high elevations within Locus C. One last notable trend is the presence of moderately dense St. Johns Plain assemblages within STP 23-10 and 26-11 near the sand mound. As the only pottery found near the mound, th ese lend credence to the h ypothesis that the mound was constructed during the St. Johns period, perhap s during St. Johns I times. In contrast to the widespread distribution of St. Johns Pl ain, only 71 St. Johns Check Stamped sherds were found in a total of 44 STPs (Figure 49). Excluding one STP in Locus A (13-4) and Locus B (19-3), St. Johns Check Stamped sherds were restricted to the terrace edge and Locus C. Although the check stamped sherds we re recovered at elevations above 6 m, on top of the ridge, the vast majo rity were actually recovered farther to the west at the survey tract border. This is an interesting trend, as little to no shell was encountered in this location. CONCLUSIONS Systematic shovel testing along Juniper Cl ub property fronting the south side of Silver Glen Run shows that subsurface arch aeological deposits are distributed widely across the 11.6-ha survey trac t. Roughly 80 percent of the 238 shovel test pits (STPs) excavated in the tract yielded pre-Columb ian artifacts and/or anthropogenic shell deposits. The latter was observed in 133 STPs the vast majority within 140 m of the spring run, but also sporadically at distan ces over 200 m from th e run. Shell density

PAGE 130

Reconnaissance Survey 119 Figure 4-9. Distribution of St. Johns Check St amped sherds recovered during the shovel test survey. varied markedly in STPs across the survey tr act. Dense subsurface shell coincides with the footprint of the mined Mount Taylor sh ell ridge designated Locus A, but it also occurs across the terrace slopes of Loci B and C and in their respective shell domes to the south, forming the apex of adjacent ridge noses. Vertebrate fauna coincide with shell due largely to the enhanced pres ervation of organic matter a fforded by the acid-neutralizing affects of degraded shell. Several areas de void of shell are notewor thy. West of Locus A, the Mount Taylor shell ridge, STPs lacking shell may signal the actual termination of this oldest shell deposit, but we hasten to add that shell-mining operations in this locus may have created an artificial void. More m eaningful perhaps is the small shell void at the apex of Locus C. Ongoing work in this location is providing evidence for a St. Johns II-period village with a presumptive central plaza Shell was largely absent as well to the west of Locus C, on the western margin of th e club property. We have not yet to conduct secondary testing in this location, but based on the density of check-stamped St. Johns pottery (see below), a nonshell component coeval with the Lo cus C village appears to be present. Like shell, subsurface pre-Columbian arti facts are distributed widely across the survey tract, and reveal spatial patterning indicative of distinct archaeological components. This is most evident in the di stribution of pottery. Sherds are generally absent in Locus A, the location of a precer amic shell ridge. Occas ional St. Johns period

PAGE 131

120 St. Johns Archaeological Field School 2007 sherds in this locus may signal a reuse of the shell ridge after an extended period of abandonment, but this remains speculative because the upper portion of the ridge was compromised by shell mining. The oldest po ttery, that of the Orange series, is concentrated in Loci B and C, largely in th e shell nodes of each locus, but also in the downslope portion of Locus C. St. Johns pottery is likewise dist ributed across Loci B and C, with especially dense occurrences in the downslope aspects of both loci. Checkstamped St. Johns pottery is con centrated in Locus C and especially in the shell-free ridge nose to the west of Locus C, most notably in STPs of transects 31-34. In sum, reconnaissance survey of 8LA1-West shows that the entire landform fronting the spring run contains intact subsurf ace deposits with few lacunae. Variation in the composition and density of subsurface shel l and artifacts enables us to subdivide 8LA1-West into three loci (Loci A, B, and C) and to implement for each a program of secondary testing and data recove ry. Provided in the balance of this report are the results of testing at Loci A and B; testing at Locus C began in ear nest only this past summer (2011) and will continue in 2012 and possibly beyond. The results of this work will be provided in a later report.

PAGE 132

CHAPTER 5 SILVER GLEN RUN, LOCUS A (8LA1-WEST) Kenneth E. Sassaman and Asa R. Randall Locus A at Silver Glen Run is the remnan t of a ~200-m-long shel l ridge dating to the middle part of the Mount Taylor period (ca. 6300-5750 cal BP). Much of the deposit was mined for shell in the 1923, when 8LA1-East was also mined, but substantial portions of the margins of this ridge remain intact. As shown in Figure 5-1, the outline of the ridge is marked today by a series of discontinuous escarpments and isolated remnants of mounded deposits. The core of this depos it has been largely denuded through mining, although scattered shell on the su rface attests to intact subsur face strata (Figure 5-2 top). Shell continued to be removed from this porti on of the site until recently, usually in small loads taken with a bucket fitted to a tractor. The margin of th is ridge fronting Silver Glen Run expresses the greatest level of continu ity, with linear escarpments as much as 2 m tall (Figure 5-2 bottom). It is difficult to j udge the original height of the ridge, but it no doubt rose higher than the tops of the extant escarpments. Irrespec tive of actual height, testing at Locus A revealed substantial subsur face deposits below the grade of the mining pit. We estimate that Locus A originally had at least three to four meters of stratified deposits. Our goal in testing portions of the ridge in 2007 and 2008 was to expose and sample as much stratigraphy as possible from locations where mining operations resulted in the steepest escarpments. Three locations were examined in this fashion (Figure 5-1), totaling 24 m2 of plan excavation and 12 meters of profile. This chapter reports the results of these efforts. TEST UNIT EXCAVATION Six 2 x 2-m test units were excavated in the mining escarpments of Locus A in three different locations: two at the east end (Test Units 5 and 8), one near the west end (Test Unit 6), and three arranged as a trench in an intermedia te location (Test Units 9, 10, and 15) (Figure 5-1). In all cas es test units were oriented s quare with the orientation of the escarpment and placed far enough into th e escarpment to afford a clean, vertical cut of above-ground deposits. Test units were al so excavated below the grade of the mining pit to examine subsurface deposits that were spared the damage of mining. The initial exposure of each escarpment was accomplished by the removal of a wedge-shaped level (referred to as Profile Cut in the text and tables that follow below), which was passed through -inch hardware cloth. All artifacts and vertebrate fauna were retrieved, bagged, and returned to the lab for analysis; freshw ater shell was not co llected. Once intact deposits were encountered, excavation procee ded in 10-cm arbitrary levels, usually within only a 1 x 2-m subunit to the shallow side of the escarpment, to prevent wall collapse (Figure 5-3). Throughout excavation, z ones of distinct deposition were removed and processed as subsamples of levels. Pi t features and other di screte deposits were mapped, sectioned, and sampled individually. At the close of excavation, all profiles were photographed and drawn to scale. Becau se the goal of testing at Locus A was to examine stratigraphy, profiles were give n more than the usual attention. 121

PAGE 133

122 St. Johns Archaeological Field School 2007-2010: Silver Glen Run Figure 5-1. Map of 8LA1-West, Locus A, showing locations of test units excavated in escarpments of mining operations.

PAGE 134

Silver Glen Run, Locus A (8LA1-West) 123 Figure 5-2. View facing west of the mining pit in the core of Locus A (top), and view facing northeast of the mining escarpment fronting the spri ng run, into which Test Units 5 and 8 were dug (bottom). Test Units 5 and 8 Placed at the east end of the shell ridge, Test Units (TU) 5 and 8 were excavated in sequence over two summer sessions (2007 a nd 2008, respectively) to provide a threedimensional view of ridge stratigraphy. The un its shared a corner but were offset to create two 4-m-long profiles set orthogonally. Test Unit 5 was excavated completely to basal sands, exposing a 3+ m-deep sequence of well stratified, anthr opogenic deposition. Test Unit 8 was suspended before reaching sterile substrate due to encounters with human remains, but ~1.6 m of the upper portion of the profile was exposed. The details of this latter unit are prov ided following a description of stratigraphy in TU5.

PAGE 135

124 St. Johns Archaeological Field School 2007-2010: Silver Glen Run Test Unit 5 was placed on an east-facing vertical escarpment. The unit was set back into the escarpment approximately 30 cm to ensure that intact stratigraphy would be exposed throughout. Initially, the unit was ex cavated to a depth of ~170 cm below the surface (cmbs), which approximates the eleva tion of the surrounding terrain to the south of the unit. The unit was then subsecti oned into two 1 x 2-m units, the eastern (downslope) half of which (TU5-E) was continue d to a depth of ~307 cmbs (Figure 5-3). After excavation we recorded the stratigraphy. Photographs and drawings of the west, south, and north profiles are presented in Fi gures 5-4, 5-5, and 5-6 respectively. Strata color and matrix composition descriptions are presented in Table 5-1. An inventory of objects recovered during excavation is pr esented in Table 5-2. Several strategies were employed during excavation. Th e upper 170 cm was removed as a single Profile Cut, and all materials were kept together as one provenience. No attempt was made to discriminate between intact and disturbed deposits. Profile cutting operations were ceased at approximately 170 cmbs when a stratigraphic change, characterized by an incr ease in sand and charcoal, was encountered. At this point, we excavated TU5-E in arbitrar y 10-cm levels (beginning with level A). We also recognized several distinct zones that were excavated and bagged separated. Surficial and potentially disturbe d fill was designated Zone A. It was recognized during excavation as relatively homogeneous and composed of organically enriched sand, Figure 5-3. Example of stepped excavation in Test Unit 5 to prevent wall collapse and allow for mapping of upper strata.

PAGE 136

Silver Glen Run, Locus A (8LA1-West) 125 Figure 5-4. Photograph and line drawing of west profile of Test Unit 5, 8LA1-West. fragmented shell, and living Juniper tree roots. In prof ile, Zone A corresponds with Strata (hereafter Str.) I a nd Ia. Beginning with Level A, we recognized several other zones. Zone B was described as a gray/bro wn sand with crushed shell. Zone C was described as whole and crushed shell. Zone D was described as tan sand. After Level A we simplified the zonation, such that Zone A was the disturbed matrix and Zone C was all shell-bearing matrices. Beginning with Level H we also recognized Zone E, which was characterized as a gray/brown sand with varying shell density. All sediment in Level H through Level M was associated with this zone designation. Excavations were ceased at the bottom of Level M. Although occasional shell and bone fragments were still encountered, the soil had become both very light in color and wet, and vertebrate bone frequency had significantly decreased. Excavation of TU5 produced a wide array of cultural materials, including bifaces, lithic flakes, marine shell, rock fragments, modified bone, over 3 kg of vertebrate faunal

PAGE 137

126 St. Johns Archaeological Field School 2007-2010: Silver Glen Run Figure 5-5. Photograph and line drawing of south profile of Test Unit 5 (East), 8LA1-West. bone, a few sherds, and some historic and mo dern materials such as nails and plastic (Table 5-2). Vertebrate fauna was found in the greatest density in the profile cut, based on relative volume. We also recovered bone concretions in appreciable quantities. In many cases, these concretions were simply seve ral bones with ash or other sediment that tightly bonding them together. In at least one case (Level G), bone concretions may have represented whole fish that were cemented in place. Finally, excluding the recent objects and sherds of the Orange and St. Johns series, this invent ory fits comfortably in the known range of Mount Taylor period assembla ges (see examples in photographs in the closing section of this chapter). The ma jority of non-Mount Taylor objects were recovered from surficial or scree-slope deposits excavated either in the Profile Cut or in Zone A. The one exception is Level G, whic h yielded two Orange Pl ain sherds. Level G was the first level in which Zone A was not visu ally recognized or excav ated. It is likely, however, that these sherds were recovered from disturbed contexts. As seen in the north

PAGE 138

Silver Glen Run, Locus A (8LA1-West) 127 Figure 5-6. Photograph and line drawing of north profile of Test Unit 5 (East), 8LA1-West. and south profiles, Str. I and Ia have an irregular and dipping contact with intact stratigraphy, and ther e may have been some deposits of Str. I remaining in Level H which was not easily visible in plan view. A total of 28 strata were discernible in profile (Table 5-1). For the purposes of discussion, these can be grouped into four macrostratigraphic units that share similar structure. From top to bottom these include : surface deposits, upper shell and sand, lower shell, and basal shell and sand. Overlying a ll intact strata in TU5 was Stratum I, an organically enriched very dark grayish brow n loamy fine sand with abundant roots, in addition to fragments of shell in varying quantities. Related to this matrix is Str. Ia, which was identified in the TU5-E profiles. The extent to which Str. I represents the original mound surface or is a product of mining operations th at homogenized and organically enriched sediments is unclear. Based on stratigraphic data and assemblage content, two post-depositional pr ocesses are likely represented here. In the west profile, Stratum I varies in thickness between 30 a nd 50 cm, and its lower margin generally mimics the surface dip. Moreover, the contact be tween Str. I and Str. II is not sharp, and there is no clear evidence that Str. II or III we re truncated. In this location, Str. I may thus represent surficial deposits that were expos ed for millennia and subject to bioturbation.

PAGE 139

128 St. Johns Archaeological Field School 2007-2010: Silver Glen Run Table 5-1. Stratigraphic Units of Test Unit 5, 8LA1-West, Locus A. Max. Depth Munsell Stratum (cm BS) Color Description I 68 10YR3/2 very dark grayish br own loamy fine sand with abundant roots, occasional crushed Viviparus Pomacea and Unionid shell Ia -10YR3/2 redeposited Stra tum I with additional shell II 41 10YR4/3 brown ashy fine sand, partially concreted, no shell III 81 10YR5/4 abundant whole and crushed Pomacea and Unionid with occasional Viviparus shell in yellowish brown fine sand IV 60 10YR2/2 very dark brown fine sand with Viviparus and occasional Pomacea shell, charcoal; partially concreted V 85 10YR3/1 very dark gray fi ne sand with occasional crushed shell and charcoal VI 74 10YR5/3 brown fine ashy sand with abundant crushed Viviparus shell and charcoal (5320 30 BP) VII 99 10YR3/2-3 very dark grayis h brown to dark brown fine ashy sand with abundant whole Viviparus shell and charcoal VIII 109 10YR3/2 very dark grayis h brown fine ashy sand with abundant whole Viviparus shell and charcoal IX 107 10YR6/4 light yellowish br own fine sand with 10YR4/3 brown mottles, small flecks of charcoal and crushed shell X 146 10YR4/4 dark yellowish brown fine ashy sand with whole and crushed Viviparus shell and occasional charcoal XI 184 10YR5/2 grayish brown fine sand with Viviparus and occasional Pomacea and Unionid shell and flecks of charcoal XII 155 10YR8/3 abundant whole and crushed Pomacea and Unionid with occasional Viviparus shell in very pale brown fine ashy sand XIII 168 10YR5/4 yellowish brown fine sand with occasional flecks of charcoal and crushed Viviparus shell XIV 164 10YR2/1 charcoal in fine sand with occasional crushed shell XV 177 10YR7/6 yellow fine sand with 10YR4/3 brown mottles and rare flecks of charcoal

PAGE 140

Silver Glen Run, Locus A (8LA1-West) 129 Table 5-1. Continued. XVa 166 10YR6/8 brownish yellow sand with 10YR6/1 gray mottles and abundant crushed shell, flecks of charcoal, and ash XVI 199 10YR7/6 yellow fine sand with 10YR4/3 brown mottles and rare flecks of charcoal XVIII 204 10YR6/4 whole Pomacea and Unionid in minimal light yellowish brown sand with occasional charcoal XIX 218 10YR4/2 dark grayish brown fine ashy sand with whole Viviparus and occasional crushed Pomacea and Unionid shell and flecks of charcoal XX 226 10YR4/1 dark gray fine ashy sand with whole Viviparus and occasional crushed Pomacea and Unionid shell and flecks of charcoal XXa 237 10YR5/4 abundant whole Pomacea and Viviparus shell in yellowish brown fine sand XXI 246 10YR3/2 abundant crushed she ll in very dark grayish brown fine ashy sand with charcoal throughout XXII 250 10YR4/3 crushed Unionid, Pomacea and Viviparus shell in brown fine sand with flecks of charcoal throughout (5290 40 BP) XXIII 265 10YR3/2 very dark gray ish brown fine sand with occasional crushed shell and charcoal XXIIIa 302 10YR3/2 very dark gray ish brown fine sand with occasional crushed shell and charcoal XXIV 278 10YR3/1 whole and crushed Viviparus and occasional Pomacea and Unionid shell in very dark gray fine ashy sand XXV 307+ 10YR5/4 yellowish brown fine sand The relationship between Str. I/Ia and in tact stratigraphy is different for the TU5E profiles. In both the north and south walls, Str. I and Ia lie unc onformably, at roughly a 30-degree angle, upon relativ ely flat-lying strata that have been truncated. Also significant is that recent Juniper roots were dist ributed throughout Str. I at this elevation, indicating that it was less compact and easier to grow through (see Fi gure 5-6). Because we did not excavate the western 1 x 2-m unit, the relati onship between the upper and lower Str. I/Ia remains indeterminate. Fi nally, it is notable that almost all pottery recovered from TU5-E was associated with St r. I deposits (Zone A) further indicating

PAGE 141

130 St. Johns Archaeological Field School 2007-2010: Silver Glen Run

PAGE 142

Silver Glen Run, Locus A (8LA1-West) 131 that this matrix was redeposited from elsewh ere. No sherds were recovered during the profile cut excavations. Strata II through XVIII comprise the upper shell unit, a stratigraphic assemblage internally marked by variable thickness and composition. Fo r example, strata are as diverse as the predominantly whole and unbroken Pomacea and Viviparus in Str. XVIII to a mostly shell-free yellow sand in Str. XV, and range from ~5 cm thick to more than 50 cm. Although a complex arrangement, thes e strata share seve ral structural and sequential characteristics that suggest a repeated pattern. In particular, it appears that a certain arrangement of surface features was bein g maintained. Most of the strata in this sequence have a distinctive topographic expression: they dip from north to south (Figure 5-4). Whether this dip has a corresponding st rike to the east is unknown due to mining, but there is similar patterning evident in TU 8 to the south and west (see below). Within the upper shell unit are two repe ated depositional subunits: thin burned layers and extensive shell and sand layers. The base of the sequence is composed of sand-only layers in association with whole shell. These different depositional subunits are discussed below. One component of this sequence includes thin strata (ca. 3 cm overall), that show evidence for extensive bur ning, and have variable amounts of shell. Str. II, for example, is characterized by a brow n ashy fine sand with no shell. In contrast, Str. XIV is composed almost exclusively of charcoal, with limited sand or shell. Most others, however, have crushed sh ell of variable quantity, incl uding Str. IV, V, VI, and IX. A sample of Str. VI was collected from the pr ofile. In the laborat ory a sample of wood charcoal was removed and submitted for an AMS assay to Beta Analytic. It returned a conventional age estimate of 5320 30 BP. These lenses are frequently concreted, providing further evidence of both burning and th e presence of ash. In general, these lenses mimic the contour of the underlying st ratum. In the eastern profile, they are typically discontinuous across these surfaces, but are most strongly expressed in the southern half of the west prof ile. In this sense, these lenses appear to be strongly associated with areas of highest topography, as if they were emplaced directly upon a mounded surface. A second subunit to the upper shell sequence consists of relatively thick (ca. 10 50 cm) and laterally extensiv e layers of shell with variable amounts of non-shell inclusions. The vast majority of thes e are characterized by whole and crushed Viviparus Pomacea and bivalve shell in a yellow-brown to brown fine sand. Often the non-shell matrix is ashy, and contains flecks of charcoal Layers matching this description include Str. III, VII, VIII, X, and XI. Some laye rs, however, are composed primarily of whole shell with limited non-shell matrix. Str. XII, for example, was expressed as several small pockets of whole and crushed Pomacea and Unionid shell with the occasional Viviparus shell. A more thick and extensive whole shell lens is Str. XVIIII, which is composed almost exclusively of whole Pomacea and Unionid shell. Crushed shell was present in this stratum, but was concentrated directly beneath Str. XIV and XV. The whole shell in Str. XVIII is closely associated with the third type of deposit that is composed almost exclusively of yello w-brown sand. In some cases, such as Str.

PAGE 143

132 St. Johns Archaeological Field School 2007-2010: Silver Glen Run XIII, the sand is discontinuous. Most striking, however, is the 15-cm thick Str. XV that is composed of yellow fine sand with rare char coal. It is a lenticular-shaped stratum that lies above whole shell in Str. XVIII and below the burned ch arcoal of Str. XIV. The presence of ashy sand (Str. XVa) at the cont act between Str. XIV and XV suggests that a fire was constructed directly upon the Str. XV surface. A related sand layer is Str. XVI, a ca. 15-cm thick yellow fine sand with brown mo ttles and rare charcoal. The lateral extent of Str. XVI is unknown because it was truncated by mining, but it is visible in the north profile. Taking these strata t ogether, it would seem that a la rge layer of w hole shell (Str. XVIII) was deposited on a mostly flat surf ace (Str. XIX). A fire may have been constructed on top of Str. XIX prior to the deposition of shellfish, as there was ash at the contact. Subsequent to the shell deposition, loads of yello w sand were deposited on top of this shell. This layer was then covered with hot coals. This sequence of (1) shell deposition, (2) sand deposition, and (3) burni ng may provide a model for subsequent deposition. That is, the overlying thick shell/ sand layers appear to be linked with thin lenses of charcoal and crushed shell. Fo llowing the model considered above, it would seem that loads of sand and shell were emplaced as small heaps, and then the surface of each heap was burned. Whether the burning oc curred just after the surface was created, or was conducted in advance of a subse quent deposit remains unknown at this time. Beneath Str. XVIII, the character of strata shifts drastically from elevated and discrete deposits to laterally extensive, fl at-lying, and heterogeneous strata between 10 and 20-cm thick. The composition and disposi tion of these strata are suggestive of a depositional pattern that did no t include the differentiation of shellfish species, but did involve significant mechanical crushing of she llfish remains. For example, Str. XIX is composed of ashy sand with whole Viviparus, and crushed Pomacea and Unionid shell. Str. XX is similarly composed of whole Viviparus with minor shell, and it grades into Str. XXa, a layer of whole Pomacea and Viviparus shell in a yellowish brown fine sand. These units are underlain by the re latively thin (ca. 8-cm thic k) Str. XXI and XXII, which are characterized by abundant cr ushed shell and charcoal. In some instances, crushed Unionid shell was present in patches. A sample of charcoal was collected from a crushed shell lens in Str. XXII, and submitted for an AMS assay. It retu rned a conventional age estimate of 5290 40 BP. Alt hough the age intercept for this date is later than the overlying Str. VI, both estimates overlap at 2sigma, and are thus statistically coeval. What this suggests is that the deposits in this portion of Locus A were rapidly emplaced. The final macrounit revealed in TU5 excavations provides evidence for intermittent occupation of this landform prior to the extensive deposition described above. During excavation of Level H (ca. 240 cmbs) the crushed shell of Str. XXII gave way to Str. XXIII, a ca. 10-cm thick a da rk grayish brown, organically enriched fine sand with occasional crushed shell and charcoal. This sediment was found to overlie another extensive deposit (Str. XXIV) of whole and crushed Viviparus Pomacea and Unionid shell. This shell lens was consis tently observed in the west, south, and east profiles, but was discontinuous in the northern profile. Bene ath the shell we encountered another organic soil with limited shell (Str. XXIIIa). This graded into a yellowish brown fine sand towards the base of the unit. Alt hough it is possible that the shell-free organic sediment was anthropogenic in origin, and purposefully emplaced like Str. XV and XVII,

PAGE 144

Silver Glen Run, Locus A (8LA1-West) 133 there are several lines of evid ence that indicate both repres ent A-horizons. First, Str. XXIII and XXIIIa were indistinguishable in the field. They were both organically enriched and characterized by a notable lack of, or comparable redu ction in vertebrate fauna. Secondly, a Euglandina shell was recovered from shell in Level H. This terrestrial gastropod preys upon other terrestrial univalves We have argued elsewhere (Sassaman et al. 2005) that Euglandina are most likely to be found in association with stable, exposed surfaces that foster colonization of prey species. There is a drop off in the density of vertebrate fauna beginning with Level H, although objects continue to be found in appreciable qu antities. We recovered two Fl orida Archaic Stemmed hafted bifaces, 2 lithic flakes, and 3 marine shell fr agments, and four modified bone pieces in this macro-unit. TU8 was oriented to the west of TU5 (Figure 5-7). As before, one goal of excavation was to expose strati graphy in this portion of th e shell ridge. Drawings and photographs of the north and east profiles are presented in Figure 5-8, and a description of each stratum is presented in Table 5-3. An inventory of objects recovered during excavation is presented in Table 5-4. Because of the surface topography in this location, excavation of TU8 followed slightly diffe rent protocols. Unlike TU5, which was emplaced on a sharp escarpment, the surface beneath the TU8 footprint sloped gradually. As a result, we did not remove a profile cut, but instead excava ted in arbitrary levels from the start. Levels A and B we re removed as 20-cm levels. Thereafter, all levels were removed in 10-cm increments. Each level wa s a wedge cut, which increased in volume Figure 5-7. View facing northwest of the excavation of Test Unit 8, 8LA1-West.

PAGE 145

134 St. Johns Archaeological Field School 2007-2010: Silver Glen Run Figure 5-8. Photographs and line drawings of the north and east profiles of Test Unit 8, 8LA1West. as excavation proceeded. Beginning with Le vel F (70 cmbs) we recognized two zones that were excavated, screened, and bagge d separately. Zone A corresponds to the organically enriched mining spoil recognized in TU5. It was removed first as each level was removed. All intact stratigraphy not incl uded in Zone A was ex cavated as Zone B. We ceased excavation in TU8 upon the discovery of two subadult burials (see below). The material culture assemblage recovere d from TU8 is complementary to that recovered in TU5. As enumerated in the ar tifact inventory from TU8 (Table 5-4), all post-Mount Taylor pottery sherds of the Orange and St. Johns series, as well as historic artifacts, were recovered from Zone A. Vertebrate faunal bone was the most frequently recovered object (n = 8898, 3437.1 g). This de nsity is consistent with the profile cut from TU5 which yielded a similarly large quantity of faunal bon e. Otherwise, the

PAGE 146

Silver Glen Run, Locus A (8LA1-West) 135 Table 5-3. Stratigraphic Units of Test Unit 8, 8LA1-West, Locus A. Max. Depth Munsell Stratum (cm BS) Color Description I 53+ 10YR3/2 very dark grayish br own ashy sand with moderate whole and crushed Viviparus and occasional Unionid shell II 32 10YR4/2 dark grayish brow n ashy sand with occasional crushed shell III 36 10YR6/6 abundant crushed Unionid with occasional whole and crushed Viviparus shell in brownish yellow ashy fine sand IV 44 10YR3/1 very dark gray fine sand with occasional whole and crushed Viviparus and Unionid shell with flecks of charcoal throughout V 63 10YR4/2 abundant whole and crushed Viviparus and occasional Pomacea and Unionid shell in dark grayish brown ashy sand with occasional flecks of charcoal VI 64 10YR7/6 concre ted crushed and burned Viviparus and occasional Unionid shell in yellow ashy, gritty sand VII 70 10YR5/4 yellowish brown ash y, gritty sand with lenses of whole and crushed Viviparus and Unionid shell VIII 75 10YR3/2 very dark grayis h brown fine ashy sand with abundant whole Viviparus shell and charcoal IX 84 10YR4/2 light yellowish br own fine sand with 10YR4/3 brown mottles, small flecks of charcoal and crushed shell X 89 10YR3/1 dark yellowish brown fine ashy sand with whole and crushed Viviparus shell and occasional charcoal XI 58 10YR4/2 grayish brown fine sand with Viviparus and occasional Pomacea and Unionid shell and flecks of charcoal XIa 35 10YR5/4 yellowish brown gritty sand with occasional crushed shell intercalated with 10YR3/2 very dark grayish brown sand with occasional crushed shell XII 112 10YR5/2 grayish brown medi um ashy sand with moderate whole and crushed Viviparus occasional whole and crushed Unioinid and Pomacea shell, and flecks of charcoal XIII 115 10YR5/1 gray ashy, gritty sand with abundant whole and crushed Viviparus shell

PAGE 147

136 St. Johns Archaeological Field School 2007-2010: Silver Glen Run Table 5-3. Continued. XIV 64 10YR4/1 dark gray sand with moderate crushed and whole Pomacea and Viviparus shell and occasional charcoal XV 115 10YR5/3 mostly whole Viviparus and occasional Pomacea and Unionid shell in a trace of brown sand XVI 136 10YR4/3 brown ashy sand with abundant whole and crushed Viviparus and occasional whole and crushed Pomacea and Unionid shell XVII 66 10YR4/1 abundant crushed Uni onid shell and charcoal in trace of dark gray sand XVIII 69 10YR3/2 very dark grayish brown ashy sand with trace of crushed shell and occasional flecks of charcoal (subadult interment?) XIX 140 10YR4/1 abundant crushed Unionid shell and charcoal in trace of dark gray sand (probably co terminus with Stratum XVII) XX 149 10YR6/6 brownish yellow fine sand with occasional whole and crushed Viviparus XXI 150 10YR5/3 crushed shell in brown ashy sand (528040 BP) XXII 161+ 10YR5/2 abundant whole Viviparus and Pomacea shell, with minor crushed shell and abundant charcoal in trace of grayish brown sand XXIII 87+ 10YR5/2 abundant whole Pomacea shell, with minor crushed shell and abundant charcoal in trace of grayish brown sand material culture assemblage included lithic waste flakes, two hafted bifaces, marine shell fragments, and a few modi fied bone fragments. Excavation of TU8 revealed 24 stra tigraphic units that were mostly complementary to the disturbed and intact upper shell macrounits documented in TU5. Three upper and disturbed strata were recognized. St r. I is an organi cally enriched and homogeneous ashy sand with moderate amounts of crushed shell. As with TU5, this stratum likely represents both a pre-mining inta ct surface, particularly above Str. II, in addition to post-mining scree deposits. In bot h the north and west profiles, Str. I lies unconformably above truncated inta ct strata. Sherds of the St. Johns and Orange series were restricted to Zone A, which corresponds to Str. I. Another kind of disturbed deposit was recognized in Str. XI and XIa, visible in the eastern profile. These deposits are composed predominantly of gray to ye llow brown sand with occasional shell and

PAGE 148

Silver Glen Run, Locus A (8LA1-West) 137

PAGE 149

138 St. Johns Archaeological Field School 2007-2010: Silver Glen Run charcoal. They too rest unconformably upon int act strata lying more or less horizontally in the eastern portion of the uni t. The origin of this sa nd mass is unknown. Presumably, the sand was emplaced incidentally during the mining process, and perhaps escaped from the steam shovel bucket. On cl ose inspection, there are disaggregated, but intact, clods of whole strata embedded within Str. XI. Just like TU5, the upper intact shell unit consisted of thin and thick strata of varying composition. In TU8, the relationship between the massive layers and thin strata is made more evident based on surface topogra phy and content. If the thin and thick strata are parts of the same depositional se quence (as argued above), then somewhere between five and six depositional events are represented in the TU8 profiles. Based on excavations at the Hontoon Dead Creek Village (8VO215) and the Hontoon Island North site (8VO202), these elevated surfaces may represent discrete house mounds and floors. Although no architectural features such as post-holes have been identified, similar deposits of shell, roughly 20-50 cm thick and fr equently covered with crushed shell, were identified at these other sites. In TU8, the uppermost depositional sequence is represented in Str. II-V. Although varying in composition, they all share the same slope and angle of repose, which can be seen tre nding upwards to the east and south. The thin strata (II, III) are composed of ashy fine sand with varying amounts of crushed shell. These are likely expressed in TU5 as Str. II The more massive strata (IV and V) are composed of whole and crushed Viviparus, Pomacea and Unionid shell with gray and yellow sand. A secondary sequence is repr esented by Str. VI-XII, wherein there are many thin complicated strata (VI-X) emplaced upon one massive deposit of ashy fine sand with whole and crushed Viviparus, Pomacea and Unionid shell (Str. XII). As seen in the northern profile, the ch arcoal and crushed shell layers trend upwards to the west, and appear to follow the elevated surface contour of Str. XII. Finally, Str. XII has a large root cast that appears to originate at or near the stratums surface, although it has been partially truncated by Str. XIa. This is likely the same surface from which root casts originate in TU5 (Str. X), giving further s upport for a temporary occupational hiatus. Str. XII is situated upon the Str. XIII surface, which is composed of ashy sand with abundant whole and crushed Viviparus shell. This laterally extensive and thin deposit begins a new series of depositional couplets. Str. XVII, XIX and XXI are composed predominantly of crushed bivalve with abundant charcoal. They are present across both the north and east profile. A char coal sample was extracted from Str. XXI and submitted for an AMS assay to Beta An alytic. It returned a conventional age estimate of 5280 40 BP. The intervening st ratigraphic units (Str XVI, XX) are more massive and composed of both sand and whole and crushed shellfish remains. In the profiles that we exposed, the lower stratigrap hy identified in TU8 are mostly flat lying (except for Str. XV and XVIII, see below). Based on the presence of the sand/shell nodes in TU5 at this elevation, it is likely that the strata observed in TU8 represent lateral elements of similar shell domes, and away from the zone of elevated deposition. The base of the test unit just barely expos ed an extensive layer of mostly whole Pomacea and Viviparus shell, with limited non-shell matrix (S tr. XXII, XXIII). This layer of shell corresponds with Str. XVIII in TU5 in terms of depth and composition.

PAGE 150

Silver Glen Run, Locus A (8LA1-West) 139 As noted previously, excavation of TU8 was terminated prior to reaching basal sands due to the presence of two subadult inte rments. The first was not recognized in the field. Back in the laboratory as samples were being clean ed, several bones from Level M (140-150 cmbs) were identified as human subadult skeletal elements. At the time of this discovery, we were midway through excavati ng Level O (160-170 cmbs). As we were cleaning down the base of Level O we encounter ed yet another subadult interment, this time in situ After consultation with the State Arch aeologist, all skeletal elements were reinterred, and we stopped excavations here a ltogether. After reco rding the stratigraphy, the unit was backfilled. The stratigraphic re lationship of the second interment remains unknown because it was encountered at the base of the unit, although presumably it was associated with the whole shell deposits of St r. XXII. Based on field notes, there is little doubt that the first subadult interment was asso ciated with Str. XVIII, described as very dark grayish brown sand with only a trace amoun t of crushed shell and some charcoal. In profile, this stratum has the appearance of a small basin, and is likely a pit into which both the subadult and gray sand were emplaced. Importantly, the burial pit lies directly beneath Str. XV, which is composed of mostly whole shell with limited non-shell matrix. In profile, Str. XV also has a basin shape, and appears to lie unconformably upon Str. XVI, which appears to be late rally truncated by Str. XV. Based on thes e stratigraphic relationships, it would appear that a pit was dug through an existing shell node, and then a subadult was buried in sand at the base of the pit. The burial pit was then filled with whole, and apparently clean sh ell. These burial practices ar e the microcosm of interment procedures afforded adults at coeval cemeter ies such as the Harris Creek mortuary mound on Tick Island (Aten 1999). Interestingly, subadults were greatly underrepresented at Harris Creek. Based on the discoveries in TU8, it may be the case that Mount Taylor communities buried juveniles in residential spaces, perhaps under house floors. In sum, excavation of TU5 and TU 8 exposed a complicated, 3+ m deep stratigraphic sequence composed of sand, shellf ish remains, and other objects. In this portion of the Locus A ridge, at least four macro-stratigraphic units are evident from bottom to top: shell midden above a burie d A horizon, a lens of emplaced sand, stacked sequence of house floors and living surfaces, and a post-abandonment stratum. Radiocarbon assays from thr oughout the sequence suggest th at Locus A emerged rapidly (see below), although there may have been intermittent abandonment. The mined disposition and manner in which we excavated the units limits what we can say about material culture frequency, but a few points are worthy of note. With one exception, all post-Mount Taylor material culture was re stricted to near-surface of mining scree deposits. Where sherds of the Orange and St. Johns series came from initially remains unknown. One possibility is that they were relocated from somewhere else on the landform, such as Locus B or C, during mi ning operations. Alternatively, they could have come from surficial deposits within Locus A. Indeed, many large Mount Taylor shell ridges have pottery in the upper 50 cm (Wyman 1875). Secondly, the material culture was diverse and consistent with the earlier Mount Taylor phase. Tools characteristic of the later Thornhill Lake phase, such as microliths and Strombus celts were not present. Finally, there was compar ably more vertebrate fauna recovered from the upper macrostratigraphic units than in the basal midden.

PAGE 151

140 St. Johns Archaeological Field School 2007-2010: Silver Glen Run Test Unit 6 A single 2 x 2-m test unit was excavated in 2007 toward the west end of the Locus A shell ridge in an escarpment running pa rallel with the long axis of the ridge and the spring run. Test Unit 6 (TU6) generall y recapitulated the stratigraphic sequence observed in TU5 and TU8, but with less complexity in the upper unit and an overall shallower profile. Nonetheless, a sequen ce of basal midden followed by emplaced sand and shallow pit features is fully coeval wi th the lower sequence at the east end of the ridge. The artifact assemblage from TU6 is likewise consis tent with the Mount Taylor assemblage recovered from the east end of the ridge. Similarities in the stratigraphy between the two locations far outweigh the differences, supporting the inference that similar processes of deposition account for the sequences and that they occurred at roughly the same time. It follows that the spatial extent of Mount Taylor activities leading to mounded deposits encompasses upwar ds of 200 m of the spring run terrace. Excavation of TU6 generally followed the same protocols as the TU5. A Profile Cut was made from the top of the escarpment to a depth of 107 cmbs, roughly at the point where the southeast corner of the extant mined surface was encountered (Figure 5-9). This wedge-shaped unit crosscut multiple she ll-bearing strata and brown sand, and scree deposits. Shell was generally looser than in TU5 and with less associated sand matrix. No pottery was recovered in th e Profile Cut, but several lith ic artifacts were found, along with marine shell fragments, some worked bone, and relatively abundant vertebrate fauna, although somewhat less per unit volume than in TU5. The floor plan at the base of the prof ile cut (ca. 107 cmbs) consisted almost entirely of brown medium sand with only minor shell. Evident at this level were several pit features, including possible postholes. Ultimately, seven circular stains or areas of concreted shell were designated features (F eatures 6-12; Figure 5-10), but not all were investigated because only the south half of TU6 was excavated below the level of the Profile Cut. As with TU 5, TU6 was divided into two 1 x 2-m subunits, with the downslope portion (TU6-South) excavated through th e removal of 11 10-cm levels to a depth of 220 cmbs. Figure 5-11 provides a photograph and drawing of the stepped profile. Strata identified in this profile ar e described in Table 55, and an inventory of artifacts, vertebrate fauna, and miscellaneous item is provided in Table 5-6. In the paragraphs that follow we describe the strati graphy of TU6 and follow with a description of the features. Fifteen distinct strata were recognized in the profile of TU6. As with TUs 5 and 8, the strata of TU6 can be gr ouped into four different macrounits, or depositional types: surface deposits, upper shell and sand, lower shell layers, and basal midden. The upper 25 cm (Str. I) consists of very dark grayish brown fine sandy loam with moderate amounts of whole Viviparus shell, followed by another 25 cm of similar matrix with a greater density of shell (Str. II). During excavati on these strata were assumed to be a postmining deposit, either fill that was redepos ited by heavy equipment during the mining operation, or simply the development of tops oil through natural pedo genic processes. However, as discussed earlier, neither of th ese scenarios seems very likely in retrospect.

PAGE 152

Silver Glen Run, Locus A (8LA1-West) 141 Figure 5-9. View facing northwest of Profile Cut of Test Unit 6, 8LA1-West, Locus A. Figure 5-10. View facing northwest of plan at base of Profile Cut, Test Unit 6, 8LA1-West, Locus A, showing Features 6-11 and related zones of disturbance; not shown is Feature 12, to the south of this view.

PAGE 153

142 St. Johns Archaeological Field School 2007-2010: Silver Glen Run Figure 5-11. Photograph and line drawing of north profile, Test Unit 6, 8LA1-West-Locus A.

PAGE 154

Silver Glen Run, Locus A (8LA1-West) 143 Table 5-5. Stratigraphic Units of Test Unit 6, 8LA1-West, Locus A. Max. Depth Munsell Stratum (cm BS) Color Description I 25 10YR3/2 very dark grayish brown fine sandy loam with surface root mat and moderate whole Viviparus shell II 50 10YR4/2 very dark brown fi ne sandy loam with abundant whole Viviparus shell III 61 10YR4/3 abundant whole Viviparus with occasional Pomacea shell and charcoal in trace of brown fine sand IV 88 10YR4/3 brown fine sandy loam with abundant whole Viviparus and occasional Pomacea shell and charcoal V 66 10YR5/3 brown medium ashy sand with abundant whole Viviparus and occasional charcoal VI 78 -abundant whole Viviparus shell with trace of charcoal and no matrix VII 105 10YR3/3 dark brown fine sand with abundant whole Viviparus and occasional Pomacea shell and abundant charcoal VIII 105 10YR2/2 very dark brow n fine sandy loam (possibly leaching from above) IX 129 7.5YR4/4 brown medi um sand with trace of shell X 102 10YR4/2 dark grayish brown medium sandy loam with abundant charcoal, trace of shell, and clasts of concreted sand; probable lens of fire-re ddened sand along western margin XI 144 10YR3/2 very dark grayis h brown medium sandy loam with crushed shell, including Unionid, and moderate charcoal XII 145 10YR4/3 brown medium sandy loam with moderate whole and crushed Viviparus shell and moderate charcoal XIII 200 10YR4/2 heterogenous mix of whole and crushed Viviparus shell with discontinuous stringers of crushed Unionid shell and occasional charcoal in dark grayish brown medium sandy loam XIV 211 10YR3/2 very dark grayish brown fine sandy loam with only trace of shell; grades into Stratum XV XV 219 10YR3/3 dark brown fine sandy loam lacking shell

PAGE 155

144 St. Johns Archaeological Field School 2007-2010: Silver Glen Run

PAGE 156

Silver Glen Run, Locus A (8LA1-West) 145 The contact between Str. I/II and the underlying layers does not express the unconformity of a cut-and-fill event that happened less than a century ago. Likewise, we cannot imagine that enough time has elapsed since th e mining to allow for the accumulation of as much as 50 cm of clastic material, especi ally considering the el evated position of the locations we tested in Locus A. Add to this the limited amount of clastic material available for upward translocation (through bioturbation) from stra ta immediately below Str. I/II and we are compelled to infer th at the upper unit was de posited by humans, much in the same fashion as the deeper sand strata. No pottery was found in Str. I/II of TU6 to corroborate the St. Johns associ ation inferred for this stratu m in TUs 5 and 8. Only two small St. Johns crumb sherds were recovere d from the entirety of TU6, one from the upper level of TU6 South (in sc ree deposits), and a second in explicably from Level J of TU6 South. On balance, the su rface stratum of the remnants of ridge in Locus A appears to be anthropogenic, apparently emplaced e ither during the St. Johns I period or well before, but was thereafter neither truncated nor buried. The second macrounit in the TU6 profile consists of Str. III through X, a complex array of shell and sand of variable composition, density, and thickness. Unlike its counterparts in TUs 5 and 8, the strata of TU6 do not clearly expr ess a dip-and-strike character, but they are not clearly horizontal either. Rather, this second macrounit includes several pit features that intruded upon shell and sand layers, resulting in profiles that do not clearly express th e structure of original depo sition. The major shell-rich stratum is the ca. 40-cm-thick Str. IV, consisting of abundant whole Viviparus and occasional Pomacea shell in a brown fine sandy loam. Below this across most of the unit is ca. 40-cm-thick stratum (Str. IX) of brown medium sand with only a trace of shell. Intercepting both of these strata is a large pit feature in th e northeast corner that appears to emanate from the top of Str. IV. It is not cl ear if this feature is truly cultural (in fact, it was thus not assigned a feature number in the field). Recorded in the profile as Str. VVII, this intrusion may have been caused by a tree-throw or similar disturbance, but if so it must have occurred before Str. I/II was emplaced. We are more inclined to interpret this as cultural given the apparent thermal activity represented in Str. V, at the top of the feature, similar in many respects to therma l features observed in TU5. A less ambiguous example of a large pit feature in TU6 is Feat ure 6, which is not seen in the north profile of Figure 5-11, but was sectioned, sampled, a nd dated, as discussed further below. Another possible intrusive feature in the north profile is reco rded as Str. X, a dark grayish brown medium sand with charcoal, a trace of sh ell, and clasts of concreted sand. Unlike the intrusion recorded as Str. V-VII, Str. X appears to emanate from the top of the sand layer (Str. IX), rather than the overlying shell. Because stratigraphic excavation in the south half of TU6 did not commence until we reached the sand layer (Str. IX), we cannot comment in detail on the distribution of bone a nd artifacts in the overl ying shell except to note that vertebrate fauna were relativel y numerous (n = 2028; 876.4 g in Profile Cut sample; Table 5-6), and pottery was entirely absent. One biface fragment, 8 flakes, 6 pieces of modified bone, and 4 fragments of ma rine shell were also recovered from the Profile Cut. The third macrounit of TU6 consists of Str. XI-XII, roughly 50 cm of well stratified shell and sand depos its consisting of laterally extensive, flat-lying, and

PAGE 157

146 St. Johns Archaeological Field School 2007-2010: Silver Glen Run heterogeneous strata of variable thickne ss (5-25 cm), much lik e those of TU5. The uppermost stratum (Str. XI) consists of ve ry dark grayish brown medium sandy loam with crushed shell, including Unionid, and char coal. This gives way to a 5-10-cm thick stratum (Str. XII) of brown medium sandy loam with moderate amounts of whole and crushed shell followed by a 25-cm-thick stra tum (Str. XIII) of dark grayish brown medium sandy loam with whole and crushed Viviparus and discontinuous stringers of crushed Unionid shell and charcoal. Throughout these strata were observed in plan small (10-20 cm) pockets of charcoal rich matrix and concreted shell that mimic the structure and disposition of post holes. Although these anomalies were recorded as possible cultural features in the plan dr awings of level forms, experi ence has since led us to infer that these are actually mineralized and/or burned root casts comm on to shell-bearing deposits across the site. Examples are illust rated in the north profile of TU6 (Figure 511) at the base of Str. XIII. The irregular outline of these anomalies in profile attests to their status as tree roots. De spite disturbances such as thes e, the strata of this third macrounit of TU6, like that of TU5, is indicati ve of repeated and generally consistent deposition of shell, vertebrate fauna, as h, charcoal, and occasional preceramic-aged artifacts on relatively flat, accretional surf aces. The density of vertebrate fauna and artifact is not terribly great throughout this macrounit, and, in fact, diminishes with depth (Levels B-E; Table 5-6). Still, repeated micr olayers of crushed shell, charcoal, and ashy sand attest to relatively intensive, repeat ed activity throughout the time this macrounit accumulated. The basal macrounit in TU6 consists of Str. XIV and XV, very dark grayish brown fine sandy loam with a tr ace of shell that lightens w ith depth as shell disappears altogether. This is essentially the burie d ground surface on which the upper strata accumulated. Although this fourth macrounit is generally devoid of shell, vertebrate fauna register an increase with depth over the macrounit above (Levels F-K; Table 5-6), accompanied by several lithic ar tifacts and pieces of modifi ed bone. A crumb sherd with spiculate (St. Johns) paste from Level J is clearly intrusive. Despite the lack of shell in this basal unit, we are confident that the ve rtebrate fauna and artifacts contained therein are not merely translocated down from above but are instead indi cative of occupations that slightly predate the accumulation of shell and associated materials. Because TU6, like all other test units in Locus A to date, was excavated for stratigraphic purposes, observati ons on matrices in planview were minimal, precluding the detection and sampling of all possible feat ures. However, TU6 revealed a relatively high frequency of pit features, some of whic h were recorded, sect ioned, and sampled as such. The largest of these was Feature 6, a 18+ cm deep basin some 60 by 75 cm in plan. Feature 6 appears to have been dug from the top of Str. IX, the brown sand layer, although it may very well have emanated from higher up, like the one noted above in the north profile. After recording it in plan, Feature 6 was sectioned and the south half removed for processing through -inch screen. The fill c onsisted of brown to dark brown fine-medium sand with moderate amounts of Viviparus and traces of other shell and a moderate density of vertebrate fauna. No artifacts were recovered from the south section. The north section was left unexcavated in the north pedestal of the unit. In profile, the basin expresses a flat bottom and slightly fl aring walls (a photograph of

PAGE 158

Silver Glen Run, Locus A (8LA1-West) 147 which can be seen in the north profile of Figure 5-11). Some concreted shell was observed along the southeast edge of the base, although it was not clear if this was truly a basal deposit or part of the matrix of the unde rlying stratum (Str. XI). A sample of wood charcoal from the fill of the south half of Feature 6 was submitted for AMS dating and returned a conventional ag e estimate of 5290 40 BP. Three other features were sampled in the south section of TU6; none of the features besides Feature 6 recorded in the north section of TU6 was excavated and thus remain intact beneath backfill. Feature 10 is a 20 x 35-cm circular pit whose upper portion was truncated by mining operations (F igure 5-10). The remnant observed at ca. 107 cmbs was sectioned and the south half re moved for -inch screening, and the north half recovered for flotation. The very dark grayish brown fine sandy loam of the fill contained equal amounts of bivalve and Viviparus shell, along with a moderate density of vertebrate fauna (including seve ral catfish spines), and some charcoal, but no artifacts. Feature 11 is a 15 x 15-cm circular patch of concreted shell in a brown sandy matrix. Only about 5 cm deep, the concreted mass wa s removed in its entirety to reveal an amorphous outline. It is likely that this mass, like the concreted masses at the base of Str. XIII, is merely a portion of a root cast. Finally, Feature 12 is a 21 x 22-cm circular pit with abundant charcoal and a sma ll amount of vertebrate fauna and Viviparus shell in a black sandy matrix. Upon s ectioning the feature and re moving the south half we determined it was likely to be a burned tree root. In sum, TU6 generally recapitulated the re sults of stratigraphic testing in TUs 5 and 8, but with less complexity to the upper sh ell and sand strata and a greater frequency of pit features. The ar tifact content and radiometric age estimate of strata in TU6 also matches those of TUs 5 and 8. Deviations be tween the two units in terms of stratigraphic sequence became interpretable following the ex cavation of Test Units 9, 10, and 15, to which we now turn. Test Units 9, 10, and 15 The largest exposure of st ratigraphy in the Locus A sh ell ridge came in 2008 with the excavation of three contiguous 2 x 2m units (TUs 8, 10, and 15) in a mining escarpment in the west-central portion of the ridge (Figure 5-1). The escarpment in this location is one of the few that deviates from the tendency for escarp ments to run parallel with the length of the ridge and the spring r un. Not quite perpendicu lar to the long axis of the ridge, the rectangle formed by these three units provided a good opportunity to view stratigraphy in the core of the deposit. Moreover, the six-meter-long profiles of these contiguous units enabled observations to be made about the horizontal relationships among depositional units lying at the same stratigraphic level. What came to be known to field school students as the trench pr oved to be insightful about stratigraphic variations observed in the earlier, smaller e xposures, particularly the relationship between emplaced sand and shell. The strategy for excavating the trench was similar to that used for the other test units but with some modifications. Having st aked out three contiguous 2 x 2-m units in

PAGE 159

148 St. Johns Archaeological Field School 2007-2010: Silver Glen Run Figure 5-12. View facing south of the excavation of Test Unit 9 (foreground) and Test Unit 10 (background), 8LA1-West, Locus A. the slope of the escarpment, the end units were opened first, TU 9 at the north end, TU10 at the south end (Figure 5-12). The intervening unit (TU15) was left intact until units of either side of it were completed in order to preserve the resp ective north and south profiles. Otherwise, excavation procedures followed those outlined above for the earlier test units. Each of the three test units wa s subdivided into west and east subunits, with the western (upslope) ones unexcavated below the depth of the Profile Cuts (~50 cmbs), and the eastern (downslope) subunits excavated completely to base. Excavation of the subunits proceeded without surprise as the lo wer deposits mimicked those of earlier test units, with exception of a greater level of c oncretion near the base, requiring field school students and supervisors to wiel d rock hammers and pick axes to penetrated to the base. Considerable effort was expended on reco rding stratigraphy of the trench upon completion of excavation. Profile drawing fo llowed the usual protocols, but was tedious given the length and complexity of the exposure (Figure 5-13). Photographing the trench profiles, however, required mo re than the usual method. To capture the detail of the 6-mlong profile, we collected a se ries of close-up digital shots at ~ 40-cm intervals along stacked horizontal transects. Individual photographs were stitched together in Adobe Photoshop to create the photo mosaic of the entire profile shown in Figure 5-14. The corresponding profile drawing is presented in Figure 5-15, with descriptions of all recorded strata provided in Table 5-7 and an inventory of all artif acts, vertebrate fauna, and miscellaneous items given in Tables 5-8 and 5-9.

PAGE 160

Silver Glen Run, Locus A (8LA1-West) 149 Figure 5-13. Recording stratigraphic profiles of th e trench (TUs 9, 10, 15), 8LA1-West, Locus A. The same four macrounits observed in profiles at the other two test locations are evident in the west profile of the trench. Howe ver, intervening in the center of the profile of the trench is a massive di sturbance, labeled in Figure 515 as Feature 21. This anomaly was initially interpreted as a large pit feature, but once the lower por tion was exposed, the tap root to a mature tree became apparent. Unfortunately, the disturbance is positioned directly between two distinct deposits in what we refer to as the second macrounit in the other test units. Although the facies of th ese depositional units is compromised by the disturbance, enough of the profile is pres erved to make sound inferences about the sources of lateral variation, its relationship to the macros trata or other units, and, ultimately, its relevance to interpreting site use. The upper macrounit of the trench is sim ilar to those of the other units and therefore warrants no further di scussion except to note that th e massive disturbance of the tree extends to nearly the surface. This does not imply that the tree wa s recent if we are to accept that this upper macrounit was emplaced l ong ago, and not simply the byproduct of mining operations in 1923. Comprised of ca. 20 cm of very dark grayish brown fine sandy loam with only occasional whole Viviparus shell, Str. I at the very top is consistent with upper strata observed elsewhere, but the remaining portion of this macrounit (Str. II) varies wildly in color, content, and struct ure. The seeming disarray of these underlying strata, particularly at the north end of the profile, is probably related to the adjacent tree disturbance, but that remains uncertain. Cl early the disturbance ha d its greatest impact on the upper macrounit, arguably affecting mo re than three-fourths of the profile.

PAGE 161

150 St. Johns Archaeological Field School 2007-2010: Silver Glen Run Figure 5-14. Composite photograph of the west profiles of Test Units 9, 10, and 15, 8LA1West-Locus A.

PAGE 162

Silver Glen Run, Locus A (8LA1-West) 151 Fi g ure 5-15. Line Drawin g of west p rofile of Test Units 10, 15, and 9, 8LA1-West, Locus A.

PAGE 163

152 St. Johns Archaeological Field School 2007-2010: Silver Glen Run Table 5-7. Stratigraphic Units of West Profile of Test Units, 9, 10, and 15, 8LA1-West, Locus A. Max. Depth Munsell Stratum (cm BS) Color Description I 47 10YR3/2 very dark grayish brown fine sandy loam with surface root mat and occasional whole Viviparus shell IIa 75 10YR3/1 heterogeneous matrix of very dark gray sandy loam with Viviparus shell, charcoal, sand stringers, and roots IIb 70 10YR4/2 dark grayish brown sand with whole Viviparus shell IIc 92 10YR4/2 whole and crushed Viviparus and Pomacea shell in dark grayish brown sand IId 76 10YR3/2 hetereogeneous shell in very dark grayish brown loam IIe 54 10YR5/3 whole Viviparus and Pomacea shell with charcoal in trace of brown sand IIf 74 -crushed and burned Un ionid shell with ash and no sand III 110 10YR3/2 very dark grayish brown sandy loam IIIa 118 10YR5/3-4 brown to yell owish brown ashy sand with crushed Viviparus shell IIIb 139 -whole Viviparus with little matrix IIIc 133 10YR4/3 brown sandy loam IIId 98 10YR3/2 hetereoge neous shell in very dark grayish brown loam IIIe 98 -whole and crushed Pomacea shell, no matrix IV 157 10YR4/2 heterogeneous whole and crushed shell including abundant crushed Unionid in dark grayish brown sand V 160 -whole and crushed Viviparus and Unionid shell with minimal matrix and occasional charcoal (5130 BP) VI 158 -whole and crushed Viviparus and Unionid shell with minimal matrix VII 167 10YR4/2 heterogeneous, abunda nt crushed shell in dark grayish brown sand (515050 BP) VIII 180 -whole and crushed Viviparus and Pomacea with minimal matrix

PAGE 164

Silver Glen Run, Locus A (8LA1-West) 153 Table 5-7. Continued. IX 208 -whole and crushed Pomacea with occasional Viviparus shell in minimal matrix IXa 196 10YR6/6 whole Viviparus shell in trace of brownish yellow sand X 195 10YR5/4 yellowish brown fine ashy sand with abundant whole Viviparus and occasional crushed Unionid shell and charcoal throughout XI 228 10YR4/2 dark grayish brow n ashy sand with moderate whole and crushed Viviparus Pomacea and Unionid shell and charcoal throughout; crushed Unionid shell lens at top of stratum XII 255 10YR5/6 yellowish brown sand mottled with 10YR4/2 dark grayish brown sand with occasional crushed shell and charcoal XIII 256 10YR5/2 whole and crushed Viviparus Pomacea and Unionid shell in grayish brown ashy sand with charcoal throughout XIIIa 259 10YR4/2 whole and crushed Viviparus Pomacea and Unionid shell in dark grayish brown sand with occasional charcoal (540040 BP) XIV 305 10YR4/3 concreted ashy br own sand with occasional whole and crushed Viviparus and Unionid shell and flecks of charcoal throughout XIVa 270 10YR3/2 very dark grayis h brown sand with occasional whole and crushed Viviparus shell XV 310 10YR6/2 light brownish gray medium sand lacking shell The second macrounit observed elsewhere cons ists of complex layers of shell and sand, often in tilted or dipping depositional struct ures. Portions of the trench profile not affected by the tree disturbanc e reflect the full range of va riation seen in the second macrounits of the other profiles. At the s outh end of the profile, Str. IV-VIII represent alternating thin layers of sh ell in varied condition and w ith varying amounts of clastic material, similar to those seen in the west pr ofile of TU5. In the tren ch profile, the strata of these microunits dip to the no rth. At the top of this sequence (Str. IV) is a relatively thick layer of dark grayish brown sand with whole and crushed shell, including abundant crushed Unionid shell. This stratum lies conformably on a th in lens of crushed shell, followed immediately by whole and crushed Viviparus and Unionid shell with minimal matrix and occasional charcoal (Str. V). Ch arcoal pulled from a bulk sample of Str. V

PAGE 165

154 St. Johns Archaeological Field School 2007-2010: Silver Glen Run

PAGE 166

Silver Glen Run, Locus A (8LA1-West) 155

PAGE 167

156 St. Johns Archaeological Field School 2007-2010: Silver Glen Run

PAGE 168

Silver Glen Run, Locus A (8LA1-West) 157

PAGE 169

158 St. Johns Archaeological Field School 2007-2010: Silver Glen Run

PAGE 170

Silver Glen Run, Locus A (8LA1-West) 159

PAGE 171

160 St. Johns Archaeological Field School 2007-2010: Silver Glen Run returned an AMS age estimate of 5130 40 BP. The three successive strata beneath Str. V are each separated buy a thin lens of crushed shell, the next (Str. VI) consisting of whole and crushed Viviparus and Unionid shell with minimal matrix, followed by (Str. VII) abundant crushed shell in a dark grayish brown sand, and then (Str. VIII) whole and crushed Viviparus and Pomacea shell with minimal matrix. A piece of charcoal pulled from a bulk sample of Str. VII return ed an AMS age estimate of 5150 50 BP, statistically identical to the age estimate of the charcoal two layers above. The last four strata just described at the south e nd of the profile, ca. 157-180 cmbs, appear to represent rapidly successive deposition acts, each followed by activity that led to the crushing of shell at the surf ace (e.g., trampling). Charcoal is prevalent in two of these crushed shell lens es. A series of root casts interrupt the sequence in three places, but the larger subunit was not truncated by the central tree disturbance (Feature 21), as was apparently the overlying Str. IV. The base of this mounded sequence of sh ell takes a dip upward to the north, against the dip of its upper strata. The base of Str. XIII is likewise underlain with a thin lense of finely crushed shell, which was tr uncated by the central tree disturbance, but continues on the north end of the profile, where its upward dip flattens. Below this crushed shell across the southern three-fourths of the profile is a thick stratum of shell. The southern subunit of this larger, deeper stratum (Str. IX) consists of crushed and whole Pomacea shell and occasional Viviparus shell with virtually no clastic matrix. Continuing northward the stratu m becomes dominated by whole Viviparus shell in a trace of brownish yellow sand (Str. IXa). This stratum overlaps the south-dipping Str. X, consisting of yellowish brown fine ashy sand with abundant, whole Viviparus shell, occasional crushed Unionid shell and particul ate charcoal throughout. Strata IX and X complement one another to form a 20-30 cm thick layer, with Str. X clearly emplaced first. Beneath this layer is yet another thin lens of crushed shell, this one extending across the entire profile at a flat grade. This deepest crushed shell lense, ~200 cmbs, marks the base of the second m acrounit and the top of the third. As described earlier for the other test units, the third macrounit of the ridge consists of laterally extensive and hetero geneous shell and sand layers with abundant charcoal, ash, and occasional concreted shell/sand, and with a vertebrate faunal assemblage that decreases with depth. All this applies to the third macrounit in the trench (Str. XI-XIII). However the trench deposits we re far more concreted in places compared to those of TUs 5 and 6. Because of th e concretion, it was not easy to distinguish between this macrounit and the basal unit, in terpreted elsewhere as a sub-ridge midden, conformant with the ground surface on whic h sand and shell was emplaced in large quantities. Moreover, a series of pit-like features appears to extend beyond the base of the third macrounit, evident in the west profile as an undula ting contact between Str. XIV (concreted ashy brown to grayish brown sand with shell) and Str. XV (light briowning gray sand lacking shell). A sample of charco al pulled from the bulk sample of Str. XIIIa of the third macrounit returned an AMS ag e estimate of 5400 40 BP, the oldest age estimate for the ridge, and indeed the entiret y of 8LA1. Minimally, about two centuries separate the deposition of th e second and third macrounits in the trench, while the age

PAGE 172

Silver Glen Run, Locus A (8LA1-West) 161 estimates for the third macrounits in TU5 and the trench overlap statistically by a large margin. The distribution of artifacts and vertebrate fauna in the trench provide insight on the processes resulting in differential depositi on. Gross comparisons across the test units from north to south are instructive. The northernmost unit, TU9, had the lowest density of vertebrate fauna (n = 2998; 1589.0 g), while the units to the south (TUs 10 and 15) each had over three times the bone (TU10: n = 10,588; 4278.9 g. TU15: n = 10,646; 5125.4 g). Test Unit 9, however, had its share of arifacts including five bifaces/biface fragments, eight uiltized flakes, and 13 pieces of marine shell. The adjoining TU15 had a comparable artifact assemblage but TU10, to the s outh, yielded only a few tools and less marine shell. Of course, these figures are not terribly indicative of va riations within each test unit, and indeed there are measurable di fferences in the density of vertebrate fauna and artifacts from top to bottom within units. We noted already the diminished density of vertebrate fauna with depth in the third m acrounit. The method of excavation did not permit nuanced comparisons of strata crosscut by excavation levels, that is, those with dipping profiles. Still, we can refine the comparison across at least one macrounit for levels within intact subunits. For instance, the combined faunal assemblage of Levels AE in TU9 (the north half of the second m acrounit, dominated by the thick sand-and-shell stratum, Str. X) amounts to 988 pieces weighi ng 520.6 g. In its count erpart at the south end of the profile, where thinly stacked shell layers are separated by crushed shell (Str. V-VIII), the frequency of bone is more than three times greater by count (n = 3551) and nearly three times by weight (1304.8 g). Th e contrast in tool frequency noted above holds as well, albeit among a small subset of the total assemblage from the trench. In all aspects, the stone, bone, and shell assemblage is consistent with the Mount Taylor estimates from AMS assays, as well as the material culture of the tradition found throughout 8LA1 and elsewhere in the region. Finally, we note that the trench actually yielded a relatively large assemblage of potte ry (n = 130), although ju st under two-thirds (n = 83) are crumb sherds. The remaining inventory consists of 12 small Orange Plain sherds, and 35 small St. Johns Plain sherds. As with the other test units, sherds in the trench were concentrated in the scree deposit s of the mining escarpment. In addition, the central tree disturbance accounted for a large fr action of the assemblage. On balance, the artifact inventory and AMS assays from the trench substantiate the inference that the entire shell ridge of Locus A, with the possible ex ception of its upper macrounit of topsoil, formed during the Mount Tayl or Period, ca. 5400-5150 BP, or 6300-5750 cal BP. INTERPRETATION OF STRATIGRAPHY Thus far we have available for observation three stratigraphic profiles of the shell ridge at Locus A that reached the pre-ridge surface: A 2 m-wide section at the eastern end of the ridge (TU5); a 2-m section near the western end of the ridge (TU6); and a 6-mwide section in between (TUs 9, 10, 15). The shallowest profile (TU6) is a bit over 2 m tall; the other two are just over 3 m tall. The previous sections provided details on each of the stratigraphic sequences, artifact and ve rtebrate fauna distribu tions, and radiocarbon assays. In this section we provide an interpretive sketch of the stratigraphic sequences

PAGE 173

162 St. Johns Archaeological Field School 2007-2010: Silver Glen Run observed to date. Although our sample is relatively small considering the large size of the total deposit, our test units were distri buted widely across available escarpments, stretching nearly 100 m from east to west. Despite obvious variation in the composition of each profile, the similarities far outweigh the differences, and radiocarbon assays show that all accumulations are coeval. Figure 5-16 is a schematic illustration of all profiles arranged in geographic order from west to east, and scaled for relative elevation using our arbi trary values from the site-wide grid (with Datum A at 10.00 m). At the base of each profile lies a pre-ridge surface stratum (i.e., buried A horizon) that has been organically enriched through the addition of vertebrate fauna, ash, charcoal, paleofeces, occasional shell, and related byproducts of human activity. Th e density of artifacts and food remains is not especially great in this basal midden, and in many places it is partially concreted. Accumulated over this original groun d surface/midden are a series of thin, lateral extensive and flat strata of shell and sand, again with abundant ash, some charcoal, and a tendency to be concreted in some places. A two-sigma ca librated age range of 6190-5940 cal BP from a charcoal sample in TU5 overlaps 70 years wi th the calibrated age range of a charcoal sample from a counterpart pr ovenience in the trench (6290-6 120 cal BP). More samples are needed to substantiate the inference th at occupation of the landform at this time extended the full length of the shell ridge, but nothing in the stra tigraphy and its dating thus far undermines that inference. The tops of these accretional macrounits in all test units began to receive deposits of brown sand without delay an d on the tops of sand strata some 20-40 cm thick shallow pits (basins?) were often dug and presumably used for functions involving heat (e.g., fire hearths). Only one such pit feature has been dated: Feature 6 in TU6, with a calibrated age range of 6190-5940 cal BP, ag ain a 70-year overlap with the lower age estimates and thus all possibly coeval and not likely separated by more than a few decades. What are the possible sources for the sand emplaced on accretional midden? Sand is hardly at a premium in the immediate vicinity, but if the th ree profiles we have observed truly reflect the emplacement of 20-40 cm of sand over midden across the entire extent of the shell ridge, then about 2000 cubic meters of sand are implicated. Remarkably, that is about the volume of the s inkhole that lies only 50 m south of the very center of the ridge (Figure 5-16 inset). Assumed to be a na tural collapse feature, this ~2500 m3 depression may actually be the borro w pit from which sand was excavated long ago. An effort to investigate th is possibility is clearly warranted. The presence of a ridge-wide stratum of emplaced sand implicates the excavation and transport of a large volume of fill 6000 years ago. Elsewhere in the region, Mount Taylor communities mounded sand, apparently as part of mortuary practice (e.g., Aten 1999). Sand may have been inci dental to a larger progra m of shell mounding early on, but by the Thornhill Lake phase of the Mount Taylor period (ca. 5500 cal BP, a century or two after Locus A at 8LA1-West reached its full form), sand mounds were at least occasionally built expressly for mortuary purposes (Endonino 2010). The shell ridge at 8LA1-West was never expressly mortuary, and despite the presence of two subadults in

PAGE 174

Silver Glen Run, Locus A (8LA1-West) 163 Figure 5-16. Schematic model of the stratigraphic sequences observed in test unit profiles of 8LA1-West, Locus A.

PAGE 175

164 St. Johns Archaeological Field School 2007-2010: Silver Glen Run TU8 at the east end of the ridge, we have no indication that other burials exist in the ridge, let alone interred in dedicated cemeter ies. Rather, the emplacement of sand may be more generally tied to traditions of renewal or rejuvenation, as in the cleansing of an extant living surface on which food remains an d other debris had accumulated. Certainly such actions can be considered practical, but worthy of consideration is the possibility that the emplacement of sand was more a symbolic act that a practical affair. The pattern of deposition changed markedly shortly after the accretional deposits were laid and after the first sand layers we re put in place. Deposits of shell and sand thereafter accumulated in nodal fashion, pres umably in low mounds no more than a few meters in diameter and about one-half mete r high. We have not exposed enough of a profile to see one of these uni ts in full cross-section, but we have a good sense of their form and internal structure from investiga tions at Hontoon Dead Creek Village (8VO215) to the south (Randall 2010). At this form er field school site sa nd may not have been mobilized to build house mounds because of the prevalence of shell, but no matter the material used, the outcome is the same: a slightly elevated platform on which some sort of structures may have been built. We remain frustrated by the lack of direct evidence for architecture (e.g., postholes), or bona fide house floors. We do, however, have ample evidence for activity surfaces in the form of crushed shell lenses. The stacked sequence of crushed shell lenses in the south ha lf of the trench profile is perhaps a good example of the accretion of shell midden adj acent to houses sited on low sand mounds. If so, there would appear to have been a more spatially differentiated use of the ridge after the sand was emplaced that involved separation s of different materials, perhaps under different material circumstances or new cultural preferences. The emphasis of investigation at Locus A has been on the stratigraphic sequence of the mining escarpments. Little attention has been given thus far to the artifact and vertebrate fauna assemblages other than to no te they are consistent with a Mount Taylor cultural affiliation. There is one other very important point to be made about the artifact assemblage recovered from our test units of the ridge. Figures 17 -19 present respective samples of the lithic, modified bone/antler, a nd marine shell artifacts recovered from all test units. This may seem like a small assemb lage for the amount of testing undertaken, it is far greater in density and diversity than the assemblage from Hontoon Dead Creek Mound (8VO214), whose excavated volume was more than twice that of Locus A. We have argued that mounding at this earlier M ount Taylor mound was ritualized, and that communities responsible for its accumulati on did not live directly on the mound (Sassaman and Randall 2012). That would not seem to be the case at Locus A, the apparent dwelling of communities that built houses atop accretional ridge, and processed foods, burned fires, discarded inedible wa ste and broken tools, and all other tasks associated with domestic living. The cont rast between Hontoon Dead Creek Mound and the ridge at Locus A reminds us that internal differences in what appear to be similar deposits (i.e., linear shell ridges) are to be expected, reflecting both the mudane and ritual aspects of Mount Taylor living. Moreover, the emplacement of sand at Locus A does to show that the line be tween ritual and mundane practice cannot be drawn too sharply for people who were not subject to the sens ibilities of the modern distinction.

PAGE 176

Silver Glen Run, Locus A (8LA1-West) 165 Figure 5-17. Select lithic artifacts from test units of Locus A. a. TU8-L-Zone A; b. TU15B-A-5; c. TU15A-profile cut; d. TU5-profile cut; e, f. TU9A-D; g. TU5 East-H-Zone C; h. TU5 East-JZone E; i. TU9A-H; j. TU6 South-B; k. TU9A-H; l. TU5 East-K-Zone E; m. TU9A-N.

PAGE 177

166 St. Johns Archaeological Field School 2007-2010: Silver Glen Run Figure 5-18. Modified bone and antler from test units of Locus A. TU6-South: a. C-3; b. D-1; c. E-1. TU9A: d. J-1; e. G-1; f, g. B-Zone A-2. TU10B: h. C-1; i. A-7. TU10A: j-l. I-2. TU15A: m, t. profile cut, Zone B-1; r. profile cut-1. TU 10: n-q. profile cut-1; TU15B: s. baulk Zone B-2; u. C-1; 10B: v. A-7; TU5: w, x, ff, gg. profile cut; TU5 East: y, z. D-Zone A-2; aa, bb. J-ZoneE2; TU8: cc. D-2; dd, ee. D-1.

PAGE 178

Silver Glen Run, Locus A (8LA1-West) 167 Figure 5-19. Marine shell artifacts from test units of Locus A. a. TU8-M-Zone A; b. TU9A-M; c. TU10A-D-1; d. TU8-K-Zone B; e. TU6 Surface.

PAGE 179

168 St. Johns Archaeological Field School 2007-2010: Silver Glen Run Finally, the cap of earth on top of all the profiles begs explanation for the accumulation of upwards of 50 cm of clas tic material on the landform apparently removed from significant sources of alluvial colluvial, and aeolian deposition. We at first attributed the upper stratum (topsoil in Figure 5-16) to pedogenesis following mining, that is, soil development since 1928. Given the lack of a obvious source of natural deposition, this scenario seems unlikely, nor is it likely that this surface stratum was the spoil of mining, because it is too pe rvasive and uniform to have been merely happenstance. We suspect it was emplaced by humans, in a capping event not unlike the earlier caps of sands, and not unlike th e shell capping of Locus B (see Chapter 6). There is certainly sufficient development of this stratum (organic enrichment and bioturbation with shell strata below) to sugg est it has been in place for a long time. The occasional Orange and St. Johns period sherds in this stratum may signal a post-Mount Taylor activity, but it seems equally likely that Mount Taylor communities capped the ridge after abandoning it as a place of dwelling, and that later Orange and St. Johns period dwellers in the vicinity occasionally used the ridge for activities involving pottery. CONCLUSION Despite extensive damage from shell mining in 1923, the ridge at Locus A contains remnants of upwards of 3 m of stra tified deposits with excellent archaeological potential. Six 2 x 2-m test units excavated in three locations of the ridge reveal a consistent sequence of basal midden, accretional shell and sand, house mounds and associated midden accumulation, and capping with sand, all elapsing over a three-to-five century period of the Mount Taylor phase, ca. 6300-5750 cal BP. All indications are that Mount Taylor communities actually resided on this ridge as it accumulated, eventually with but at first apparen tly without constructing house mounds and imposing a formal spatial order on the emplacement of sand, shell, and the outputs of daily living. This pattern of dwelling stands in co ntrast to Mount Taylor shell ridges that lack evidence for domestic activities, but compares favorably to the one known linear village (8VO215), which involved the use of small shell mounds, presumably for domestic dwelling. Much remains of the shell ridge at Locu s A, and further work is warranted. Before delving into additional mining escarpm ents, however, two other areas of inquiry demand attention. First, stratigraphic ex cavations at Locus A have emphasized the vertical record of Mount Tayl or site use (i.e., change over time), and lacking have been data on the spatial structure of dwelling at any given moment of time. The trench profile shows good promise for locating evidence for sp atial patterning in the siting of houses, middens of secondary deposition, and related domestic activities. Although most such evidence was carted away long ago by shell miners, mining stopped well short of the basal deposits, so good potential does exist fo r examining laterally ex tensive areas within a single stratum. A large block excavation in the mining pit will be needed, perhaps preceded by some remote sensing to detect s ubsurface features such as hearths and pits. The second pressing issue is the possible bo rrow pit to the south of the ridge. Some strategic coring and remote sensing may help to detect evidence that sand was removed from this depression 6000 years ago, but we can start by simply comparing the sand from the mound to a profile adjacent to the depression to see if it matches the texture and color

PAGE 180

Silver Glen Run, Locus A (8LA1-West) 169 of the emplaced sand. Of course, pedogenic processes since the time of excavation and emplacement of sand may have obscured rele vant evidence, so it may take broader sampling in the vicinity to know how much pe dogenic variation can be expected under a range of edaphic and topographic cond itions. Additional conclusions and recommendations for more work at Locus A are included in the concluding chapter of this report.

PAGE 181

170 St. Johns Archaeological Field School 2007-2010: Silver Glen Run

PAGE 182

171 CHAPTER 6 SILVER GLEN RUN, LOCUS B (8LA1-WEST) Zackary I. Gilmore Locus B occupies a relativel y flat, well drained ridge nose less than 200 m to the southwest of Locus A and approximately 80 m south of Silver Glen Spring run. It consists of a small, roughly crescent-shaped shell node and the extensive archaeological deposits that surround it. A previously unknown archaeological resource, Locus B was initially recorded as a result of reconnaissance survey conduct ed by participants in the St. Johns Archaeological Field School in 2007 (Chapter 4). Despite being relatively modest in size and depth compared to the shell ridges at 8LA1-East and Locus A, Locus B contains well-stratified and largely intact de posits dating primarily to the Late Archaic Orange (4600-3600 cal BP) and preceramic Mt. Taylor (7300-4600 cal BP) periods, although subsequent St. Johns peri od artifacts are also present. Given the especially well preserved deposits encompassing multiple culture-historical components, this portion of the site presents a virtually unparalleled oppo rtunity for investiga tion of Late Archaic ritual and domestic practices conducted outside of the more extensively studied shell mound contexts. Moreover, its close proximity to the concu rrently utilized shell mound at Locus A renders Locus B a uniquely approp riate setting for studyi ng the relationship between these two contrasting t ypes of Archaic places. Following shovel testing in 2007 (see Chapter 4), Locus B was recognized as an area that warranted more intensive investigatio n due to the identification of an arcuate or circular concentration of Orange ceramic sherds presumed to be indicative of an Orange period habitation or village site. Consequently, between 2007 and 2011, Locus B was the target of rigorous field investigations that included three primary strategies: (1) topographic mapping and closeinterval coring for the purpose of establishing the horizontal and vertical extent of cultural deposits; (2) extensive test unit excavations to determine the vertical structure of these deposits and their variability across Locus B; and (3) intensive block excavations intended to expose relatively fine-scale horizontal and vertical patterning of cultural materials, as well as eviden ce of architectural remains and features. Although unequivocal evidence for an Orange period village has proven elusive, the results of these investiga tions have revealed three successive, and fundamentally distinctive patte rns of site use. Together, these patterns encompass virtually the entire Late Archaic, a dynamic period of region-wide material and social transformation in the middle St. Johns Valley. This chapter details the methods and initial results associated with each of the three testing strategies through 2010, as well as descriptions of feature contents and artifact assemblages for samples analyzed to date. In addition, preliminary interpretations are offered regarding the historical circumstances surrounding the shifting uses of Locus B as well as their relationship to coeval developments at other areas of the Silver Glen Run complex a nd the broader region. Excluded from this chapter are the results of excavations in 2011, analyses of which are ongoing and will detailed in a subsequent report.

PAGE 183

172 St. Johns Archaeologi cal Field School 2007-2010: Silver Glen Run SITE MAPPING AND SUBSURFACE AUGERING The mapping of Locus B was conducted in accordance with a permanent eastwest baseline established at 8LA1E in 2007 (see Chapter 3). Two permanent reference points (Datum A and Datum B) were created to form this line and the western point (Datum A) was assigned the coordina tes of N1000.00 E1000.00, with an arbitrary elevation of 10.00 meters. Surface mapping of Locus B was conducted in the spring of 2009 using a Nikon DTM-310 Total Station. Tw o additional permanent data (Datum C and Datum D) were established in the bait fiel d at the north end of Locus B, and several temporary stations were established to a llow for relatively comprehensive mapping of Locus B while minimizing the extent of vegeta tion removal required in order to establish clear lines of site. Pin flags were used to ma rk and keep track of r ecorded points in order to ensure complete coverage. In total, 335 transit points were recorded across Locus B. The resulting map largely recapitulates the topography discernable from LiDAR data collected by the Volusia County Public Works Department (2006). As illustrated in Figure 6-1, Locus B forms a slightly ar cuate and relatively subtle topographic prominence that opens northward, toward Silver Glen Run. It stretches for approximately 60 m along its longest axis (eas t-west), roughly paralleling the natural terrace on which it sits. At its highest point Locus B rises a pproximately 1 m above the surrounding terrain. It slopes downward relatively steeply to the nor th toward the spring run (at least partially due to the presence of a modern gravel road) and more gently in all other directions. Extensive subsurface testing was perf ormed in 2008 by field school students using an Oakfield soil tube with a 3/4-in ch diameter and a maximum depth of 85 cm. Although originally planned to cover all of Locus B and the surrounding area, the difficulty and time involved in punching the soil tube through dense and often concreted shell midden hampered this goal. Tests were conducted at 2-meter intervals within larger 10 x 10-m blocks. Each of these blocks was oriented along approximate cardinal directions using a sighting compass and measur ed out using 30-m cloth measuring tapes. Pin flags were used to mark the corners of ea ch block as well as th e intervening locations of each planned auger test. Ultimately, six contiguous 10 x 10-m blocks were completed that together cover the hypothesized core ar ea of Locus B from the western edge of the topographically visible shell node to its central apex (Figure 6-2). All auger tests were conducted to the maximum depth allowed by the soil tube (approximately 85 cm) except for the instances in which impenetrable concreted shell midden was encountered. For each test, information was recorded on a log sheet regarding the constituents of each strata en countered (i.e., the type of soil matrix, the density and condition of shell, and the occurrence of artifacts and other cultural materials) along with the depths of stratig raphic transitions. Unfortunate ly, as tests and logs were completed by a number of different field schoo l participants and the detail with which observations were made and recorded vari ed significantly from person to person, the quality of data conferred by the auger survey was not entirely consis tent across the tested area. Nevertheless, a great deal of useful in formation was gathered regarding the vertical extent of recent near-surface disturbance, the thickness and density of shell midden

PAGE 184

Silver Glen Run, Locus B (8LA1-West) 173

PAGE 185

174 St. Johns Archaeologi cal Field School 2007-2010: Silver Glen Run deposits, and the basic morphology of the Locus B surface prior to the initiation of shell deposition. While virtually every auger test indicated the presence of shell, a tremendous amount of variation was demonstrated in terms of the depth and thickness of shell deposits. Much of this variation can be s een in Figure 6-3, which shows fence diagrams of subsurface deposits along two perpendicular transects at Locus B based on auger tests and subsequent test unit excavations. Each of the vertical black bars represents the position of an auger test or te st unit corner along the transect although some are offset by up to 2.5 m perpendicular to the actual transe ct line. The diagrams show a consistent zone of disturbance related to historic plowing and recent bioturbation ranging from approximately 5-30 cm thick across this enti re portion of the site. The intact shell midden underlying this disturbed stratum varies significantly in thickness along both transects. Shell deposits are most substantia l along the western por tion of the west-east line, exhibiting a maximum thickness of aroun d 1.5 m in an area subsequently found to contain large numbers of Late Archaic shell-filled pit features. Along the eastern margin of this transect, which coincides approximately with the apex of Locus Bs shell node, shell extends for just over 1 m below the su rface. Along the north -south transect, shell deposits are most substantial in the north and then taper gradually toward the south, eventually reaching a thickness of only about 20 cm. Aside from the areas dramatically altered by Archaic period pit-digging activit ies and those where the midden could not be completely penetrated by the auger, these diagrams indicate that shell was deposited on a roughly level sand surface with an absolute elevation of approximately 7.5-7.7 m (NAVD1988; based on local site datum). If this is accepted as the natural predepositional surface, then virtually all of Lo cus Bs modest topographi c relief at present can be attributed to the depositional activities of the sites Late Archaic inhabitants. Additional stratigraphic eviden ce supporting this conclusion is discussed below. Figure 6-2. Topographic map of Locus B showing locations of 2008 auger tests.

PAGE 186

Silver Glen Run, Locus B (8LA1-West) 175 Figure 6-3. Fence diagrams showing cross-sectional profiles of archaeological deposits at Locus B based on auger and test unit excavations. EXPLORATORY TEST UNIT EXCAVATIONS During the 2007 St. Johns Archaeological Field School, one 1 x 2-m test unit (Test Unit 4) was excavated at Locus B in an area that shovel te sting had indicated contained a concentration of Orange fiber-tempered pottery This excavation revealed

PAGE 187

176 St. Johns Archaeologi cal Field School 2007-2010: Silver Glen Run significant, stratigraphically intact cultural de posits dating to the Late Archaic Orange period. To further assess the nature and spatial ex tent of these deposit s, in 2009 and 2010 six more 1 x 2-m test units (Test Units 12-14, 19, 21 and 22) were excavated perpendicular to Test Unit 4, along a north-south transect bisecting Locus B (Figure 6-4). These units were not contiguous but were instead placed two meters apart in order to preserve intervening stratigraphic data. In addition, a six-meter gap was left between Test Units 14 and 21 and the location of the northernmost unit (Test Unit 22) was offset one meter to the west of the original north-south transect in order to avoid two large trees. Finally, in 2010 field school st udents excavated two additi onal exploratory test units intended to broaden our understa nding of other areas of the site. These included one 1 x 2-m unit (Test Unit 46) to the east near the apex of Locus Bs shell node and one 2 x 2-m unit (Test Unit 57) to the northwest in a cu rrently cleared and plowed bait field. All test units were hand ex cavated by trowel in arbitr ary 10-cm levels with the exception of the uppermost level, which, bein g heavily disturbed by thick roots and modern farming activities, was shoveled-scrap ed. A datum was set at the highest corner of each test unit from which level depths were measured. Fill was processed through 1/4inch screens and all artifacts, vertebrate fauna, and other cultural materials (excluding freshwater shells) were collected and bagge d according to provenience. Where clear archaeostratigraphic zones were identified within levels, these were mapped and the respective fills and artifacts kept separate. The floors of each level were inspected for these zones as well as the presence of cultura l or natural features. When recognized as such during excavation, features were mapped in plan view, bisected vertically, and then drawn in profile. Where possible, one half of the feature fill was then collected for 1/8inch water screening while the remaining half was removed as a bulk sample for flotation processing. In instances where large feature size preven ted complete sampling, bulk samples were systematically collected from di fferent areas of the feature (usually upper, middle, and lower sections), while the rema inder of feature fill was processed through 1/8-inch and 1/4-inch screens. Excavation in all test units proceed ed until reaching sterile or virtually sterile subsoil. In addition, 50 x 50-cm column samples were taken from the west profile of Test Unit 22, the west profile of Test Unit 43, and the north profile of Test Unit 46. Unlike the general test unit excavations, columns were excavated stratigraphically. Column strata that exceeded 10-cm in thickness were excavated in 10cm levels. Within each level, a one gallon sample was collected for flotation while the rest was removed for 1/8-inch water screening. Processing and analysis of bulk column samples has not yet been completed. The following discussion of individual test units begins with Test Unit 13, the southernmost unit in the north-south transe ct, and then proceeds northward, as this follows the general south-north progression of the Locus B shell midden from relatively thin and stratigraphically simple to thicke r and more complex. Test Units 47, 4, and 57 are then discussed individually because of th eir relatively detached spatial positions as well as the unique interpretive ch allenges that each presents.

PAGE 188

Silver Glen Run, Locus B (8LA1-West) 177 Figure 6-4. 2008 field school crew excavating a nor th-south transect of test units at Locus B, 8LA1W. Test Unit 13 Test Unit 13 (TU13) is a 1 x 2-m test unit placed approximately 10 m to the southeast of shovel test pit 22-2, in an attemp t to catch the southern margin of the shell midden at Locus B as indicated by shovel testing a nd auger data. Oriented north-south, this unit was excavated to a depth of 100 cm below datum (cmbd). Composite drawings and photographs of the stratigraphic profiles from all four of TU13s walls are shown in Figure 6-5, and desc riptions of the major stratigraphic units are provided in Table 6-1. Artif act counts for each level and z one are shown in Table 6-2. Corresponding with excavation Levels A and B, Stratum I is a 15 to 20-cm-thick A-horizon consisting of dark brown sand with occasional whole and fragmentary Viviparus (banded mystery snail) an d bivalve (freshwater mussel) shell. Dense root mat permeates the entire stratum with larger roots appearing near the bottom. The few cultural materials that were recovered from this stratum include sparse vertebrate fauna and highly fragmented plain and check-stamped St. Johns ceramic sherds. It was discovered subsequently, durin g the 2009 block excavation (dis cussed below), that Locus

PAGE 189

178 St. Johns Archaeologi cal Field School 2007-2010: Silver Glen Run B and its surrounding area had been subjected to historic plowing and that the upper 2030 cm of this part of the site has been signifi cantly disturbed. Cons equently, the sporadic occurrence of shell and artifact s throughout Stratum I may be at least partially a result of the repeated scraping and churning of the very top of the underlying midden. Stratum II appears to represent the top of the undisturbed shell midden at Locus B. It corresponds primarily with excavation levels C, D, and E and consists of an approximately 20 to 30-cm-thick layer of gray sand with abundant whole and crushed Viviparus shell mixed with a smaller amount of crushed bivalve. In terms of cultural materials, this stratum contained only sparse vertebrate fauna and a small number of plain St. Johns and Orange ceramic sherds. With in Stratum II, discrete areas containing abundant charcoal and burned shell were en countered along the northern and eastern margins of the unit that were collectively designated Zone A in the field. The clean, fresh appearance of the charcoal from Zone A, the irregular shape of the deposits, and the overall similarity of the zone to the rest of Stratum II (if evidence for burning is excluded) all suggest that Zone A is a rela tively recent intrusive disturbance, probably a tree that burned in place. A roughly circular pocket of whole and crushed Viviparus along the southern wall of the test unit was designated Zone B. Although initially thought to be of cultural significance, its horiz ontal orientation perpe ndicular to the test unit profile suggests instead that it is a natural disturbance. The soil and shell matrix constituting Zone B is identical to the gene ral fill of Stratum II and was probably dragged down by a burrowing animal, most likely a gopher tortoise ( Gopherus polyphemus) Zone B is surrounded by a conspicuous area of virtually shell-free, dark grayishbrown sand, visible in the south profile drawin g (Figure 6-5) as Stratum III. Occupying the same basic vertical position as Stratum II, this organically enriched pocket of soil is likely a byproduct of the activities of the bu rrower noted above. The majorities of both Stratum II and Stratum III correspond to excavation Levels C and D, although they do extend into lower levels in multiple places. A sharp stratigraphic break is visible between Stratum II and Stratum IV, the latter composed of lighter brown medium sand with only sporadic and isolated pockets of whole Viviparus shell. Shell is most common near the top of this stratum and decreases with depth, disappearing entire ly near the base of the test unit. Stratum IV largely corresponds with excavation Levels F through J. In locations where Stratum II deposits drop down and penetrate into Level F or below, these were designated Zone D, while the remainder of the test unit deposits, composed of Stratum IV sediments, were labeled Zone C. Overall, Stratum IV contains sparse vertebrate fauna and several lithic flakes but no pottery, making it likely that its constituent deposits are preceramic in age. This Stratum underlies the entire test unit, whic h was terminated at approximately 100 cmbd, although three lithic flakes were recovered from the bottom excavation level.

PAGE 190

Silver Glen Run, Locus B (8LA1-West) 179 Figure 6-5. Stratigraphic drawings and photogr aphs from profiles of TU13, 8LA1W. (Note: photographs are not to scale)

PAGE 191

180 St. Johns Archaeologi cal Field School 2007-2010: Silver Glen Run Table 6-1. Stratigraphic Units of Test Unit 13, 8LA1W. Stratigraphic Max. Depth Munsell Unit cm BD1 cm BS2 Color Description I 28 28 7.5YR2.5/2 Very dark brown historically plowed A horizon; abundant roots; occasional whole Viviparus II 68 67 7.5YR3/2 Dense whole and crushed Viviparus in a medium dark brown sandy matrix. III 58 57 7.5YR3/2 Dark brown organically enriched sandy matrix with no shell or visible cultural materials. IV 107 107 7.5YR4/6 Strong brown fine sand with isolated pockets of dense concreted shell. Table 6-2. Cultural Materials Recovered from Test Unit 13, 8LA1W. Level St. Johns Plain Orange/ T. I. Incised Orange Plain Crumb Lithic Flake Lithic Biface Marine Shell (g) Vert. Fauna (g) Historic Artifact A 3 15 38.0 B 3 25 2 33.0 1 C 7 45 2 1 117.9 D 2 2 49 2.4 103.9 E Zone A 5.1 E Zone B 0.3 E Zone C 2 2 31 2 0.3 66.1 F Zone B 1 2.9 F Zone C 11 1 44.7 G Zone C 6 1.1 18.2 G Zone D 0.4 H Zone C 4 19.7 I Zone C 2 2.9 J Zone C 3 Total 18 2 2 176 22 1 3.8 453.1 1

PAGE 192

Silver Glen Run, Locus B (8LA1-West) 181 Test Unit 12 Situated two meters to the north of TU 13, Test Unit 12 (TU12) exhibits similar macro-stratigraphic units but with some adde d complexities. Composite drawings and photographs of the stratigraphic profiles from all four walls of TU12 are shown in Figure 6-6, and descriptions of the major stratigrap hic units are provided in Table 6-3. Artifact counts for each level and zone are shown in Table 6-4. Seven distinct strata were identified in this test un it. Like TU13, Stratum I of TU12 consists of a dark brown organic A hor izon lying completely within the historic plow zone. It contains abundant small to medium juniper and palmetto roots along with occasional whole and fragmentary Viviparus and bivalve shell. Stratum I corresponds with excavation Levels A, B, and the upper portion of Leve l C from which were recovered a variety of cultural materials in cluding highly fragmented St. Johns Plain and Check-Stamped ceramic sherds, lithic debita ge, and a small amount of vertebrate faunal remains. The top of Stratum II marks the upper boundary of the intact shell midden in TU12. It consists of dense predominantly whole Viviparus shell in grayish brown medium sand. Cultural materials in this stratum are limited to St. Johns ceramics and vertebrate fauna. Stratum III is distinguis hed from Stratum II by the addition of occasional whole Pomacea (apple snail) and bivalve to the shell matrix as well as the appearance of plain and incise d Orange fiber-tempered ceramics. During excavation, a small area of dense concreted shell was noted in the northeastern corner of the test unit that penetrated into underlying strata but wa s never discrete enough to receive a feature designation. This area was labeled Zone B while the rest of the test unit was considered Zone A. Zone B produced no non-shell cu ltural materials and was terminated at the bottom of excavation Level G. Together, Strata II and III co rrespond roughly with excavation Levels C through F. Across most of the test unit, Stratum III s its atop a thin layer of dense burned and crushed bivalve mixed with a small amount of brown fine sand and occasional whole unopened bivalve shells most clear ly visible in the northern half of the unit (Stratum IV). Stratum IV slopes gently downward from north to south, para lleling the modern topography. It also th ins out and becomes wispier in that direction, eventually tapering out completely as evidenced by its complete absence from Test Unit 13. The point of contact between Strata III and IV was the most artifact rich level within TU12, yielding a variety of materials including Orange ceramic s, a bone tool, a marine shell bead, lithic debitage, and relatively abundant vertebrate fauna. Paleofeces were also recovered. The diversity of materials associated with ever yday activities, along with the finely crushed roughly horizontal layer of sh ell suggest that Stratum IV probably represents a Late Archaic living surface. At approximately the same elevation as Stratum IV, Stratum V consists of a discrete pocket of very dense whole Viviparus and abundant vertebrate fauna with only a small amount of interspersed fine sand. This deposits discreteness and its apparent

PAGE 193

182 St. Johns Archaeologi cal Field School 2007-2010: Silver Glen Run Figure 6-6. Stratigraphic drawings and photogr aphs from profiles of TU12, 8LA1W. (Note: Photographs are not to scale.)

PAGE 194

Silver Glen Run, Locus B (8LA1-West) 183 Table 6-3. Stratigraphic Units of Test Unit 12, 8LA1W. Stratigraphic Max. Depth Munsell Unit cm BD1 cm BS2 Color Description I 25 18 7.5YR3/2 Dark brown historically plowed A horizon; abundant roots; occasional whole and fragmentary Viviparus and bivalve shell. II 50 49 7.5YR2.5/1 Abundant whole Viviparus shell in a black medium sandy matrix. III 65 60 7.5YR4/3 Abundant whole and crushed Viviparus shell in a brown fine sandy matrix; infrequent Pomacea and bivalve shell. IV 72 67 7.5YR4/2 Dense burned crushed bivalve with a few whole unopened shells in a brown fine sandy matrix; abundant vertebrate fauna. V 80 81 7.5YR4/3 Dense whole Viviparus with very sparse brown fine sandy matrix; abundant vertebrate fauna. VI 127 127 7.5YR4/4 Brown fine sand with discrete pockets of dense shell and vertebrate fauna. VII 96 91 7.5YR4/3 Dense whole Viviparus with very sparse brown fine sandy matrix; infrequent vertebrate fauna. vertical correspondence with a hypothesized Late Archaic surface, suggest that Stratum V is probably actually a feature of some kind (perhaps a pit or an infilled post hole), although its position in the corner of the unit ma de this difficult to recognize in the field. The basal stratum throughout all of TU12 (Strat um VI) is a thic k layer of brown medium sand with occasional whole Viviparus, Pomacea, and bivalve shell. Within TU12, Stratum VI corresponds roughly with ex cavation Levels G through L. Although appearing in profile as relatively undiffe rentiated, Stratum VI contains abundant vertebrate fauna that decreases in density from top to bottom and exhibits scattered, apparently isolated deposits of shell and bone throughout. Beginning in excavation Level G, two seemingly discrete con centrations of concreted whole Viviparus shell were encountered. The first, Zone C, was located near the center of the test unit while the second, Zone D, was positioned approximately 25 cm to the southeast and intersected the east wall of the unit. By the bottom of Le vel I, (90 cmbd) these deposits had converged and were grouped together into Zone D. Neither of these zones contained significant non-shell cultural materials outside of trace amou nts of vertebrate fauna. In the east wall profile drawing (Figure 6-6), Zone D is labe led Stratum VII and appears as an amorphous feature descending down from Stratum IV. Its irregular shape and in tersection with Zone C suggest that Zone D is in all likelihood an old infilled animal burrow rather than a cultural feature.

PAGE 195

184 St. Johns Archaeologi cal Field School 2007-2010: Silver Glen Run Table 6-4. Cultural Materials Recovered from Test Unit 12, 8LA1W. Level St. Johns Plain Orange/ T. I. Incised Crumb Lithic Flake Mod. Marine Shell Marine Shell (g) Modified Bone Vert. Fauna (g) PaleoFeces (g) A 12 1 14.1 B 7 17 3 5.0 46.0 C 2 6 1 56.5 D 3 47 38.5 E 2 16 5 230.5 6.1 F Zone A 1 5 2 11 5.5 2 364.9 4.5 G Zone A 1 3.7 1 780.9 34.7 G Zone C 8.4 H 2 69.8 I Zone A 1 42.6 I Zone C 29.7 I Zone D 1 3.1 J Zone A 1 0.6 129.9 J Zone D 2.1 J Zone E 2.0 K 22.6 K Zone F 98.5 L 3 2.1 Total 13 2 106 18 1 14.8 3 1942.2 45.3 1Marine shell disk bead intersection with Zone C suggest that Zone D is in all like lihood an old infilled animal burrow rather than a cultural feature. Zone E is another well-defined deposit of concreted shell ( Viviparus, Pomacea and bivalve) and sand extending out of the west wall of TU12 near the southwest corner. Eventually designated Feature 17, this roughly 15-cm deep deposit may represent a preceramic pit feature, although no pit margins could be defined in the test unit profile either above or below the shell itself. And finally, within ex cavation Level K (100-110 cmbd), an isolated pocket of shell (Zone F) was identified in the northeastern corner of the test unit that contained a relative abunda nce of vertebrate fauna that included fish, bird, and mammal. Interestingly, much of the bone from Zone F consists of rabbit ( Sylvilagus spp.) appendages that re main articulated, having apparently been cemented by the calcium carbonate leached down from overlying shell depos its. The concentration of bone and shell constituting Zone F was not accompanied by any detectable change in soil color or texture. It is possible that it is simply a collapsed rabbit den. Test Unit 14 Test Unit 14 (TU14) is located two mete rs to the north of TU12. Following the general south-north trend, TU14 shares a num ber of stratigraphic similarities with TU12

PAGE 196

Silver Glen Run, Locus B (8LA1-West) 185 but also exhibits a certain amount of a dded complexity. Composite drawings and photographs of the stratigraphic profiles from all four of TU14s walls are shown in Figure 6-7 while descriptions of the major stratigraphic units are provided in Table 6-5. Summations of artifact counts for each level and zone are shown in Table 6-6. Seven distinct strata were identified in the TU14 excavation. Stratum I once again consists of a dark brown, organically enrich ed A-horizon that has been disturbed by modern plowing. It is permeated with small to medium tree roots and contains occasional whole and crushed Viviparus, as well as moderate amounts of highly fragmented St. Johns Plain and Check-Stamped ceramics, lithic debi tage, and vertebrate fauna. Shell density is highest in the north and decreases gradually toward the south. Stratum I ranges between 20 and 25 cm thic k and corresponds r oughly with excavation Levels A and B. Stratum II consists of a dense hom ogeneous layer of mostly whole Viviparus shell in dark brown loamy sand. It slopes gently upward from south to north and contains relatively small numbers of St. Johns Check-S tamped, St. Johns Plain, and plain Orange fiber-tempered ceramics in addition to sparse vertebrate fauna. Two discrete areas of charred black soil and shell encountered in Stratum II (one along the west profile and one along the east) were determined to have resulted from relatively recent burning subsequent to the stratums or iginal deposition. The charred area along the east wall penetrated deep into underlyi ng strata and likely resulted fr om a tree root that smoldered in place. At the bottom of Stratum II, near its contact with Stratum III, the Viviparus shell, while retaining its high density, transitions from primarily whole to primarily crushed. At approximately the same eleva tion, the vertebrate fauna dens ity increases significantly and the first Pomacea and bivalve specimens appear within TU14. A hafted biface and a few fiber-tempered crumb sherds were also recovered. All of this suggest s the possibility that at the time of initial Stratum II deposition, this location may ha ve experienced a relatively intense level of depositional activity and trampling after which additional, perhaps more rapid, shell deposition took pl ace that was not trampled a nd did not undergo the same level of diminution. Beginning in excavation Level E, two discrete zones of whole Viviparus shell were identified within the otherwise crushed shell floor of TU14 (Zone A). The first of these (designated Zone B) extended approximate ly 30 cm out of the west profile in the southwestern quadrant of TU14. It proved to be onl y about 15 cm thick in profile and to represent a location where Stratum II dipped slightly, penetrating underlying strata (see the west profile drawing in Figure 6-7). The second area of whole Viviparus shell (Zone C) was located along the south profile of TU 14. Unlike Zone B, Zone C descended more than 50-cm beneath the rest of Stratum II. Although not recognized as such in the field, in the south profile of the test unit (Figure 6-7), Zone C is revealed as a roughly straightsided vertical deposit of shell, alm ost certainly a result of the infilling of a pre-existing open pit feature. The apparent homogeneity of the Zone C deposit, along with its dearth of non-shell cultural materials in cluding food remains, indicate that it, along with most of

PAGE 197

186 St. Johns Archaeologi cal Field School 2007-2010: Silver Glen Run Figure 6-7. Stratigraphic drawings and photographs from profiles of TU14, 8LA1W. (Note: Photographs are not to scale.)

PAGE 198

Silver Glen Run, Locus B (8LA1-West) 187 Table 6-5. Stratigraphic Units of Test Unit 14, 8LA1W. Stratigraphic Max. Depth Munsell Unit cm BD1 cm BS2 Color Description I 28 22 7.5YR2.5/2 Very dark brown historically plowed A horizon; abundant roots; occasional whole and fragmentary Viviparus shell. II 91 86 7.5YR3/2 Very high density whole Viviparus shell in dark brown loamy sand. III 56 56 7.5YR4/3 Brown fine sand with small amount of finely crushed Viviparus shell. IV 97 92 7.5YR4/2 Moderate density crushed bivalve and Pomacea shell in brown sandy matrix; abundant vertebrate fauna. V 101 97 7.5YR4/2 Burned, crushed and concreted bivalve shell interspersed with very sparse brown fine sandy soil matrix. VI 110 110 7.5YR4/4 Virtually shell-free brown fine sand. VII 102 98 7.5YR4/4 Discrete deposit of mostly whole Viviparus shell with small amount of brown fine sand. Stratum II, are a result of rapid, large-sc ale deposition rather than the gradual accumulation of debris from everyday living. In other Archaic places in the St. Johns Valley, similar deposits of clean shell have been interpreted as instances of ritualized mounding at ceremonially significant lo cations (Aten 1999; Randall and Sassaman 2005; Russo 1994, 2004). The possibility of ritual deposition at Locus B is further discussed later in this chapter. Across much of TU14, Stratum II sits atop Stratum III, a thin horizontal layer of medium brown sand with a tr ace amount of finely crushed Viviparus shell. This stratum may constitute a buried A-horizon, indicating a substantial period of abandonment during which natural soil development was allowed to take place in the absence of human disturbance. Stratum III is thicker and better developed in the north and grows more faint toward the south. Although not easily discer nable in the south profile, Stratum III may be the surface from which the Zone C pit feature descends. Directly beneath Stratum III, Stratum IV is a 15 to 30-cm-thick layer of grayishbrown medium sand with a moderate density of fragmented Pomacea and bivalve shell. This stratum contains abundant vertebrate fa una, occasional lithic debitage, and one bone tool but no ceramics. Lining th e base of Stratum IV in seve ral spots are thin lenses of concreted, burned and crushed bivalve shell coll ectively labeled Stratum V. Stratum V in

PAGE 199

188 St. Johns Archaeologi cal Field School 2007-2010: Silver Glen Run Table 6-6. Cultural Materials Recovered from Test Unit 14, 8LA1W. Level St. Johns Plain Orange Plain Crumb Lithic Biface Lithic Flake Marine Shell (g) Modified bone Vert. Fauna (g) A 2 6 1 11.1 B1 5 47 5 0.7 55.5 C2 3 32 4 30.1 D 1 2 10 41.0 E 1 22.7 E Zone A 4 2 0.3 90.6 F Zone A 1 2 0.9 13 92.0 F Zone B 17.2 F Zone C 0.9 2.7 G 3 14 152.4 G Zone C 6.0 H 1 2 56.3 H Zone C 1 4.3 I 69.0 I Zone C 2.3 J 29.7 J Zone D 1.7 J Zone E 7.0 K 3.3 Total 11 2 100 2 20 2.8 2 694.9 1also two sand-tempered sherds 2also one St. Johns Check-Stamped sherd 3bone bead. 4bone pin TU14 is similar in elevation and stratigraphi c position to Stratum IV in TU12 and is most likely a continuation of the same preceramic living surface. In one location along the west profile, Stratum IV drops down appr oximately 20-25 cm below the rest of the stratum, revealing the presence of a small basin-shaped pit. The base of the pit is lined with the same burned and concreted bivalve co mprising Stratum V, indicating that the pit was contemporaneous with the inferred crushed shell surface. In the field, the concreted bivalve lens lining the pit was designated Zone D. The basal stratum of TU14, Stratum VI, is comprised of medium brown sand with only a trace amount of shell. The upper por tion of the stratum contains moderate amounts of vertebrate fauna and lithic debitage. Both shell and bone densities decrease with depth and eventually disappear comple tely below 100 cmbd. A discrete deposit of dense whole Viviparus encountered along the east profile at approximately 80 cmbd (Stratum VII in Figure 6-7) was designated Zone F. Zone F is similar in stratigraphic position and composition to the isolated shell de posits discovered near the base of TU12.

PAGE 200

Silver Glen Run, Locus B (8LA1-West) 189 Whether it represents a cultural feature or a na tural disturbance of some sort could not be determined. TU14 excavations were halted at 110 cmbd. Test Unit 21 Test Unit 21 (TU21) is located six meters to the north of TU14 in a slight depression of unknown origin just north of the westernmost extension of Locus Bs shell node. The northern edge of the test unit falls on the downward slope of this depression so there is an overall uphill trend from north to south. The po ssibility exists that this depression is a result of modern earth-moving activities and that the cultural deposits intersected by TU21 have been truncated to some extent. Excavated to a depth of 110 cmbd, TU21 contained seven distinct stratigraphic units. Composite drawings and photographs of the stratigraphic profile s from all four of the units walls are shown in Figure 6-8, and descriptions of the major stratigraphic units are provided in Table 6-7. Summations of artifact counts for each level and zone are shown in Table 6-8. Stratum I-A consists of the dark brown loamy A-horizon permeated by dense root mat. It varies significantly in thickne ss between 10 and 28 cm, perhaps contributing supporting evidence of modern surface modificati on in this location. Stratum I contains occasional whole Viviparus shell, a few small St. Johns ceramic sherds, and a trace amount of vertebrate fauna. In the southeastern quadrant of TU21, a small palm stump penetrated this and underlying strata and was not removed un til excavation Level D. The criteria distinguishing Stratum I-B from I-A include a reduction in the density of the root mat and an increase in the density of shell. The shell constituents of Stratum I-B consist primarily of whole and crushed Viviparus and rare crushed bivalve. Non-shell cultural materials are similar to those recovered from Stratum I-A and incl ude sparse St. Johns ceramics and trace vertebrate fauna. Togeth er, Strata I-A and I-B correspond to excavation Levels A through C and parts of Level D. Beginning in Level B and continuing th rough Level D, excavations uncovered an area in the southwestern corner of the unit consisting of gray, ashy sediment with moderate amounts of charcoal, shell and sa nd concretions, and whole and crushed Viviparus (some of which shows signs of burning). Within Level C, this anomaly spread amorphously to the east with depth and was dete rmined to be a relatively recent root burn. Beginning in Level C, this disturbance was designated Zone B while deposits in the rest of the test unit were labeled Zone A. Zone B was no longer visible by the bottom of Level D. Throughout much of TU21, Stratum I-B is unde rlain by Stratum II, a thin layer of dark brown medium sand with sparse whole and crushed Viviparus and abundant roots. Stratum II is roughly horizontal, although in the north profile, it can be seen sloping downward from west to east. It likely constitutes a burie d A-horizon corresponding to the one identified within TU14. The sharply defi ned contact between Strata I-B and II falls within excavation Levels D and E, two of th e most artifact-rich levels encountered in

PAGE 201

190 St. Johns Archaeologi cal Field School 2007-2010: Silver Glen Run TU21 excavations. Recovered artifacts include mostly incised Orange ceramics, lithic debitage, modified bone, and relatively abundant vertebrate fauna. As hypothesized for TU14, the relatively high artifact content ma y be indicative of an Orange period living surface directly atop the buried A-horizon. Beneath Stratum II, Stratum III is a highly heterogeneous layer consisting of dark grayish-brown sand and containing numerous di scontinuous pockets of shell, concretion, and mineralized roots. The shell from Stratum III is mostly whole Viviparus although crushed Viviparus and bivalve are also present in small amounts. The few cultural materials recovered from this stratum include a lithic biface, a small fragment of marine shell, and a moderate amount of vertebrate fauna. No ceramics were found below Stratum II. At approximately the same elevation as Stratum III, a discrete pocket of very dense shell was encountered along the north profile of TU 21. Eventually designated Feature 25, this pocket was first noted near th e base of excavation Level E and continues down into Level H. It cons ists exclusively of whole Viviparus shell in brown medium sand except at its base, which contains a highly concreted mixture of whole Viviparus and Pomacea In the southern half of TU21, Stra tum III is interrupted again by a thick mottled grayish brown and dark grayish brow n layer containing a relatively low density of whole and crushed Viviparus Labeled Stratum IV in Figure 6-8, this deposit is permeated by a number of light gray minera lized roots and amorphous sand concretions. Rather than a completely separate stratigraphic unit, it is likely that Stratum IV represents a largely disturbed portion of Stratum III where the large number of mineralized roots led to a distinct mottled appearance. Stratum V consists of a 15-30 cm thick discontinuous layer of fine to medium sand with a high density of mostly crushed Viviparus shell. It is interrupted in the north by Feature 25 and Stratum IV and pinches out abruptly in the south. Stratum V is penetrated by a few small live and mineralized roots and contains few artifacts, although a Marion or Newnan-type point that had been re worked into a drill, few lithic flakes, and sparse vertebrate faunal remains were recovered. Underlying Stratum V, the basal stratum within TU21 (Stratum VI) is composed of dark yellowish-brown fine sand that is la rgely shell-free but contains some isolated deposits of whole concreted bivalve and/or Viviparus shell. Artifacts in this stratum are limited to small amounts of lithic debitage and vertebrate fauna. One of the shell concentrations, occurring n ear the base of excavation Level K (100-110 cmbd), was surrounded by an faint amorphous halo of slightly darker sand, possibly reflecting elevated organic content. Two additional shell concentrations within Stratum VI were given their own unique stratigraphic designations. Stratum VII-A is a small disc rete deposit of very dark grayish-brown sand and dense shell lying at the contact between Strata III and VI in the northeastern corner of TU21. It contains primarily whole Viviparus but exhibits abundant crushed bivalve along its upper and lower margins. It appears to extend out of

PAGE 202

Silver Glen Run, Locus B (8LA1-West) 191 Figure 6-8. Stratigraphic drawings and photographs from profiles of TU21, 8LA1W. (Note: Photographs are not to scale.)

PAGE 203

192 St. Johns Archaeologi cal Field School 2007-2010: Silver Glen Run Table 6-7. Stratigraphic Units of Test Unit 21, 8LA1W. Stratigraphic Max. Depth Munsell Unit cm BD1 cm BS2 Color Description I-A 33 28 7.5YR3/1 Very dark gray medium loamy sand with abundant whole and crushed Viviparus shell; abundant roots. I-B 46 38 10YR3/2 Very dark grayish brown medium loamy sand with abundant whole and few crushed Viviparus shell. II 56 49 10YR3/3 Dark brown medium sandy loam with low density whole and crushed Viviparus shell; abundant roots. III 91 84 10YR4/2 Dark grayish brown medium sandy loam with discrete pockets of whole Viviparus and crushed bivalve shell; occasional ashy deposits; abundant roots and mineralized root casts. IV 90 90 10YR5/2 10YR4/2 Mottled grayish brown and dark grayish brown sandy loam with low density whole and crushed Viviparus shell; frequent mineralized root casts and concreted sand; occasional ashy deposits. V 100 89 10YR4/3 Brown fine to medium loamy sand with high density mostly crushed Viviparus shell; few live and mineralized roots. VI 114 114 10YR4/4 Dark yellowish brown fine loamy sand with discrete deposits of whole bivalve and Viviparus shell. VII-A 81 70 10YR3/2 Very dark grayish brown medium sandy loam with high density whole Viviparus shell throughout and abundant crushed bivalve along lower margin; common fine charcoal particles. Probable animal burrow. VII-B 113 102 10YR4/4 Dark yellowish brown medium loamy sand with common whole Viviparus shell throughout and concreted whole and crushed Viviparus Pomacea and bivalve shell along lower margin. Feature 25 and is most likely the result of an animal burrowing through the feature deposit and dragging shell into the underlying sand. Stratum VII-B is another pocket of shell near the northeast corner of the TU 21. It consists mostly of whole loose Viviparus shell but also contains frequent pockets of concreted whole and crushed Viviparus Pomacea and bivalve along its lower margin. Several of the preceramic pit features discussed below also exhibit a lining of concreted shell al ong their bases, and Stratum VII-B is likely an additiona l example of these pits th at was not recognized during excavation.

PAGE 204

Silver Glen Run, Locus B (8LA1-West) 193 Table 6-8. Cultural Materials Recovered from Test Unit 21, 8LA1W. Level St. Johns Plain Orange / T. I. Incised Orange Eroded Crumb Lithic Biface Lithic Flake Marine Shell (g) Modified Bone Vert. Fauna (g) A 4 0.9 B 3 4 4.3 C1 1 1 3 2.5 C Zone A 1 1 3 1 14.5 D Zone B 4 7 27 1 28.9 E2 4 1 2 63.4 F 1 1 0.1 66.7 G 4 29.2 H 1 2 10.7 I 1 5.8 J 1 4.0 K 2 4.0 Total 8 5 9 41 2 12 0.1 4 234.9 1plus one St. Johns Check-Stamped 2plus one Orange Plain Test Unit 19 Test Unit 19 (TU19) in another 1 x 2-m unit placed two meters to the north of TU21 in a relatively flat area just north of Locus Bs shell node. Excavations of TU19 proceeded to a depth of 130 cmbd and revealed eight distinct stratigraphic units. Composite drawings and photographs of the stra tigraphic profiles from all four of TU19s walls are shown in Figure 6-9, and descripti ons of the major stratigraphic units are provided in Table 6-9. Summations of artifact counts for each level and zone are shown in Table 6-10. In line with the previously discussed units in this transect, Stratum I of TU19 consists of a 10 to 20-cm-thick A-horizon that has been churned by hi storic plowing. It consists of dark brown loamy sand with a bundant small to medium roots and occasional whole and fragmented Viviparus. Artifact content is relatively low in Stratum I and includes small St. Johns Plain ceramic sherds, a few lithic flakes, and sparse vertebrate fauna, all of which are likely to have been displaced by the plowing. Stratum I sits atop four distinct strata in different parts of TU19, each of which is distinguished from the others based primarily on the density and condition of their shell constituents. Most clearly visible in the south and west profiles, St ratum II consists of a roughly horizontal layer of dark br own loamy sand with dense whole Viviparus shell. Stratum IV contains similar soil matrix but is distinguished from Stratum II by a higher proportion of crushed to whole Viviparus shell. It shares the same basic elevation as Stratum II and is visible primarily in the nort h and east profiles. Se parating Strata II and IV in the northwestern corn er of TU19 is a discrete pocket of very dense whole Viviparus

PAGE 205

194 St. Johns Archaeologi cal Field School 2007-2010: Silver Glen Run and bivalve that contains little to no soil matri x. The portion of this deposit visible in the west profile contains virtua lly no non-shell sediment and was designated IIIA, while the portion in the corner that contains some bur ned shell mixed with a small amount of very dark gray ashy sediment was labeled IIIB. Strata IIIA/IIIB extend approximately 25 cm out of the northwest corner of the test un it. Within excavation Levels D and E, this deposit was separated as Zone B while the rema ining general level fill was referred to as Zone A. These distinctions are reflec ted in the artifact counts in Table 6-10. All occurring at approximately the same elevations, Strata II, IIIA, IIIB, and IV correspond primarily to excavation Levels C through E and a portion of Level F. The density and condition of shell in these strata suggest rapid massive deposition rather than gradual accumulation. The fre quency of Orange fiber-tempered ceramics in these levels exceeds that found in the more southerly units while the density of other artifact types remains relatively low. Vertebrate faunal remains, in fact, are significantly less abundant in TU19 than in comparable deposits from th e test units already discussed. Also in contrast to previous units, several of th e fiber-tempered ceramics recovered from TU19 exhibit Tick Island style surface decorations featuring bold curvilinear incisions and punctations. Underlying these strata, Stratum V is a probable buried Ahorizon visible in TU19s west, south, and east profiles. Stra tum V at least roughly follows the general surface topography, sloping downward from south to north. It consists of a layer of dark brown loamy sand of variable thickness and contains only sparse whole Viviparus and virtually no other cultural materials. This stratum shares a similar stratigraphic position to presumed buried A-horizons in Test Unit s 12 and 14 and probably represents a period of site abandonment or very low-intensity use between the Late Preceramic and Early Ceramic occupations of the site. Stratum VI is a small isol ated lens of dense whole Viviparus located within the otherwise virtually shell-free matrix of St ratum V, along the southwestern margins of TU19. Located within an apparently natural pedogenic deposit, Stratum VI is best explained as resulting from an animal bu rrowing through the overlying shell midden and intruding into the soil below. Throughout much of TU19 Stratum V sits atop Stratum VII, a thin horizontal layer of dense burned and, in some places, concreted bivalve stretching across the southeastern half of the test unit. Designated Zone C during ex cavation, Stratum VII was first encountered at the base of excavation Le vel D as a discrete pocket of dense bivalve located in the southeastern corner of th e unit. As excavation proceeded, it expanded laterally across almost half of the test unit before eventually terminating near the top of Level G. The only non-shell cultural mate rials recovered include a small amount of vertebrate fauna and paleofeces. In te rms of composition, morphology, and elevation, Stratum VII seems to correspond well with th e preceramic crushed sh ell surfaces noted in TUs 12 and 14. In all three cases, the crus hed shell layers are positioned directly underneath a buried A-horizon and exhibit a complete absence of ceramics, suggesting that they may all be extensions of the same preceramic surface.

PAGE 206

Silver Glen Run, Locus B (8LA1-West) 195 Figure 6-9. Stratigraphic drawings and photographs from profiles of TU19, 8LA1W. (Note: Photographs are not to scale.)

PAGE 207

196 St. Johns Archaeologi cal Field School 2007-2010: Silver Glen Run Table 6-9. Stratigraphic Units of Test Unit 19, 8LA1W. Stratigraphic Max. Depth Munsell Unit cm BD1 cm BS2 Color Description I 26 23 7.5YR2.5/2 Very dark brown historically plowed A horizon with abundant roots and occasional whole and fragmentary Viviparus shell. II 56 53 7.5YR3/2 Dense whole Viviparus shell in a dark brown loamy sand. III-A 46 43 Extremely dense whole Viviparus shell and occasional bivalve with virtually no soil matrix. III-B 46 43 7.5YR3/1 Extremely dense whole Viviparus and bivalve in very dark gray ashy, burned soil matrix. IV 46 41 7.5YR3/2 Dense whole and crushed Viviparus shell in a dark brown loamy sand. V 88 85 7.5YR3/2 Possible buried A horizon; dark brown organically enriched sa nd with sparse whole Viviparus shell. VI 57 52 7.5YR3/1 Dense whole Viviparus shell in very dark gray sandy matrix. VII 75 77 7.5YR3/3 Dense crushed and burned bivalve concreted in some places; sparse fauna; paleofeces. VIII 127 136 7.5YR3/3 Dark brown, virtually shell-free fine sand; occasional fauna and chert flakes, decreasing with depth; large Busycon cooking vessel. The basal stratum of TU19 (Stratum VIII) is composed largely of the same fine to medium brown sand that characterizes the basal deposits of the test units already discussed. Overall, this stratum cont ains very sparse whole and crushed Viviparus and bivalve along with occasional chun ks of concreted shell near th e center of the test unit. Artifact density is slightly higher than in ot her units and includes l ithic debitage, marine shell, and infrequent vertebrate fauna. In addition to these items, a Newnan point and a burned out Busycon shell vessel were recovered fr om excavation Levels I and J respectively. Both of these are characteristic of the preceramic Mt. Taylor period. Shell and artifact density decrease wi th depth, eventually dwindling to almost nothing at the bottom of the unit (130 cmbd). Test Unit 22 Test Unit 22 (TU22), the northernmost un it in the north-south transect bisecting Locus B, is another north-south oriented 1 x 2-m unit located 2 m to the north of TU19 at the northern edge of the ridge nose just before the land begins to slope downward toward the spring run. It is offset one meter to the west of the rest of the tran sect units in order to

PAGE 208

Silver Glen Run, Locus B (8LA1-West) 197 Table 6-10. Cultural Materials Recovered from Test Unit 19, 8LA1W. Level St. Johns Plain Orange/ T. I. Incised Orange Plain Crumb Lithic Biface Lithic Flake Marine Shell (g) Vert. Fauna (g) Paleofeces (g) A 1 4 1 5.7 B 2 1 3 1 12.8 C 4 3 9 101.5 4.0 D Zone A 6 10 2 24.0 0.1 D Zone B 2.2 E Zone A 2 15.8 E Zone B 1.5 E Zone C 0.4 F Zone A 2 3 3 28.7 F Zone C 2.7 G Zone A 1 1 14.4 35.1 G Zone C 0.9 48.5 10.9 H 28.1 I 1 1 2 1.4 14.3 J 726.2 18.4 K 1 13.4 L 1 1 9.2 M 10.7 Total 3 10 8 31 3 11 844.4 275.5 11.0 avoid a large tree. Excava ted to approximately 175 cmbd, TU22 was found to contain seven distinct stratigraphic units, four of which exhibit th eir own internal divisions. Composite drawings of the stratig raphic profiles from all four of TU22s walls are shown in Figure 6-10, while descriptions of the ma jor stratigraphic units are provided in Table 6-11. Summations of artifact counts fo r each level are shown in Table 6-12. Stratum I in TU21 is the 10-20 cm thick active A-horizon covering virtually all of Locus B. It slopes gently downward from s outh to north, following the general trend of the surface topography on which it is located. Like previously discussed test units, this stratum lies entirely within the modern plow zone and has thus been disturbed to some extent. Located near the base of a large J uniper tree, the plow z one in this area is permeated by a dense mat of small to large root s. It corresponds to excavation Levels A and B and yielded very few artifacts. These are limited to a few small St. Johns Plain and Orange Plain ceramics and a trace of vertebrate fauna. Consistent with the rest of the transect, the plow zone here sits atop a thick, dense layer of shell that extends acr oss much of Locus B. In TU22, this layer (Stratum II) consists of mostly whole unconsolidated Viviparus and occasional whole Pomacea within a dark gray fine sand. In some loca tions (Stratum IIA), sh ell density is so high that minimal soil matrix can be discerned. Al ong the east and south profiles, Stratum II is

PAGE 209

198 St. Johns Archaeologi cal Field School 2007-2010: Silver Glen Run Figure 6-10. Stratigraphic drawings of profiles from TU22, 8LA1W.

PAGE 210

Silver Glen Run, Locus B (8LA1-West) 199 Table 6-11. Stratigraphic Units of Test Unit 22, 8LA1W. Stratigraphic Max. Depth Munsell Unit cm BD1 cm BS2 Color Description I 34 20 10YR3/2 Very dark grayish brown medium to fine sand with moderate whole Viviparus shell and dense root mat. II 71 74 10YR4/1 Dark gray fine sand with dense whole Viviparus and low to moderate density crushed Viviparus shell; occasional Pomacea. II-A 54 52 Whole Viviparus shell with minimal soil matrix. III 80 78 10YR2/2 10YR3/2 Fine sand with moderate density whole Viviparus shell; grades from very dark brown to very dark grayish brown toward north. IV 191 179 10YR3/3 Dark brown fine sand with low density whole Viviparus and occasional bivalve and Pomacea. IV-A 144 138 10YR3/2 Very dark grayish brown fine sand with moderate density Viviparus shell. V 104 106 10YR4/3 Brown fine sand with abundant whole and broken Viviparus shell. Undesignated pit feature. V-A 115 117 7.5YR4/6 Dark yellowish brown fine shell-free sand with probable heat oxidation. VI 200 192 10YR4/6 10YR6/6 Fine sand (sterile substrate) that grades from brownish yellow at the top to dark yellowish brown near the bottom. VII-A 191 182 10YR3/3 Dark brown fine sand with small shell concretions; sparse charcoal. VII-B 196 182 7.5YR5/6 Yellowish brown heat-oxidized sand. VII-C 198 186 10YR4/4 Dark yellowish brown fine sand; zone of leaching. interrupted in spots by thin horizontal lens es of burned and mostly crushed bivalve. These lenses provide virtually the only evidence of a depositional discontinuity within an otherwise massive deposit of gastropod shell. Stratum II deposits contain occasional St. Johns Plain and abundant Orange Plain and Incised ceramics but little else in terms of material culture. Although dominated by highly fragmented crumb sherds with indeterminate surface modifications, Orange ceramics from this stratum include both straight-line and curvilinear Ti ck Island style motifs. Other recovered materials include a single lithic tool and spar se vertebrate fauna.

PAGE 211

200 St. Johns Archaeologi cal Field School 2007-2010: Silver Glen Run Across most of the test unit, Stratum II transitions gradually to Stratum III, a stratigraphic distinction marked by the latters contrasting color and slightly lower density of shell. Stratum III consists of dark grayish brown fine sand that grades to an even darker brown in th e northern half of the test unit. It contains a moderate density of whole Viviparus shell along with a small amount of fiber-tempered ceramics, a few small fragments of marine shell a nd modified bone, and a noticeable, albeit modest, increase in vertebrate fauna density. Along the east wall of TU22, Stratum III is intersected by St ratum V, a roughly cone-shaped deposit of sand and shell that penetrates approximately 45 cm into the underlying sand. Stratum V is composed of medium brown fine sand with abundant whole and broken Viviparus. It appears to originate from the basal margin of Stratum III and probably represents an undesignated pi t feature emanating from a Late Archaic occupational surface at that elevation. Additio nal evidence for a once stable surface at the base of Stratum III occurs in the form of numerous mineralized roots visible in the TU22 profile drawings in the upper portion of Stratum IV. As discussed in Chapter 3, these mineral deposits are thought to form as large amounts of shell are dumped on a surface concealing live roots. As minera ls leach down from the overlying shell, mineralized casts of the original roots are preserved. The size and position of these root casts in Stratum IV of TU22 resembles that of the live roots located just below the contemporary surface in Stratum I. In all li kelihood, this is the same surface as that is most clearly identified at the base of Stra tum II in TU14 and inferable from the profiles of the other test units already discussed. In all of these cases, massive deposits of snail shell containing fiber-tempered pottery were dumped upon a previously existing horizontal sand substrate at a depth of betw een 50 and 75 cm below the modern surface. Underlying the Strata II and III shell deposits in TU22 is a massive ca. 120-cm thick layer of brown sand that at first glance appears to si mply represent the sterile sand substrate observed at the bottoms of all the aforementioned test units. In contrast to this massive undifferentiated layer, however, excavati on of the first few levels of the sand in TU22 revealed at least two large overlapping soil stains that were determined to be infilled pits. Although overlapping and only subtly distinct near their tops, these pits eventually diverged, exposing two discrete dark brown feat ures (Features 26 and 27) penetrating more than one meter into the yellowish-brown substrate (see Figure 6-11). An additional pit (Feature 28), which was indi scernible at its top, was identified near the bottom of TU22 based on a discrete area of oxidized, charcoal impregnated sand and concreted shell. Although impossible to determine accurate dimensions because of the restricted view offered by the 1 x 2-m test unit boundaries, Feature 26 has a diameter of at least 60-cm while Feature 27 exceeds one me ter in diameter. These pits contain little material culture aside from a small amount of Orange Plain pottery in Feature 27 and occasional vertebrate fauna. A thin lens of Viviparus shell (Stratum IV-A) runs through the center of both of these features, sugges ting that their infilling may have occurred slowly in stages rather than in a single depositional episode. Subs equent excavations of similar pits in different areas of Locus B suggest that they func tioned as large-scale

PAGE 212

Silver Glen Run, Locus B (8LA1-West) 201

PAGE 213

202 St. Johns Archaeologi cal Field School 2007-2010: Silver Glen Run Figure 6-11. Photographs of the east and south profiles of Test Unit 22 showing Features 26 and 27. roasting or steaming facilities us ed in the processing of bivalve. This interpretation helps to explain their general lack of artifact content as well as the burning evidence often exhibited at their bases. An AMS assay (s ee all radiocarbon data in Appendix B) from charcoal obtained near the bottom of Featur e 26 yielded an age estimate of 3970 40 rcybp (4520-4300 cal BP). This estimate indica tes that unlike the other test units in the same transect, TU22 contains no significant M ount Taylor component. Rather, its entire stratigraphic sequence resulted from Or ange period depositional practices. Following the general level excavations of TU22, a 50 x 50-cm column sample was removed from the east wall of the test unit (precise location of column shown in Figure 6-10) in order to collect fine-grained subsistence data. The column was located so as to capture the stratigraphic data present in the upper strata of the test unit as well as to intersect with the undesignate d pit feature (Strata V a nd VA) descending down from Stratum III in the east profile. The column was terminated at ca. 125 cmbd, as bulk samples had already been removed from the underlying pit features. Discussion of North-South Transect (TUs 12, 13, 14, 19, 21, and 22) The six test units making up the north-so uth transect at Locus B offer a 26-m cross section of this portion of 8LA1W and pr ovide a solid basis for inferring this areas depositional history. The overall stratigraphy of the transect cons ists of three major ethnostratigraphic units. Beginning at the bottom of the sequence, a substantial preceramic Mount Taylor occupation is visibl e primarily in the form of a thin crushed

PAGE 214

Silver Glen Run, Locus B (8LA1-West) 203 shell surface laid down upon a locus-wide st erile substrate of yellowish-brown sand. This crushed shell stratum is visible in all of the test units ex cluding TU13, where the shell midden thins out considerably along Lo cus Bs southern margin and TU22, where Mount Taylor deposits may have been oblit erated by subsequent Orange period pit digging activities. Mount Taylor deposits in this area include a wide variety of cultural materials generally associated with everyda y life including lithic, bone, and shell tools, abundant vertebrate fauna, paleofeces, and a numb er of small pit features suggesting that Locus B served as a place of domes tic habitation during this period. This Mount Taylor occupation was followed by a substantial, albeit indeterminate, period of abandonment or lowintensity use at Locu s B as evidenced by the development of the organi cally enriched A-horizon visible in TUs 14 and 19. The interval represented by this stratum must necessarily have entailed relative surface stability and a lack of large-scale depositional activities. Further evidence for extended surface stability during this time is provided by the numerous mineralized root casts present just below Orange Period shell deposits, which indicate that substantial vegetation was in place shortly befo re these deposits were made. Following this abandonment period, in tensive utilization of Locus B was resumed by people utilizing Orange fiber-tem pered ceramics. Along the northern margin of the excavated transe ct, this utilization involved the digging of extremely large pits, presumably for the processing of freshwater bivalve. After these pits were filled in, massive amounts of Viviparus shell were deposited over the entire area, resulting in a thick mantle of whole, often unconsolidate d, shell that shows no signs of having been trampled or intensively lived upon. In contrast to earlier Mount Tayl or period activities, those carried out by Orange people apparent ly involved a narrower range of material culture items and resulted in the depos ition of less vertebrate fauna and nonViviparus shellfish. All of these factors suggest the possibility of a more specialized use for Locus B during its Orange period occupation. Finally, a relatively thin and poorly unders tood St. Johns component exists at the top of the stratigraphic sequence revealed by th is series of test units. Unfortunately, Locus Bs extensive near-surface disturbances have largely obscured the nature of this occupation as well as its relationship to th e underlying Archaic components. There are some indications, such as the depression noted at the location of TU 21, of modern surface modifications that may have truncated St. Johns deposits, although the extent of such activities is unclear. At the very least, the en tire area appears to have been plowed and its upper 20-30 cm of sediment churned and displaced. In summary, the 1 x 2-m test units comp rising the north-south tr ansect at Locus B revealed three distinct ethnostratigraphic units that respectively correspond to the archaeologically defined Mount Taylor, Oran ge, and St. Johns periods. The Mount Taylor and Orange components in this area ap pear substantial and well-preserved while the St. Johns stratum is relatively wispy and has been disturbe d. A more nuanced understanding of these different components and their historic al relationships has been

PAGE 215

204 St. Johns Archaeologi cal Field School 2007-2010: Silver Glen Run achieved through the excavation of two exploratory test units in other parts of Locus B, as well as larger-scale and more intense block excavations. Test Unit 46 Test Unit 46 (TU46) is a 1 x 2-m test un it located approximately 13 m to the east of TU21. Excavated in 2010, TU46 was placed near Locus Bs topographic high point at the apex of the shell node in order to assess the nodes age and cultural affiliation. It was also intended to help determine what th e natural morphology of the Locus B landscape was prior to the deposition of shell during the Archaic. Composite drawings and photographs of the stratigraphic profiles from all four of TU46s walls are shown in Figure 6-12, and desc riptions of the major stratigraphic units are provided in Table 6-13. Summations of artifact counts for each level and zone are shown in Table 6-14. Based on a 1941 aerial photograph and the nearby presence of a very large and presumably old hardwood tree, the location of TU46 is likely to have escaped the historic plowing that affected the rest of Locus B. The uppermost stratum in TU46 (Stratum I) consists of a 10 to 25-cm-thick A-Horizon an d contains the highest density of St. Johns ceramics thus far encountered at Locus B. Ben eath this stratum, pottery is largely absent, with only one sherd recovered that is larger than crumb-size (>2 cm in diameter). Fibertempered pottery is scarce throughout TU46, suggesting that the Orange component may be less substantial in this part of the site. Directly beneath the active A-horizon li es a thick deposit of mostly whole Viviparus that is divided into multiple distinct strata based primarily on soil color differences and the presence of additional she llfish species. Much of this differentiation may be a result of several disturbances intr uding into the deposit from above including live tree roots, trees that burned in place, and animal burrows. In some instances, these disturbances penetrate more than a mete r into the TU46 deposits, obscuring the upper portion of the stratigraphic profil e. Burned ashy deposits in th e south half of the test unit and orange oxidized sand and sh ell along the west profile ar e perhaps the most severe examples. A shell-filled pit featur e (Feature 52) also interrupts this Viviparus-dominated stratum in the southwest corner of TU 46. Originally designated Zone A during excavation, the pit itself is high ly stratified, indi cating that it was filled in multiple episodes involving alternately the deposition of whole and broken bivalve and whole Viviparus Although containing no ceramics, this feature appears to originate high in the stratigraphic sequence, perhaps emanating from the St. Johns deposits near the modern surface. A few mineralized roots are present at the base of and di rectly underneath this Viviparus layer. A series of thin, horiz ontal layers of crushe d and burned bivalve and Pomacea are observable underneath th e thick stratum of Viviparus shell. These bivalve lenses are most clearly visible in the west and north prof iles as Strata V, VIII, and X but exist in all four profiles to some extent. When first en countered as a seemingly discrete pocket of

PAGE 216

Silver Glen Run, Locus B (8LA1-West) 205

PAGE 217

206 St. Johns Archaeologi cal Field School 2007-2010: Silver Glen Run Table 6-13. Stratigraphic Units of Test Unit 46, 8LA1W. Stratigraphic Max. Depth Munsell Unit cm BD1 cm BS2 Color Description I 36 30 10YR2/2 Very dark brown fine sand with low density whole Viviparus II-A 73 72 Dense crushed bivalve with virtually no soil matrix. II-B 81 80 Dense whole Viviparus with virtually no soil matrix. III 54 46 10YR3/6 Dark yellowish brown medium sand with moderate density whole and broken Viviparus IV 52 50 10YR6/2 Light brownish gray ashy fine sand with moderate to high density whole Viviparus and crushed bivalve. V 72 59 10YR5/3 Brown fine sand with moderate to high density crushed bivalve. VI 133 121 10YR3/3 Dark brown fine sand with very low density broken shell. VII 103 101 10YR4/3 Brown fine sand with medium whole Viviparus and broken shell. VIII 109 104 10YR4/3 Brown fine sand with moderate to high density whole and crushed bivalve and Pomacea IX 153 148 10YR3/3 Dark brown, shell-free fine sand. X 101 96 10YR4/3 Brown fine sand with moderate to high density whole and crushed bivalve and Pomacea XI 127 126 10YR3/3 Dark brown fine sand with very low density broken shell. bivalve and Pomacea in an otherwise sand test unit floor at the base of Level G, the uppermost lens was labeled Zone B. These thin layers of shell contain a variety of cultural materials, albeit all in modest am ounts, including lithic debitage and tools, marine shell, modified bone, increased vert ebrate fauna relative to overlying strata (including a sharks tooth), and paleofeces. Stratigraphically, the crushed shell lenses alternate with thin layers of dark brown sand containing very low densities of shell or other cultural materials. This entire sequence of shell-filled and shell-free strata most likely reflects a series of domestic occupati ons with intervening periods of abandonment during which natural soil development was allowed to take place. Two charcoal samples from the intervening sand strata were s ubmitted for AMS radiocarbon assays, one from Stratum VI and one from the top of Stratu m XI and returned dates of 4490 40 rcybp (5300-4970 cal BP) and 4940 40 rcybp (5740-5600 cal BP) resp ectively. If confirmed

PAGE 218

Silver Glen Run, Locus B (8LA1-West) 207 Table 6-14. Cultural Materials Recovered from Level Excavation of Test Unit 46, 8LA1W. Level St. Johns Plain Crumb Lithic Tool Unmod. Lithic Flake Misc. Rock Marine Shell (g) Modified Bone Vert. Fauna (g) A1 4 12 1 1 1 22.4 B2 1 10 1 1 1 13.8 C3 9 10 48.8 10.6 D 22 0.3 23.9 E 1 3 3.4 1 36.0 F 1 1 1 16.3 44.6 F Zone A 36.2 G4 1 2 2 103.8 G Zone A 3.0 G Zone B5 10.4 H6 3 6.8 56.47 I 2 1 80.3 J 1 2.0 152.6 K 7.8 L 1 5 1.4 M 5 1 4.2 N 7 0.9 Profile Cleanup 14.9 Total 15 58 7 27 3 77.6 3 623.2 1 plus one St. Johns Incised sherd 2 plus one Orange/Tick Island Incised sherd 3plus one sand-tempered eroded sherd 4 plus 26.1 g paleofeces 5 plus 1.7 g paleofeces 6 plus one historic ar tifact; 9.6 g paleofeces 7 includes one shark tooth by additional analysis, these dates indicate a surprisingly long, roughly 450-year interval between successive occupati ons at the base of TU46. A brown sand substrate underlies these seri al occupations at the base of TU46. The color of the substrate is sl ightly darker in this area than that revealed by other Locus B test units. Also, a higher density of heavily patinated lithic debitage was observed, possibly indicating the presence of an early Holocene component akin to those noted at 8LA1-East (Chapter 3) and also directly ac ross the spring run at 8MR123 (Randall et al. 2011). To summarize the current understanding of the stratigraphic sequence at TU46, beginning at ca. 5000 B.P. this location was th e site of apparent domestic activities involving the deposition of rela tively large meaty shellfish species (primarily bivalve and Pomacea ), various tools and manufacturing debris, vertebrate fauna, and paleofeces. Eventually, however, the area was abandoned and natural pedogenic processes gradually

PAGE 219

208 St. Johns Archaeologi cal Field School 2007-2010: Silver Glen Run covered it with a thin layer of dark organic sediment. Over the next four-plus centuries, this process of settlement and abandonment re peated itself at least two additional times leading to the stacked sequence of shell and sand visible near the base of TU46. At some point, still prior to the intr oduction of ceramic technology, the use of this location changed dramatically as everyday domestic activities were discontinued and a massive mantle of Viviparus shell was placed over the former settlement, resulting in the shell node observable at Locus B today. Simila r episodes involving the capping domestic locales at the end of their us e-lives have been noted for ot her Mount Taylor period sites in the region and may mark a ritual trans ition in the use-lives of places (Randall 2010; Sassaman 2010). The only subsequent cultu ral deposition evid ent from TU46 was carried out by people using St. Johns Plain ceramics, indicating a probable hiatus in the use of the shell node throughout th e Late Archaic Orange period. Test Unit 57 Test Unit 57 (TU57) is a 2 x 2-m test unit located approximately 34 m to the northwest of TU22, the north end of the northsouth transect discu ssed above. This puts it within what is now a cleared field that sl opes down toward the spring run to the north. TU57 was excavated in 2010 in an area in between shovel te sts 23-1 and 23-6, both of which yielded substantial quant ities of Orange fiber-tempe red ceramics. These shovel tests, along with judgmentally pl aced bucket auger tests, also indicated this as an area where shell deposits extended at least a me ter beneath the modern surface. TU57 was intended to help determine the lateral extent of the Orange component at Locus B and perhaps to locate the habitation structures a nd/or domestic debris a ssociated with this component that were conspicuously absent in the test units farther to the east. Composite drawings and phot ographs of the stratigraphi c profiles from three of TU46s walls are shown in Figure 6-13. Missi ng from this figure is the north profile, which unfortunately collapsed af ter a severe rain storm late in the field season before it was photographed or drawn (see Figure 6-14). Descriptions of the major stratigraphic units are provided in Table 6-15, and artifact counts for each level and zone are shown in Table 6-16. The clearing in which TU57 is situated is maintained by the current land manager through periodic disking. Consequently, th e upper 20-25 cm of sediment has been repeatedly churned leading to a well-define d plow-zone throughout th e unit with visible plow scars near the bottom of this stratum. This plow zone (Stratum I) was excavated largely as a single unit (Level A). Directly underneath the plow zone, a clear division became visible in the floor of the unit between a dark grayish brown zone with a very high density of mostly whole Viviparus shell (Zone A) and a dark brown loamy sand with a much lower density of primarily crushed Viviparus (Zone B). Initially, at the base of Level B (30 cmbd), Zone A was rest ricted to the southern quarter of the unit. It expanded with depth, however, reaching its maximum exte nt in Level E (50-60 cmbd) before again receding toward the south. Artifact distribut ion was fairly comparable between Zone A and Zone B with both containing Orange cer amic sherds, of which incised varieties

PAGE 220

Silver Glen Run, Locus B (8LA1-West) 209

PAGE 221

210 St. Johns Archaeologi cal Field School 2007-2010: Silver Glen Run Table 6-15. Stratigraphic Units of Test Unit 57, 8LA1W. Stratigraphic Max. Depth Munsell Unit cm BD1 cm BS2 Color Description I 47 31 10YR3/2 Plow zone. Very dark grayish brown medium sand with low density Viviparus II-A 55 39 10YR3/2 Very dark grayish brown medium sand with dense crushed Viviparus II-B 88 72 10YR3/2 Very dark grayish brown medium sand with dense whole Viviparus. II-C 94 60 10YR3/2 Very dark grayish brown medium sand with high density whole Viviparus and moderate density whole Pomacea II-D 60 58 10YR4/2 Dark grayish brown medium sand with dense whole Viviparus and crushed bivalve. II-E 71 66 10YR3/2 Very dense whole Viviparus in small amount of very dark grayish brown medium sand. III-A 75 63 10YR3/3 Dark brown medium sand. III-B 94 80 10YR3/2 Dark grayish brown medium sand. IV 103 101 10YR3/3 Dark brown medium sand with high density whole Viviparus V-A 140 138 10YR4/3 Brown medium sand with low to moderate density whole Viviparus. V-B 124 120 10YR4/3 Brown medium to fine sand with no shell surrounding pit feature. V-C 140 128 10YR4/3 Brown medium to fine shell-free sand. VI 155 154 10YR6/6 Yellowish-brown fine sand. outnumber the plain ones, and occasional vertebra te fauna. In profile, Zone A is visible as a thick dense mantle of she ll (Strata IIA-IIE) that, like th e upper shell strata in the 1 x 2-m units to the southwest, shows little differentiation from bo ttom to top, perhaps indicating its complete deposition in one or a few episodes over a short period of time. The thick stratum of dense Viviparus constituting Zone A sits atop a buried Ahorizon that can be observed as Stratum IIIB in the profile illustrations. This 5 to 15-cmthick layer runs throughout mu ch of the test unit at appr oximately 70-80 cm below the modern surface. Interestingly, as this ancient surface was approached during excavation, the fiber-tempered ceramic assemblage shifted from predominantly incised to

PAGE 222

Silver Glen Run, Locus B (8LA1-West) 211 Figure 6-14. North profile of Test Unit 57 following the collapse of its rain-saturated basal sands. predominantly plain, indicating two distinct depositional patterns within the Orange component in TU57. As excavations proceeded beneath the bur ied surface, the densest shell deposits remained confined to the southern half of the test unit. Initially, this shell appeared to constitute a broad continuous stratum cutting across the entire test unit, but at ca. 100 cmbd, it became more discrete, re vealing a large shell-filled pit. Originally considered two separate pits (Features 54 and 55) due to the angle at which the shell intersected the corner of the test unit, it was later determined to be one large pit. This pit is similar in scale to those observed in TU22, although it is somewhat shallower and more basinshaped. It also lacks clear evidence for ther mal alteration at its base, perhaps indicating an alternative function from its counterparts to the southwest. Detailed compositional data from the pit fill is not yet available but the fill includes dense whole and crushed Viviparus, moderate density of vertebrate fauna, a nd occasional lithic flakes. A complete bone awl was also recovered from Featur e 54/55. A zone of dark brown organic sediment with a lower density of shell (Strata VA and VB) surrounds the shell-dense portion of the feature. It is unclear whethe r this reflects the stratigraphically distinct basal portions of the pit or simply a zone of organic leaching into the underlying substrate. An AMS radiocarbon assay obtained from charcoal recovered from the base of Feature 54/55 returned an age estimate of 3690 40 rcybp (4140-3900 cal BP), situating it at least two hundred years later than Feature 26 from TU22.

PAGE 223

212 St. Johns Archaeologi cal Field School 2007-2010: Silver Glen Run

PAGE 224

Silver Glen Run, Locus B (8LA1-West) 213 Before its collapse, the north profile of TU 57 contrasted significantly with the rest of the test unit in its overal l dearth of shell. The buried A-horizon (Stratum IIIB) was clearly visible at the same elevation as the rest of the un it but instead of supporting an overlying mantle of dense shell, it was topped by a thick undifferentiated layer of dark brown sand. In the surviving profiles, this sand layer is visible as only a small lens (Stratum IIIA) in the east wall that underlies the shell stratum as it pinches out in the northeast corner of the unit. This sand was emplaced directly onto the ancient surface and must have existed as subtle topographic rise at the time that the mantle of shell (Strata IIA-IIE) was deposited. That shell thus appears to have leveled out what previously would have been a rough and uneven surface. Test Unit 4 Test Unit 4 (TU4) is an eas t-west oriented 1 x 2-m test unit located approximately five meters south (upslope) of shovel test pi t 22-1 and ca. ten meters to the west of the north-south transect described a bove. It was sited in an effo rt to investigate an Orange period occupation, possibly of circular orientation judging from the distribution of shell and fiber-tempered pottery in shovel tests. Although the first test unit excavated at Locus B in 2007, TU4 is discussed last among the e xploratory units because its stratigraphic complexity made explanation difficult until additional units were dug to provide some interpretive context. Excavation of TU4 was supervised by Neill Wallis and this discussion is adapted larg ely from his field notes. Composite drawings and phot ographs of the stratigraphi c profiles from three of TU46s walls are shown in Figures 6-15 and 6-16. Descriptions of the major stratigraphic units are provided in Table 6-17, and artifact counts for each level and zone are shown in Table 6-18. Excavation methods in TU4 differed sli ghtly from those described at the beginning of this chapter. Like the other test units, TU4 was excavated primarily in arbitrary 10-cm levels. As excavation pro ceeded, however, dense unconsolidated shell in the walls of the unit began to dislodge and in order to avoid a complete collapse, the walls were allowed to slope further and further in toward the center of the unit until it took on a bathtub shape. Subsequently, this bathtub balk was removed in natural stratigraphic layers. Conseque ntly, artifact counts (Table 6-18) are provided for both arbitrary excavation levels a nd natural stratigraphic units. TU4 exhibits a 20 to 30-cm-thick dark brown, organic plow zone containing a few displaced artifacts and occasional Viviparus shell, the density of which increases with depth. By the middle of excavation Level C (ca. 25 cmbs) this plow zone (Stratum I) begins to give way to the same thick stratum of dense whole Viviparus (Strata IIA-B, IV, and IVA-B in the profile drawings; Zones A and B during excavation) that has been observed within all Locus B test units so fa r discussed. The shell density in TU4, however, exceeds that encountered in most of the other units, leadin g to the instability issues noted above. In multiple locations along the profiles (especia lly in the east and

PAGE 225

214 St. Johns Archaeologi cal Field School 2007-2010: Silver Glen Run Figure 6-15. Stratigraphic drawings and photographs from west and north profiles of TU19, 8LA1W. (Note: Photographs are not to scale.) south) there are large shell pockets that cont ain virtually no soil matrix. Thin horizontal lenses of crushed and sometimes burned bivalv e (Strata IIIA-C; Zone C) cut through this thick Viviparus stratum in at least three distinct elevations. One of these lenses extended across the entire floor of the test unit at th e base of excavation Level E (50 cmbd) and contained some of the densest concentrations of Orange ceramic sherds (including both plain and incised varieties) an d vertebrate fauna within TU4. Together, these upper shell strata at TU4 may show ev idence of a sequence of pr epared and lived upon surfaces consisting of layers of whole Viviparus laid down to form a le vel and regular foundation upon which burned and crushed bivalve shell wa s then placed. At least three of these surfaces are discernible in the TU4 profiles, while another may have sat atop Stratum II,

PAGE 226

Silver Glen Run, Locus B (8LA1-West) 215 Figure 6-16. Stratigraphic drawings and phot ographs from east and south profiles of TU19, 8LA1W. (Note: Photogr aphs are not to scale.) but was perhaps obliterated by plowing. Unde rneath the lowermost crushed bivalve lens the shell density decreases substantially. Strata VA and VB constitute a thick massive layer of moderately dense whole Viviparus and sand distinguished only by the darker color of the latter. This layer (designated Zone D during excavation) contains abundant Orange Plain ceramics, occasional flakes and ve rtebrate fauna, and frequent charcoal. It

PAGE 227

216 St. Johns Archaeologi cal Field School 2007-2010: Silver Glen Run Table 6-17. Stratigraphic Units of Test Unit 4, 8LA1W. Stratigraphic Max. Depth Munsell Unit cm BD1 cm BS2 Color Description I 30 25 10YR3/3 Historically plowed A horizon with abundant roots and moderately dense whole Viviparus shell. II-A 52 48 Whole unconsolidated Viviparus shell with no soil matrix. Trace amounts of vertebrate fauna and charcoal. II-B 45 40 10YR3/1 High density Viviparus in very dark gray loamy fine sand. III-A 62 57 10YR5/2 Crushed and burned bivalve shell (with trace amounts of Viviparus and Pomacea ) interspersed with grayish brown fine ashy sand. III-B 60 55 10YR3/2 Identical to St ratum IIIB but with very dark grayish brown sand and slightly less dense shell. III-C 67 60 Identical to Strata IIIA and IIIB but with absolutely no soil matrix. IV 70 64 Whole unconsolidated Viviparus shell with small amounts of Pomacea and bivalve and no soil matrix. Trace amounts of vertebrate fauna and abundant charcoal. V-A 85 83 10YR4/3 Moderately dense whole Viviparus shell in brown fine sandy matrix. Abundant vertebrate fauna. V-B 106 106 10YR3/2 Moderately dense whole Viviparus shell in very dark grayish brown sandy loam; abundant vertebrate fauna. VI 131 126 10YR3/4 Dark yellowish brown fine sand with scant whole Viviparus shell; sparse vertebrate fauna. VII 165 165 10YR4/4 Dark yellowish brown fine sand with no shell; trace vertebrate fauna. sits atop a dark yellowish brown sandy substrat e (Zone E) that lightens slightly in color near the base of the test unit. In the southeastern half of the test unit, the basal sands are interrupted by Features 15 and 16, visible as a 20-30 cm thick deposit of concreted shell, ash, and charcoal in the south and east profiles. Originally designated Zone F, these features covered over ha lf of the test unit at the base of excavation Level L (120 cmbd). They consist primarily of burned whole and crushed Viviparus but also contain pockets of burned whole bivalve, Pomacea and charcoal. Portions of the features are highly concreted and had to be excavated with a pi ck. Both contain Orange Plain ceramics but no other artifacts were recove red. No obvious division ex ists between Features 15 and

PAGE 228

Silver Glen Run, Locus B (8LA1-West) 217 Table 6-18. Cultural Materials Recovered from Test Unit 4, 8LA1W. Level St. Johns Plain Orange/ T. I. Incised Orange Plain Crumb Lithic Flake Marine Shell (g) Vert. Fauna (g) Botanicals (g) Other 80CM 3 A 1 0.1 B 1 2 7 1 0.9 2.6 21 C (Zone A) 1 0.7 C (Zone B) 2 12 2 5.9 D (Zone A) 8 0.7 0.5 D (Zone B) 1 1.7 0.4 D (Zone C) 15 1.8 E (Zone A) 10 3.8 E (Zone C) 1 16 1 9.6 14.4 3.3 0.62 F (Zone C) 1 4 4.7 G (Zone C) 4 4.3 G (Zone D) 1 11 11 28.5 H (Zone D) 14 35 2 15.2 H 3 I (Zone D) 12 10 3.9 1.2 I (Zone E) 6 7 7.1 J (Str. VI) 1 17 8.3 11.7 J (Zone F) 2 K (Str. VI) 2 7 14.9 2.4 K (Str. VII) 2 24 3 33.1 1.5 L (Str. VII) 5 18 3 19.3 1.4 13 L (Zone F) 5 M (Str VIII) 3.1 M (Zone F) 8 21 7.0 1.0 N 2 0.2 Profile Clean 4 14 8.1 Str. I 1.6 Str. IIA 5.1 Str. IIB 0.5 Str. III 3 24 1.6 1.4 Str. IV+ 1 6.0 2.9 13 Str. V 10 15 1 8.9 Str. VI 14 12 2 24.5 Total 3 3 109 295 15 10.5 237.6 27.7 1one historic lead bullet; one St. Johns Check Stamped sherd 2paleofeces (g) 3one modified bone

PAGE 229

218 St. Johns Archaeologi cal Field School 2007-2010: Silver Glen Run 16, although the shell in Feature 16 in the sout heastern corner of the test unit does drop down further into the underlyi ng sand than that from Feat ure 15. Nevertheless, it is possible that these are actually di stinct portions of the same ma ssive feature. No vertical margins could be discerned within the overlying sand but it is likely that the shell and ash deposits represent the basal fill of one or more large roasting pits, similar to those Features 26 and 27 at the bo ttom of TU22. An AMS radio carbon assay from charcoal recovered from Feature 15 returned an age estimate of 3820 40 rcybp (4410-4100 cal BP), which overlaps temporally with the 2sigma calibrated range of the assay from Feature 26. Summary of Exploratory Test Units The excavation of nine individual disc ontinuous test units at 8LA1Ws Locus B between 2007 and 2010 yielded a tremendous amount of information regarding the extent, condition, and structure of the sites archaeological deposits. These test units first expand upon the auger survey in demonstrating the presence of extensiv e shell deposits at Locus B that extend at least 35 m from north to south and 50-m from east to west. Within this 1750-m2 area, deposits vary greatly with regard to both th ickness and depth, with the thickest, most complex stratigra phy occurring along the northern margin and then thinning gradually to the south. Asid e from a consistent 20-30-cm plow zone and various discrete plan t and animal disturbances, these deposits were shown to have escaped the mining activities that impacted othe r areas of the Silver Glen Run site and remain relatively intact. Test units revealed three distinct et hnostratigraphic units w ithin Locus B that together span thousands of years of the sites history. Beginning as earl y as 5700 cal BP, Mount Taylor people began using Locus B in a manner that involved the deposition of freshwater shellfish (primarily bivalve and Pomacea ), a variety of stone, bone, and marine shell tools and debitage, and modera te amounts of vertebrate fauna. Formal floors or living surfaces were constructed du ring this period through the deposition of bivalve that was either processed beforehand through burning and crushing or was altered in this way through use. Sm all cylindrical and basin-shap ed pits were dug down from these surfaces presumably fo r either storage or cooking, although there is little evidence for thermal alteration. All of these material culture objects and features are what would be expected to result from routine everyday living at this time, suggesting that the Mount Taylor component at Locus B reflects a sett lement or village occupation. At least one area of Locus B, as revealed by TU46, was occupied repeatedly in a similar manner following periods of abandonment as indicate d by a series of thin horizontal bivalve strata with intervening layers of sand. This eventually resulted in the construction of Locus Bs raised shell node, which was subsequently capped with a layer of Viviparus by Mt. Taylor or later people. The test units tentatively suggest that this Mt. Taylor component is restricted to the southeastern portion of Locus B as it was not observed in TU22 at the north end of the north-south tr ansect or in TUs 4 and 57 to the west. Eventually Locus B was abandoned or ut ilized sparingly ove r an extended enough period that a well-developed A-horizon was allo wed to form with associated large tree

PAGE 230

Silver Glen Run, Locus B (8LA1-West) 219 roots. Subsequently, perhaps as early as 4500 cal BP, this location was utilized in a wholly different manner by Orange period people using the regions earl iest pottery type. Enormous pits were dug deep into the underlying sand and sh ellfish was processed at an unprecedented scale. At the same time, fewe r and less diverse cultural materials were deposited with the shell, po ssibly suggesting a more specia lized, non-domestic function. Later in the Orange period, use of the site changed again, as the entire area was capped with a massive layer of whole Viviparus shell. In most places this capping appears to have occurred quickly as a si ngle act while in TU4 it was adde d in stages as indicated by the presence of horizontal bi valve lenses interrupting the Vivivparus stratum at multiple points. The architects of the shell cap util ized a new type of fiber-tempered pottery (Tick-Island Incised) featuring bold curvi linear incisions and punctations. It is unknown at present whether they were the same or a di fferent people than those responsible for the massive underlying pits. Although spatially ov erlapping somewhat with earlier Mount Taylor materials, Orange deposits at Locus B appear to be centered to the west of the shell node on the down-slope portion of the landform. And finally, following the Orange occ upation, Locus B was utilized to some extent by St. Johns people as evidenced by the smattering of St. Johns ceramics found throughout the sites near-surface deposits. Unfortunately, the St. Johns component lies completely within this areas plow zone and has been heavily disturbed, making any inferences regarding the nature of the St. J ohns occupation or its relationship to earlier Orange materials difficult at best. The individual test units also revealed some significant stratigraphic patterns that proved useful for interpreting the results of s ubsequent excavations. First, they provide additional support for the notion that the presen ce and distribution of the mineralized root casts that form underneath shell deposits (see also Chapter 3) can be used to infer periods of relative inactivity and approximate the elevat ions of paleosurfaces. At Locus B, root casts are generally concentrated within or just below buried A-horiz ons, thus replicating the pattern exhibited by live r oots in relation to the modern surface. In addition, these test units demonstrate that living surfaces during both the Mount Taylor and Orange periods are frequently lined with thin la yers of burned and crushed bivalve that sometimes contain concentrations artifacts and vertebrate fauna. BLOCK EXCAVATIONS Each of the individual test units de scribed above provides an important window into the basic structure of Locus B deposits at a particular location and together, they begin to offer some clues as to the distribut ion of particular stra tigraphic units and the historical relationships among them. These sm all-scale units are clearly limited, however, in the insight they convey into the contemporary processes operating within any one temporal-stratigraphic unit because of the rela tively restricted horizontal perspective that they entail. To address topi cs such as community organization, architectural features, and the horizontal patterning of cultural ma terials, it was necessary to broaden the perspective by opening up larger continuous areas for excavation. Consequently,

PAGE 231

220 St. Johns Archaeologi cal Field School 2007-2010: Silver Glen Run between 2009 and 2010 two block excavations were conducted near the presumed center of Locus Bs archaeological deposits. 2009 Block Excavation In 2009, a 4 x 4-m block was excavated in the area between the north-south transect of test units and TU4 (see Figure 6-17). It was originally laid out as a 3 x 4-m unit with its southeast corner positioned 4 m to the west of the southwest corner of TU21 but was expanded to 4 x 4 m before excavati on began. The block was intended to locate Orange period domestic featur es and architectural remain s and to document spatial patterning in features and artifacts at a fi ner scale than is possible utilizing smaller excavation units. This location was chosen because of the particularly dense Orange shell deposits and pit features indica ted by the surrounding 1 x 2-m test units. Excavation Methods. With these goals in mind, excavation methods employed in the block deviated somewhat from those described above and were tailored toward detailed documentation of spatial data. Initial ly, the block was divided into sixteen 1 x 1m units (Test Units 23-38) which were excavat ed and documented sepa rately, although at the same rate. It was quickly realized, t hough, that digging in th ese small units, while providing a fine level of sp atial control, actually obs cured the broad horizontal perspective that constituted the ultimate goal of the block excavation. Ultimately, near the base of the plow zone, the block was di vided into four 2-x-2 m units (Test Units 3942) in order to maintain adequate spatial c ontrol while still allowing for the recognition of horizontal patterning. Test unit numbers both be fore and after the sw itch from 1 x 1-m to 2 x 2-m test units are shown in Figure 6-18. The four individual units were generally excavated separately in arbitrary 10 cm levels; however, an effort was made to take them all down at approximately the same rate so that when stratigraphic zones or large f eatures were uncovered, th ey could be followed and documented across test unit boundaries. To this end, level elevations for all four units were measured from the same datum locat ed at the southeastern corner of the block with an absolute elevation of 15.81 m based on the site datum. Because the upper 30 cm of the block was lumped together as the p low zone, Level A refers to the absolute elevations 15.51-15.41 m, Level B to 15.41-15.31 m, and so on. As a result of the sloping surface on which the block was situated Levels A through E were excavated as partial levels until a level plane was reach ed across all four test units. All artifacts larger than two centimeters in diameter (excluding unmodified vertebrate fauna) that were discovered in situ were plotted with a Nikon DTM-310 Total Station, assigned a unique piece-plot number, and bagged individually. As with the exploratory units discussed in the previous section, all non-feature fill was dry sieved through 1/4-inch screen and all cultural mate rials excluding freshwater shells were collected and bagged according to provenience. Features were mapped in plan view, bisected vertically, and then drawn in profile. When possible, one half of the feature fill was removed for 1/8-inch water screening wh ile the remaining half was removed as a bulk flotation sample. For some extremely larg e features, it was not practical to collect

PAGE 232

Silver Glen Run, Locus B (8LA1-West) 221 Figure 6-17. 2009 field crew excavating a 4-x-4-m block at Locus B, 8LA1W. Figure 6-18. Test unit numbers of 1-x-1-m units and 2-x-2-m units from 2009 block at Locus B, 8LA1W.

PAGE 233

222 St. Johns Archaeologi cal Field School 2007-2010: Silver Glen Run them in their entirety. In these cases fine screen and bulk flotation samples were collected from multiple locations within the f eature while the rest of the fill was 1/4-inch screened. 2009 Block Stratigraphy. Nineteen distinct stratigraphic units were identified in the 2009 block excavation. Composite drawin gs and photographs of the stratigraphic profiles from the blocks walls are shown in Figures 6-19 through 6-25 Descriptions of the major stratigraphic units are provided in Table 6-19, and arti fact counts for each level and zone are shown in Table 6-20. For the most part, the uppermost stra tum encountered within the 2009 block (Stratum I) is consistent with that observed in the individual explorat ory units. It consists of a 20 to 30-cm thick layer very dark grayish brown organically enriched sand with a moderate density of whole Viviparus and is permeated by dense tree roots. Small, highly fragmented St. Johns ceramic sherds are th e most common artifact type, although Orange ceramics, lithic tools and debi tage, and vertebrate faunal re mains were also observed. Near the base of Stratum I, linear stringers of whole and crushed shell oriented in a northeast-southwest fashion (shown in Figure 6-26) were identified as plow scars, confirming suspicions based on earlier excava tions that Locus B was cleared and plowed in the relatively recent past. The contact between Stratum I and the int act shell deposits below is irregular in some locations, probably owing to the scraping of the top of the midden by the metal plow. It is unclear how much shell may have been removed or displaced by these activities. Regardless, Stratum I now s its atop a variably th ick (ranging primarily between ca. 30 and 50 cm) deposit of extremely dense mostly whole Viviparus with occasional bivalve and Pomacea labeled Stratum II (also IIA and IIB). This shell layer is thickest in the north half of the block and thins considerably near the southeast corner. In one discrete location visible in the west profile, Stratum II abruptly drops down into underlying strata, apparently infil ling a pit that was ope n at the time of its deposition. Shell density in Stratum II varies slightly, ranging from high (Zone A) to extremely high (Zone C), often with little to no interven ing soil matrix. As in TU4, this dense, unconsolidated shell resulted in instability issues in some locations as whole Viviparus shell would occasionally pour out of the profile if even lightly contacted. Again, shell of this density and condition s uggests rapid large-s cale deposition rather than the gradual accumulation of domestic food remains that w ould have allowed for more substantial soil development and resulted in crushed tightly packed shell deposits. The highly variable basal margin of Stratum II indicates that these shell deposits filled in what had been a rough irregular surface at the time of deposition. Stratum II, however, was probably not de posited in a single episode as evidenced by the fact that this thick layer of Viviparus, as seen in TU4, is interrupted at different levels by thin horizontal lenses of burne d and crushed bivalve (Zone B) that may represent a series of prepared shell surfaces. These bivalve lenses are clearly visible in the north and east profiles of the block but we re not observed in the west and south.

PAGE 234

Silver Glen Run, Locus B (8LA1-West) 223 Figure 6-19. Stratigraphic drawing of north profile from 2009 block (TUs 39-42), 8LA1W.

PAGE 235

224 St. Johns Archaeologi cal Field School 2007-2010: Silver Glen Run Figure 6-20. Composite photograph of north profile from 2009 block (TUs 39-42), 8LA1W.

PAGE 236

Silver Glen Run, Locus B (8LA1-West) 225 Figure 6-21. Stratigraphic drawing of west profile from 2009 block (TUs 39-42), 8LA1W.

PAGE 237

226 St. Johns Archaeologi cal Field School 2007-2010: Silver Glen Run Figure 6-22. Composite photograph of west profile from 2009 block (TUs 39-42), 8LA1W.

PAGE 238

Silver Glen Run, Locus B (8LA1-West) 227 Figure 6-23. Stratigraphic drawing of east profile from 2009 block (TUs 39-42), 8LA1W.

PAGE 239

228 St. Johns Archaeologi cal Field School 2007-2010: Silver Glen Run Figure 6-24. Composite photograph of east profile from 2009 block (TUs 39-42), 8LA1W.

PAGE 240

Silver Glen Run, Locus B (8LA1-West) 229 Fi g ure 6-25. Strati g ra p hic drawin g and com p osite p hoto g ra p h of south p rofile from 2009 block ( TUs 39-42 ) 8LA1W.

PAGE 241

230 St. Johns Archaeologi cal Field School 2007-2010: Silver Glen Run Table 6-19. Stratigraphic Units of 2009 Block (TUs 39-42), 8LA1W. Stratigraphic Max. Depth Munsell Unit cm BD1 cm BS2 Color Description I 74 23 10YR3/2 Plow zone; very dark grayish brown loamy fine sand with moderate to abundant whole Viviparus shell; abundant live roots II 111 71 Whole Viviparus shell with minimal soil matrix; traces of crushed shell throughout; thin lenses of crushed bivalve II-A 144 104 10YR4/2 Dark grayish brown loamy fine sand with abundant whole Viviparus shell and occasional Pomacea shell II-B 125 81 10YR3/2 Very dark grayish brown loamy fine sand with whole Viviparus shell; infrequent fauna. III 141 96 10YR3/3 Buried A-horizon; dark brown fine to medium sand with only trace amount of shell; grades to lighter brown color with depth. IV 141 94 10YR4/2 Dark grayish brown medium sand with abundant whole Viviparus shell and moderate density crushed Viviparus and bivalve. IV-A 179 132 10YR4/2 Dark grayish brown concreted medium sand with abundant whole Viviparus and bivalve shell and moderate density crushed Viviparus V 187 141 10YR4/3 Brown fine sand with moderate to low frequency of whole Viviparus and bivalve. VI 201 158 10YR4/3 Brown fine sand with low frequency of whole Viviparus and bivalve. VII 190 148 Concreted Viviparus (whole) and bivalve (whole and crushed) shell. VII-A 217 172 Concreted Viviparus (whole) bivalve (whole and crushed) shell; higher density bivalve than stratum VII. VIII 222 176 10YR4/4 Dark yellowish brown fine to medium sand with no shell. VIII-A 218 176 10YR4/2 Dark grayish brown medium sand with no shell. IX 231 187 10YR6/4 Light yellowish brown fine to medium sand with no shell. X 221 181 7.5YR5/8 Strong brown heat oxidized sand with high frequency of charcoal. (continued on next page)

PAGE 242

Silver Glen Run, Locus B (8LA1-West) 231 Table 6-19. (continued) Stratigraphic Max. Depth Munsell Unit cm BD1 cm BS2 Color Description XI 144 109 10YR2/2 Very dark brown medium sand with moderate density whole and crushed Viviparus and small amount of charcoal. XII 190 148 10YR3/2 Very dark grayish brown fine sand with abundant whole Viviparus and moderate density whole bivalves. Heterogeneous in color and content. XIII 144 96 10YR4/2 Dark grayish brown fine sand with abundant compacted whole Viviparus and rare crushed bivalve, Pomacea Occasional charcoal. XIV 135 99 10YR4/2 Dark grayish brown fine to medium sand with abundant whole and moderate density crushed Viviparus XV 108 85 10YR5/310YR3/3 Brown to dark brown medium sand with moderate frequency whole Viviparus and abundant mineralized ro ot casts throughout. XV-A 116 92 10YR4/3 Brown fine to medium sand with abundant mineralized root casts throughout. XVI 83 40 10YR5/3 Brown mottled crushed shell with oxidized sand and burned shell. Virtually no soil matrix. XVII 118 103 10YR3/3 Dark brown medium sand with whole and crushed Viviparus and bivalve near base. XVIII 70 54 10YR4/1 Dark gray medium sand with moderate amount finely crushed shell and whole Viviparus XIX 119 98 10YR3/3 Dark brown medium sand with moderate amount crushed shell and whole Viviparus and Pomacea. Whatever their origin, th ere is little indication, in terms of either hab itation debris or shell diminution, that they reflect le ngthy domestic occupations. In addition, at leas t three discrete thermal featur es were encountered in the excavation of Stratum II. All were ovoid in shape and consisted of a central pocket of gray ash and burned shell surrounded first by burned and crushed Viviparus and bivalve and then by orange oxidized shell and sand. Initially these were de signated as features (Features 29 and 30) and were interpreted as possible domestic cooking hearths. As excavations progressed, however, the burne d deposits began to taper and turn in unpredictable ways suggesting that they were in fact tree roots that penetrated into shell deposits and eventually burned in place.

PAGE 243

232 St. Johns Archaeologi cal Field School 2007-2010: Silver Glen Run

PAGE 244

Silver Glen Run, Locus B (8LA1-West) 233

PAGE 245

234 St. Johns Archaeologi cal Field School 2007-2010: Silver Glen Run t

PAGE 246

Silver Glen Run, Locus B (8LA1-West) 235 Figure 6-26. Plow scars in south half of 2009 block at approximately 20 cmbd. Artifacts are generally sparse within Stratum II and include Orange fibertempered ceramics, occasional lithic flakes, a nd a very low frequency of vertebrate fauna. The Orange ceramic assemblage includes both plain and incised vari eties, including relatively rare Tick Island Incised examples. The latter were found primarily within the densest shell in the northern half of the block. The thick mantle of shell constituting Stratum II sits atop a probable buried Ahorizon (Stratum III) suggesting some period of disuse at Locus B before the shell was deposited. This natural soil horizon consists of a ca. 10-15 cm thick layer of dark brown sand with only a trace amount of shell and is most clearly visible in the blocks north profile. It is roughly horizontal but trends gently upward to ward the northeast, following the slope of the modern surface in this area. Beneath the thick Viviparus layer and buried A-horizon, it becomes more difficult to discuss the block as a whole, as at this point its north a nd south halves begin to diverge stratigraphically. In the north, just below Stratum III the basal sand (Stratum IX) is intersected by a number of ma ssive pit features similar to those found in TUs 4 and 22 (see Figure 6-27). As in TU22, these pits are broadest and often overl ap at their tops so that virtually the entire bottom 1.5 m of cultural deposits in the no rthern half of the excavation block consist complete ly of pit fill. Although only five of these large-scale pits could be defined well e nough to receive feature dis tinctions, several additional examples were undoubtedly excavated. Probable pit fill that could not be confidently attributed to any specific feature was excavated as either Zone D if it consisted of shell

PAGE 247

236 St. Johns Archaeologi cal Field School 2007-2010: Silver Glen Run Figure 6-27. Plan drawing and photograph of the b ases of pit features in the north half of the 2009 block, 8LA1W (Note: photograph not to scale). free sand or Zone F if it contained both sand and shell. With one notable exception, the features that could be delineated contain mostly dark gray-brown sand with varying densities of Viviparus shell and small to moderate amounts of plain fiber-tempered pottery. They all exhibit evidence of thermal activities, with most displaying the bright red oxidized sand and charcoal deposits at th eir bases noted above for TU22. In addition, at least three of the pits contain substantial amounts of burned and concreted bivalve shell, providing strong support for their in terpretation as mussel steaming facilities. Charcoal samples recovered from two of the pits containing Orange Plain pottery (Features 33 and 36) returned respective AMS radiocarbon assays of 3730 40 rcybp (4230-3980 cal BP) and 3590 40 rcybp ( 3980-3830 cal BP).

PAGE 248

Silver Glen Run, Locus B (8LA1-West) 237 One of the pit features, Feature 38 is differe nt from the others in that it appears to have been filled in at least three distinct stages. The base of this feature is composed of fine brown sand with no shell. Above this is a dense, highly c oncreted layer of whole and crushed bivalve and Viviparous shell, followed by a less concreted layer with more sand relative to shell. Feat ure 38 is also the only pit in the 2009 block to contain decorated ceramic sherds, all of which exhibit Tick Island style designs. This pit appears to be slightly younger in age than the others because it originates from a higher surface and actually cuts into two preexisting p its. Two AMS assays of 3590 40 rcybp (39803830 cal BP) and 3670 40 rcybp (4140-3890 cal BP ) situate Feature 38, along with the Tick Island style pottery that it contains, late in the Orange period, near the end of Locus Bs Late Archaic occupation. I ndividual pits are discussed in more detail in the section below on features. In the south half of the bl ock, instead of a covering a series of large pits, Stratum II, the thick Viviparus layer, obscures a low-lying, fl at-topped mound of emplaced sand labeled Strata XV and XV-A (Figure 6-28). This sand feature is approximately 40 cm tall at its highest exposed point in the southeast corner of the block and can be seen to slope gradually down in both th e south and east profiles until it reaches the underlying basal sand (Stratum IX). Although its precise dimensions are unknown, prior to excavation the sand extended almost 1.5 m out from the southeast corner. In all likelihood, the sand was piled up as the pits evid ent in the north half of the block were dug, although what, if any, function this feature may have served is uncle ar at this time. Virtually no cultural materials were f ound either on or within the emplaced sand and no features were discove red that would suggest it se rved as an architectural foundation. There were, however, two small, shallow basin shaped pits discovered along the margins of the emplaced sand. One of these, Feature 35, ex tended out of the east wall of the block near the northe rn edge of the sand and cont ained mostly whole unopened bivalve and whole Pomacea while the other, Feature 37, was located approximately 2 m to the west and contained exclusively whole unopened bivalve. These pits emanated from the same surface as the larger steaming pits in the northern half of the block and appear to have been placed in relation to the emplaced sand feature. Their relatively small size, lack of thermal alteration, and the whole pair ed bivalves they contained, suggest that these features may reflect either short-term st orage pits that were abandoned or forgotten or perhaps even intentional votive deposits or offerings a ssociated with the shellfish production process. An AMS radiocar bon assay of 3640 40 rcybp (4080-3850 cal BP) from Feature 37 indicates general contempor aneity with the large steaming pits and possibly a coordinating function. The emplaced sand sits atop the yellowish br own layer of shell-free sand (Stratum IX) that constitutes the ster ile basal stratum throughout the excavation block. In the south half of the block this stratum was cons istently encountered at approximately 70 to 90 cmbd. In the north half, the upper portion of the basal sand was largely obliterated when the large steaming pits were dug. C onsequently, at some spots along the northern profiles it is present only below these pits at depths of 150 to 200 cmbd.

PAGE 249

238 St. Johns Archaeologi cal Field School 2007-2010: Silver Glen Run Figure 6-28. South profile of 2009 bloc k showing location of emplaced sand. 2010 Block Excavation In 2010, an additional 2 x 4-m block was excavated at Locus B by members of the St. Johns Archaeological Field School (Fi gure 6-29). Oriented north-south, the 2010 block was aligned with the eastern edge of the 2009 block and offset two meters to the south, placing it at the western edge of Locu s Bs shell node. It was positioned in an attempt to intersect the southern edge of the emplaced sand platform encountered in 2009 and thereby determine what, if any, functi on the sand may have served. As in 2009, the ultimate goal was to locate any architectural features or patterns in the distribution of artifacts and/or features from which houses or other domestic structures could be inferred. Excavation Methods. The excavation block was divided into two 2 x 2-m test units, Test Unit 43 (TU43) in the north and Test Unit 44 (TU44) in the south. The basic excavation methods employed were identical to those utilized in 2009 and were once again geared toward the collection of fine-gra ined spatial data. As with the previous year, all artifacts larg er than two centimeters in diamet er were point-plotted, although in this case plot locations were measured manua lly using folding rules and line-levels rather than a total station. Depths for both units we re measured from a local datum set at 10-cm above the northeast corn er of TU43. At approximately 50 cmbd, a pit feature (Feature 45) was encountered along the eastern portion of TU43s north wall. Within the feature, a large portion of a fibertempered ceramic vessel protruded out of the wall and into TU43. In order to expose the feature in its entirety and recover the pot wit hout destroying it, a small L-shaped test unit covering 1 m2 (Test Unit 45) was tacked onto the northeast corner of the block. Following the same excavation methods employe d for the rest of the block, it was initially excavated down to the level of the exposed pot. Once this level was reached and the sherds recovered, Test Unit (TU45) was left untouched until excavation of TU43 and TU44 was completed, so that their original profiles could be recorded. Subsequently, TU45 was excavated down to the same level as the other two test units.

PAGE 250

Silver Glen Run, Locus B (8LA1-West) 239 2010 Block Stratigraphy Eleven distinct stratigraphic units were identified in TU43 and TU44 from the 2010 block excavatio n. Composite drawings and photographs of the stratigraphic profiles from the block s walls are shown in Figures 6-30 through 633. Descriptions of the major stratigraphic units are provided in Table 6-21. Because TU45 was excavated and profiled independen tly from the rest of the block, its stratigraphic data are reported separately in Figure 6-34 and Table 6-22. Nonetheless, unless otherwise indicated, the following in-t ext discussion of the blocks stratigraphy utilizes strata designations from TUs 43 and 44. Artifact counts for each level and zone of all three test units comprising th e 2010 block are shown in Table 6-20. Figure 6-29: Field school students excavating the 2010 block at Locus B, 8LA1W.

PAGE 251

240 St. Johns Archaeologi cal Field School 2007-2010: Silver Glen Run Figure 6-30. Stratigraphic drawing and compos ite photograph of north profile from 2010 block (TUs 43 and 44), 8LA1W. (Note: photograph not to scale.)

PAGE 252

Silver Glen Run, Locus B (8LA1-West) 241 Figure 6-31. Stratigraphic drawi ng and composite photograph of east profile from 2010 block (TUs 43 and 44), 8LA1W.

PAGE 253

242 St. Johns Archaeologi cal Field School 2007-2010: Silver Glen Run Figure 6-32. Stratigraphic drawing and composite photograph of west profile from 2010 block (TUs 43 and 44), 8LA1W.

PAGE 254

Silver Glen Run, Locus B (8LA1-West) 243 Figure 6-33. Stratigraphic drawing and compos ite photograph of east profile from 2010 block (TUs 43 and 44), 8LA1W. (Note: photograph not to scale.)

PAGE 255

244 St. Johns Archaeologi cal Field School 2007-2010: Silver Glen Run Table 6-21. Stratigraphic Units of 2010 block (Test Units 43 and 44), 8LA1W. Stratigraphic Max. Depth Munsell Unit cm BD1 cm BS2 Color Description I 35 30 10YR3/2 Very dark grayish brown, organically enriched sand with moderate density whole and crushed Viviparus and abundant live roots. Plow zone. II 76 59 10YR4/2 Dark grayish brown sand with high density of mostly whole Viviparus frequent lenses of burned crushed bivalve and occasional Pomacea. III 55 50 10YR5/1 Very dense burned and crushed bivalve shell in a small amount of gray sand. IV 71 67 Very high density of whole Viviparus and crushed bivalve with virtually no soil matrix. V 83 68 10YR3/2 Buried A-horizon. Very dark grayish brown medium sand with low density whole Viviparus. VI 118 84 10YR5/3 Medium grayish sand with low density crushed bivalve, occasional whole Viviparus and Pomacea VI-A 87 80 10YR5/3 Medium grayish brown sand with low density crushed bivalve. Abundant mineralized roots, and patches of very pale brown sand (10YR8/3). VI-B 95 81 Very dense and compacted whole and crushed Pomacea and bivalve with virtually no soil matrix. VII 102 96 10YR5/3 Medium grayish brown sand with very high density whole Viviparus Smaller amounts of crushed Pomacea and bivalve. Feature 45. VII-A 103 97 10YR6/2 Light brownish-gray, ashy sand with moderate density whole Viviparus occasional crushed bivalve and frequent mineralized roots. VIII 108 92 10YR3/1 Medium very dark gray shand with very low density crushed shell. Second buried A-horizon. IX 105 97 10YR3/1 Very dark gray sand with high density whole Viviparus Feature 49. X 145 131 10YR4/3 Medium brown sand with occasional whole Pomacea and bivalve. Feature 48. X-A 125 109 10YR8/3 Mneralized roots and sand with frequent whole and crushed bivalve. Highly concreted. X-B 137 122 Very dense concreted Pomacea and bivalve with virtually no soil matrix. XI 150 137 10YR4/3 Medium brown sand with no shell. Sterile.

PAGE 256

Silver Glen Run, Locus B (8LA1-West) 245 Figure 6-34. Stratigraphic drawings and compos ite photographs of profiles from Test Unit 45, 8LA1W. (Note: photographs not to scale.) The plow zone exposed by the 2010 block ranges from 10-25 cm thick and largely corresponds to excavation Level A/B, although it extends down into Level C in some places. Like other excavated areas of Locus B, it consists of very dark, organically enriched sand with low to moderate density whole and crushed Viviparus shell and abundant small to large roots. It contains a relati vely high density of mostly plain St. Johns ceramics, several lithic tools and flak es, and a small amount of vertebrate fauna. Again, the St. Johns component is largely conf ined to this upper, disturbed stratigraphic unit.

PAGE 257

246 St. Johns Archaeologi cal Field School 2007-2010: Silver Glen Run Table 6-22. Stratigraphic Units of Test Unit 45 (2010 block), 8LA1W. Stratigraphic Max. Depth Munsell Unit cm BD1 cm BS2 Color Description I 25 18 10YR3/2 Dark grayish-brown fine sand with low to medium density whole and crushed Viviparus IIA 41 39 10YR4/2 Dark grayish brown fine sand with medium density whole Viviparus. IIB 79 74 10YR4/2 Dark grayish brown fine sand with high density crushed bivalve and occasional whole Viviparus. IIC 110 104 10YR2/2 Very dark brown fine sand with low density whole Viviparus IID 141 138 10YR3/2 Very dark grayish brown fine sand with medium to high density crushed bivalve and occasional whole Viviparus IIE 142 135 10YR2/2 Very dark brown fine sand with occasional crushed bivalve and whole Viviparus. IIIA 52 45 Very high density whole Viviparus and crushed bivalve with virtually no soil matrix. IIIB 120 113 10YR4/2 Dark grayish brown fine sand with high density whole Viviparus. IIIC 171 165 10YR2/2 Very dark brown fine sand with sporadic whole Viviparus IV 170 163 10YR3/3 Dark brown fine sand. V 162 168 10YR4/4 Dark yellowish brown fine sand. In this area, as in the 2009 block just to the north, the plow zone sits atop a massive stratum of mostly w hole shell with varying amounts of dark grayish brown sand (Stratum II). At one location in the northeast corner of TU43 (Stratum IV), shell density is so high that virtually no so il matrix can be discerned. This shell layer consists primarily of large whole Viviparus but also contains occasional Pomacea and is crosscut in several locations by thin horizontal lenses of burned and crushed bivalve. While the top of this stratum contains a significant amount of St. Johns Plain ceramics, this is most likely a result of plow disturbance. Orange fiber-tempered ceramics are present throughout, although their density is highest around the crushe d bivalve lenses near the stratums center. They include roughly equal proportions of plain a nd incised varieties. Other artifacts from Strata II/IV are sparse and include a few chert flakes, small bits of marine shell, and a low fre quency of vertebrate fauna.

PAGE 258

Silver Glen Run, Locus B (8LA1-West) 247 Table 6-23. Cultural Materials Recovered from Level Excava tion of 2010 block (Test Units 43, 44, and 45), 8LA1W.

PAGE 259

248 St. Johns Archaeologi cal Field School 2007-2010: Silver Glen Run Stratum II is interrupted in multiple spots along the west profile by patches of charcoal and burned black sand that probably resulted from tree roots burning in place. In the east profile near the southeast corner, a large branching tree root penetrated down into Stratum II from the base of the plow zone It was initially thought that this root may have been a small Archaic-age tree preserved as a result of being encased in the Stratum II shell. A sample of the root, however, re turned conventional radiocarbon assay of 120 40 rcybp (280-0 cal BP), indicati ng that it is actually the lo wer root portion of a modern tree, perhaps removed when the land wa s cleared for plowing in the early 20th-century. The shell cap in the 2010 block (Strata II/I V) thickens substantially from east to west, increasing from around 20 cm in the east profile to upwards of 45 cm in the west. This trend appears to result from the slop ing buried A-horizon (Stratum V) upon which the shell was deposited. This buried soil horizon consists of very dark grayish brown soil with low shell density and slopes upward from west to east (see south profile in Figure 633). As observed within other Locus B test units, this buried surface resembles the modern active A-horizon in exhibiting an unde rlying concentration of mineralized root casts. It is possible that the shell constituting Strata II/IV was used to level this preexisting sloped surface. A lternatively, if the shell deposits at Locus B have been truncated by plowing or other recent earth-moving activities, then the Strata II/IV may represent deposits of mounded sh ell that actually accentuated the sloping surface before they were scraped flat. It appears unlikel y, however, that signi ficant Late Archaic deposits have been removed, given the abunda nce of St. Johns materials that remain within the modern plow zone. In the northwest corner of the block the Stratum II shell drops down approximately 110 cm, filling in a huge Ora nge period pit (Feature 51). The shell demarcating the exposed portion of this pit extends approximately 50 cm out from the northwest corner into TU43 and is identical to that observed in the overlying shell stratum. At least two additional shell-fille d pits were exposed during the excavation of TU45, although they were not designated as such during excavation. One of these undesignated features can be seen in Figure 6-34 cutting down through the eastern edge of TU45s north profile and coming back up in the east, while the second partially overlaps the southern margin of the first and is visible in both the east and south profiles. This southernmost feature is similar to Feature 51 in that there is no distinction between the pit fill and the overlying shel l cap, suggesting that some pits may have been open at the time that the shell layer was deposited and were infilled in one or a few large scale depositional events. This is perhaps a furt her indication that th e shell cap was used consciously as means for obscuring existing topographical irregularit ies or renewing the existing Locus B surface. These massive pits are similar in scale and morphology to those previously excavated at Locus B and indicate widespread intensive use of this area during the Orange Period. Once again, however little was found within them in terms of typical domestic debris that would suggest they resulted from everyday habitation activities. Instead, their extraordinary size and overlapping distribution suggest intermittent pulses of extremely intensive activity, perhaps geared toward the rapid processing of large quantities of shellfish.

PAGE 260

Silver Glen Run, Locus B (8LA1-West) 249 This cap is visible across all 2010 bloc k profiles with the exception of the north profile of TU45. There, an undifferentiate d sand deposit exists between the two pit features that probably represents an inta ct portion of the emplaced sand platform observed in the southern half of the 2009 bloc k. Unfortunately, this appears to be the only section of the emplaced sand that was not obliterated by subsequent pit-digging and is not particularly informative regarding th e features function or meaning. Underneath the buried A-horiz on, excavations uncovered a layer of grayish brown sand with low (Stratum VI) to very high (Stratum VI-B) de nsities of whole and crushed Pomacea and bivalve shell. This provides a noticeable contrast with the subsequent Viviparusdominated Orange deposits and implies a distinct suite of shell deposition activities during this time. Stratum VI ranges in thickness from 10 cm to upwards of 50 cm. In spots (designated Stra tum VI-A), it is permeated by abundant mineralized root casts associated with the overl ying buried surface. No ce ramic sherds were recovered from this stratum, indicating its probable preceramic Mount Tayl or age. It does, however, contain frequent lithic and marine shell artifacts and relatively abundant vertebrate fauna compared to overlying deposits. Stratum VI was deposited on top of a s econd buried A-horizon (Stratum VIII) that is most clearly visible in the blocks east profile as a 10 to 20-cm-thick layer of very dark, organic sand. The densest shell within Stra tum VI was observed at its base along the contact with this organic layer, suggesting th at the top of Stratum VIII is most likely the original Mount Taylor period surface. An in crease in the density of cultural materials was noted as this surface was approached dur ing excavation. Recovered artifacts include the base of a Newnan point, lithic flakes, ma rine shell fragments, and vertebrate fauna. Three pit features were also found to orig inate from this surface, including a broad shallow basin (Feature 49), a small straight-sided cylindrical pit filled with Viviparus (Feature 50), and a massive roughly 2-m-wide basin that is lined with dense concreted Pomacea and bivalve (Feature 48). All three ar e devoid of ceramic s and Feature 48 contains a moderate density of vertebrate fauna remains. Charcoal samples from Features 50 and 48 returned radiocarbon a ssays of 4180 40 rc ybp (4810-4430 cal BP) and 4240 40 (4860-4650 cal BP), situating them late in the Mount Taylor Period within the recently defined Thornhill Lake Phase (5600-4500 cal BP) (Endonino 2008). The diversity of features and artif acts, along with the higher fre quencies of vertebrate fauna, suggest that the Mount Tayl or component revealed by the 2010 block resulted from sustained activities associated with everyday living. These Mount Taylor deposits are underlain by medium brow n, virtually sterile subs oil (Stratum XI). Locus B Depositional Patterns Excavation of a 4 x 4-m block in 2009 and a 2 x 4-m block in 2010 exposed evidence for the same basic culture-historica l components previously identified in Locus Bs initial exploratory test un its, albeit at a much broader and more revealing scale. When combined with the data from the explor atory test units, the added insight gained from these relatively expansiv e excavations has exposed ev idence for three distinct patterns of shell deposition at Locus B (see Figure 6-35). These patterns each exhibit

PAGE 261

250 St. Johns Archaeologi cal Field School 2007-2010: Silver Glen Run Figure 6-35. Map showing distribution of Late Ar chaic depositional patterns at Locus B, 8LA1W. distinct material cu lture assemblages and correspond to th ree fundamentally different uses of this place during the Late Archaic. Together, they divulge a great deal regarding the Locus Bs dynamic depositional history and its shifting role in larger-scale social processes. Depositional Pattern 1 (DP1) As noted above, the earliest shell deposition at Locus B is centered on the high point of this areas shell node and in fact, these cultural deposits, and not natural fluvial processes are responsible for most of Locus Bs modern topographic relief. The preceramic occupation responsible for this pattern involved a small domestic settlement where everyday act ivities were carried out off and on over a period of hundreds of years. TU46, excavated near the center of the shell node, revealed a series of thin horizontal deposits of shel l with intervening thin layers of dark, organically enriched sa nd that is virtually shell-free. In terms of composition, the thin shell layers are made up primarily of crushed freshwater bivalve and apple snail, but also include lithic tools and debitage, marine shell, modified and unmodified vertebrate fauna, and paleofeces. It is likely that these altern ating layers represent a sequence of at least 4 repeated habitations and abandonments. Approximately 20 m to the southwest, we see a continuation of this same basic pattern in deposits uncovered in the 2010 block excavation. Again, there is a horizontal layer of crushed bivalve and Pomacea shell but in this location there are also a series of pit features descending from this appare nt habitation surface. Varying in size, morphology, and content, these appear to cons titute three functionally distinct features within a preceramic domestic context. The ar tifacts recovered from this stratum and the associated pits include a diverse array of bone lithic, and shell tools and a relatively large

PAGE 262

Silver Glen Run, Locus B (8LA1-West) 251 quantity of vertebrate fauna, all of which support a domestic interpretation of these deposits. These materials were deposited on top of a well-developed intact A-horizon, the original preceramic surface at Locus B. Calibrated 2 sigma ranges from a series of four radiocarbon assays indica te that DP1 dates to the la te preceramic Thornhill Lake Phase between 5740 and 4580 cal BP. Depositional Pattern 2 (DP2). Radiocarbon and stratigraphic data indicate that this area of Locus B was abandoned near the end of the Thornhill Lake phase for perhaps a few hundred years, but at least long enough that substantial natural soil development was allowed to take place. Further evidence of this abandonment exists in the large mineralized tree roots that permeate the soil just below this second buried A-horizon. The next people to occupy Locus B did so during the subsequent Orange Period and initiated a depositional pattern wholly inconsis tent with their preceramic predecessors. Where before the sites inhabitants had deposited materials and prepared features indicative of residential domes tic activities, DP2 entailed a mode of inhab itation centered on the excavation and use of extraordinarily la rge pits. Many of these pits overlap, dug one on top of another across a broad area stre tching at least 50 m fr om the western edge of Locus Bs shell node over to TU57. Burned and concreted bivalve shell in the bottoms of multiple pits hints at their use in shellfish processing, although with dimensions sometimes exceeding a meter in both diameter and depth, these pits seem out of proportion wi th domestic food processing. This period also saw the introduc tion of the regions fi rst ceramic technology to Locus B. Aside from a modest numbe r of undecorated fiber-tempered pot sherds, however, relatively few cultural materials were deposited during this time, especially considering the apparently high-intensity activities taking place. Consequently, little evidence exists suggesting that Locus B wa s a place of residenc e during this time. Instead, the size and overlapping nature of th e pits point to successive short-term, high intensity events involving th e processing of shellfish at a communal or even extracommunal scale. Calibrated 2 sigma ranges from a series of seven radiocarbon assays situate DP2 within the Orange period between 4520 and 3830 cal BP. Depositional Pattern 3 (DP3). Shortly following the e nd of large-scale pit digging, a large quantity of mostly whole Viviparus shell was deposited across the surface of Locus B, an event marking another major transition in the sites history. DP3 forms a 30-50 cm thick, mostly homogeneous stratum of dense unconsolidated shell that at many places contains little if any soil matrix. Like the pits below it, this stratum contains only very sparse vertebrate fauna a nd artifacts, save for a small amount of fibertempered pottery. In contrast to the undeco rated pottery from the pits, however, many of the sherds recovered from this deposit exhi bit the curvilinear incisions and punctuations typical of the Tick Isla nd style of decoration. In many locations this layer of shell is completely undifferentiated from top to bottom. In others, it is crosscut by thin roughly horizontal laye rs of crushed bivalve similar in many respects to the preceramic surfaces noted in DP1. These may be an indication that DP3 did not enta il a single massive depositiona l act but rather a repeated

PAGE 263

252 St. Johns Archaeologi cal Field School 2007-2010: Silver Glen Run sequence of formal surface preparations usi ng crushed bivalve followed by deposition of clean Viviparus. Nevertheless, the clean and unfrag mented nature of the shell suggests a series of large-scale intentional capping events and not the gradual accumulation of domestic debris. The layer of shell constituting DP3 is virtua lly coextensive with the pits underlying it and in some places appears to have infilled open pits, in effect turning what must have been a rough and uneven surface into a relatively flat and smooth one. This massive mantle of shell is not unlike the ones noted earlier that cap discontinued Mt. Taylor habitation spaces (Randall 2010; Sa ssaman 2010) and perhaps constitutes the renewal of a long lived tradition of laying down whole shell over places at the end of their use lives. Two radiocarbon assays on samples recovered from Feature 38 (the only pit feature containing Tick Is land style pottery similar to that recovered from DP3 deposits) tentatively date this stratum to 4140-3830 cal BP. FEATURE ASSEMBLAGE In total, 34 features were recorded during the 2007-2010 excavations at Locus B, of which 25 were determined to be of pr e-modern cultural orig in (see Table 6-24). Following examination of excavatio n profiles, an additional four features were identified that had gone unrecognized in the field an d as a result, were not assigned feature numbers. All Locus B features have been in terpreted as infilled pits. These pits were classified according to shape and size as follows: Type 1: includes shallow basin-shaped pits with outward sloping margins; maximum diameters range from 2587 cm and depths from 8-20 cm; Type 2: includes broad, deep basin-shap ed pits with outward sloping margins; maximum diameters range from 67-230 cm and depths from 42-73+ cm; Type 3: includes small cylindrical pits w ith vertical margins; maximum diameters range from 40-45 cm and depths from 31-51 cm; Type 4: includes large cylindrical pits w ith vertical margins; maximum diameters range from 60-140+ cm and de pths range from 50-102 cm; Type 5: includes conical pits with inward sloping margins; maximum diameter of 120 cm and depth of 94 cm; Type 6: includes isolated shell pocket s of presumed cultural origin: maximum diameters range from 38-47 cm and depths from 20-28 cm. Typical examples of the various feature types and their relative stratigraphic distributions can be seen in Figure 6-36. The Thornhill Lake Phase deposits constituting DP1 include the widest variety of featur e types, even though this component has undergone the smallest amount of excavation in terms of total surface area. DP1 features include small basins (Type 1), one large sha llow basin (Type 2), small cylindrical pits (Type 3), and isolated shell pockets (Type 6) Most of these DP1 pits are found within

PAGE 264

Silver Glen Run, Locus B (8LA1-West) 253 the relatively restricted space encompassed by the southern two thirds of the 2010 block (see Figure 6-37). In this location, various pit types were dug down from a buried surface occurring at approximately 90-100 cmbd. Table 6-24. Cultural features recorded at Locus B, 8LA1W. Feature No. Type Shape Dimension (cm) Depth (cm) Cult.-Hist. Affiliation Depo. Pattern 1 1 basin 25 14 Orange 2 2 downgraded 3 downgraded 4 1 basin 45 x 27 10 Orange 2 5 downgraded 13 downgraded 14 1 basin 17 x 12 8 Orange 2 15 2 basin 131+ x 52+ 75? Orange 2 16 2 basin 100+ 80? Orange 2 17 6 amorph. 45 x 22 20 Thornhill? 1 25 6 amorph. 38 x 30 28+ Thornhill? 1 26 4 cylinder 60 65+ Orange 2 27 4 cylinder 100+ 50+ Orange 2 29 downgraded 30 downgraded 31 downgraded 33 4 cylinder 100 x 66 84 Orange 2 34 1 basin 87 x 83 12 Orange 2 35 1 basin 82 x 74 16 Orange 2 36 2 basin 120+ x 80+ ~90 Orange 2 37 1 basin 55 x 34 19 Orange 2 38 5 cone 120 x 72+ 94 Orange/T.I. 3? 39 downgraded 41 2 basin indeterminate indeterminate 42 4 cylinder 100+ indeterminate 45 4 cylinder 71 x 58 70 Orange 2/3? 48 2 basin 230 x 135+ 42 Thornhill 1 49 1 basin 54 x 23 20 Thornhill 1 50 3 cylinder 45 x 43 31 Thornhill 1 51 4 cylinder 140+ x 100+ 102 Orange 2 52 2 basin 67 x 27 65 St. Johns? 53 downgraded 54/55 2 basin 227+ x 134+ 50+ Orange 2

PAGE 265

254 St. Johns Archaeologi cal Field School 2007-2010: Silver Glen Run Figure 6-36. Schematic showing typical examples of feature types and their stratigraphic distributions within Locus B deposits

PAGE 266

Silver Glen Run, Locus B (8LA1-West) 255 Feature 6-37. Map showing horizontal distribu tion of pit feature types in the 2009 and 2010 blocks, 8LA1W.

PAGE 267

256 St. Johns Archaeologi cal Field School 2007-2010: Silver Glen Run DP2 features are far less varied and re flect a major transition in the mode and temporality with which Locus B was utili zed during the Orange period. The most striking aspect of DP2 features is their scal e, as almost all of Locus Bs truly massive Type 2 and Type 4 features are associated with this pattern. These unusually large pits are found in all excavated contexts in the north ern half of Locus B from TU22 in the east to TU57 in the west. They are tightly bunche d across much of this area and frequently overlap making the delineation of some p it boundaries impossible. This has undoubtedly led to an underestimation of their total numbe r. Smaller Type 1 features are also found within DP1 deposits, although at a lowe r frequency than Types 2 and 4. The only Type 5 feature has tentatively been associated with DP3. This is a large conical pit, similar in size to many of the DP 2 pits but with a tape red base and complex stratified fill. It stradd les the boundary between TU39 and TU40 along the northern edge of the 2009 block. All feature types and individual features are discussed in more detail in the section that follows. Type 1 Features A total of seven Type 1 features were reco rded at Locus B. While these features all share a shallow basin shape in profile, in plan view, they vary widely in size and shape, ranging from small a nd roughly circular to long an d ovoid. They also hold a variety of different fills w ith some enclosing only dark organic sands and others containing dense deposits of shell, bone, and other artifacts. The pr ecise function(s) of Type 1 features is unknown, although they may have served variously as roasting pits, small-scale storage containers, or even as receptacles for votive offerings. As noted above, they are found within both the 2009 and 2010 blocks and are distributed throughout DP1 and DP2, thus bridging the gap between the preceramic Thornhill Lake and early ceramic Orange Period occupations of Locus B. Feature 1 Feature 1 (Figure 6-38) is a small circular Type 1 pit located in the southern half of TU4. Measuring approxima tely 25 cm in diameter, it was initially recognized at 86 cmbd as a very dark gray (10YR3/1) pocket of loamy soil containing denser shell than that in th e surrounding matrix. In crosssection, Feature 1 is a shallow basin with a maximum depth of 14 cm. It contains dense whole and crushed Viviparus and bivalve shell, some of which shows signs of burning. Bits of charcoal and small fiber-tempered ceramic sherds were also recovered from the pit, dating it to the Late Archaic Orange period. It is possible that this feature func tioned as a small, temporary cooking hearth or roasting pit, although none of the matrix surrounding the pit shows any signs of having been exposed to fire. Feat ure 1 was bisected along a north-south axis. The eastern half was removed as a bulk sample for flotation, while the western half was -inch screened. Feature 4. Feature 4 (Figure 6-39) is a Type 1 pit located in the southeastern corner of TU4. Although only a quarter-section of this feat ure was uncovered, it appears amorphous in shape, with exposed dimensions of 70 x 63 cm. A large mineralized root that was originally designated Feature 5 was observed along the northern edge of the pit

PAGE 268

Silver Glen Run, Locus B (8LA1-West) 257 Figure 6-38. Drawing of the plan view of Feature 1, TU4, 8LA1W. Figure 6-39. Drawing and photograph of the plan view of Feature 4, TU4, 8LA1W. (Note: photograph not to scale.) and may be at least partially responsible for its unusual morphology. Feature 4 was not recognized until near its base but the portion vi sible in the south a nd east profiles of TU4 exhibits a maximum depth of approximately 10 cm. The pit is characterized by very dark grayish brown (10YR3/2) loamy soil matrix containing moderately dense whole Viviparus shell. After documenting the feature, the entire intact po rtion was removed as a bulk sample for flotation. Although no artifacts were observed during excavation, some

PAGE 269

258 St. Johns Archaeologi cal Field School 2007-2010: Silver Glen Run may still exist within this unanalyzed sample. Stratigraphically, Feature 4 is situated near the top of DP2 deposits, likely situ ating it within the Orange period. Feature 14 Feature 14 is visible in cross-section in the northern half of TU4s east profile. It has been tentatively interpreted as a very small and shallow basin-shaped pit but could alternatively be th e base of an infilled posthole. Feature 14 was recognized during excavation as a discre te pocket of dense whole Pomacea in slightly darker brown sand than the surrounding matrix. The shell is most concentrated near the base of the pit and is mixed with frequent lumps of char coal. Due to its small size, Feature 14 was excavated in its enti rety as a bulk sample for flotation. Feature 34. Feature 34 (Figure 6-40) is a broad shallow basin-shaped pit located in the eastern half of TU39. It is roughly circ ular in plan view, m easuring 87 x 83 cm at the top. It is characterized by abundant whole Viviparus shell in very dark grayish brown (10YR3/2) medium sand. Much of the sh ell excavated from Feature 34 is semiconcreted, especially that found near the center of the pit. One fiber-tempered plain sherd was recovered from this area and two other large sherds were found near the features southwestern periphery, near the base of Feature 29. The cl ose proximity of Feature 29 may have impacted this portion of the pit, while a mineralized root runs through its eastern half. Feature 34 was bisected along an axis running northwe st to southeast. The northeastern half of the feat ure was removed as a bulk for flotation analysis while the southwestern half was 1/8-inch water screened. Excavation of the water screen sample was carried out to a point below the actual lower margin of the feature in order to ensure that the bottom had been reached and to expos e a complete cross-section of the pit. It revealed a shallow basin with gently sloping si des that bottoms out in the northwest with a maximum depth of approximately 12 cm. Feature 35 Feature 35 (Figures 641 and 6-42) is a shallo w pit extending out of the east wall of TU42. It exhibits an unusual elongated ovoid shape that distinguishes it among Type 1 features and results in a line ar trough-like appearance. It was first recognized at an absolute depth of 14.699 m (bas ed on the local site datum) as a discrete concentration of whole bivalve and Pomacea. Excavation of the pit also revealed occasional whole Viviparus, crushed shell fragments of vari ous types, and small flakes of charcoal. No artifacts were r ecovered that can confidently be attributed to the feature but a marine shell fragment found near its edge may be associated. Feature 35 lies within DP2 deposits, right at the edge of the empl aced sand anomaly discovered in the southern half of the 2009 block and appears to origin ate from approximately the same surface as the one on which the sand was deposited. It is likely that Feature 35 was intentionally placed in relation to the emplaced sand and the two may have had interrelated functions, although it remains unclear what those may have entailed. In the field, this pit was bisected along an east-west axis, revealing the crosssection of a shallow basin w ith a maximum depth of 16 cm and regular, gently sloping margins. The cross-section grades in color from medium brown (10Y R4/3) at the top to

PAGE 270

Silver Glen Run, Locus B (8LA1-West) 259 Figure 6-40. Drawings and photograph of the plan view and excavated cross-section of Feature 34 from TU39, 8LA1W. (Note: photograph not to scale.) dark yellowish brown (10YR4/4) at the bottom. A thin 2-4-cm layer of dense bivalve and Pomacea cuts through the center. A potential disturbance exists in the form of a large mineralized root that runs along, or perhaps truncates, the northern edge of the pit and may have affected the features current shape. Feature 37. Feature 37 (Figure 6-43) is a T ype 1 pit feature located in the southwest quadrant of TU41. It stood out dur ing excavation as a dense concentration of bivalve shell and very dark grayish brown (10YR3/2) sand in an otherwise mostly shellfree stratum. This feature is unusual in that it is filled prim arily with whole bivalve shells that were paired and unopened prio r to excavation. On the west side of the pit, a layer of extremely dense whole and concreted bivalve lin e the bottom of the pit. Underneath this concreted layer is a lens of bright red oxidized sand. This type of thermally altered sand is common at the bases of larger Type 2 a nd Type 4 pits at Locu s B but is unique to Feature 37 among Type 1 features. The oxi dized sand, along with a small amount of charcoal, and the whole unopene d bivalve, may indicate th at Feature 37 was a small

PAGE 271

260 St. Johns Archaeologi cal Field School 2007-2010: Silver Glen Run Figure 6-40. Drawings of the plan view and excavated cross-section of Feature 35 from TU39, 8LA1W. Figure 6-41. Photograph of the excavated cross-section of Feature 35 from TU39, 8LA1W.

PAGE 272

Silver Glen Run, Locus B (8LA1-West) 261 Figure 6-43. Drawings and photograph of the plan view and excavated cross-sections of Feature 37 from TU39, 8LA1W. (Note: photograph not to scale.) roasting pit that was filled w ith freshwater clams, covered, and then forgotten about. A moderate density of Viviparus and small number of fiber-tempered plain ceramic sherds are the only other materials documented during excavation. A small concreted root runs through the base of the feature. Feature 37 was bisected along a north-south axis, revealin g a shallow basin with a maximum depth of approximately 19 cm. Th e west half was removed for flotation analysis and the east half was 1/8-inch water screened. A charcoal sample recovered from the bulk returned a radiocarbon assay of 3630 40 rcybp (4080-3850 cal BP), which is within the temporal range of other DP2 features. Feature 49 Feature 49 (Figure 6-44) is a Type 1 feature overlapping the eastern edge of TU44 in the 2010 block. Although its actual top sits at 88 cmbd, this feature was not recognized as such until near its base when a concentration of Viviparus and bivalve in dark grayish brown (10YR3/2) sand became a pparent in both the floor and east profile of the test unit. The exposed half of the f eature measures 54 cm in diameter and extends approximately 23 cm into TU44. It appears to be the western half of a roughly circular

PAGE 273

262 St. Johns Archaeologi cal Field School 2007-2010: Silver Glen Run Figure 6-44. Drawings and photograph of the plan view and excavated cross-section of Feature 49 from TU44, 8LA1W. (Note: photograph not to scale.) pit containing a moderate density of shell but no other observable artifacts. Bisected by the test unit wall, the feature originates fr om the lower buried A-horizon observed within the 2010 block, making it a part of DP1. It may consequently be related to two other pits (Features 48 and 50) discovered at this same level. Feature 49 is basin shaped and relatively shallow with a maximum depth of around 20 cm. Upon its recognition at approximately 100 cmbd, the entire remaini ng portion of the feature was removed and bagged as a bulk sample for flotation analysis. Type 2 Features A total of eight Type 2 features have been documented at Locus B. Like Type 1 features, these exhibit a basic basin shape in cross-section with slopi ng sides and rounded or flat bottoms. They are clearly distingui shed, however, by their scale, with maximum diameters ranging from 67-230 cm and depths from 42-73+ cm, resulting in volumes many times greater than those displayed by T ype 1 pits. Pit fill varies considerably

PAGE 274

Silver Glen Run, Locus B (8LA1-West) 263 among features, with some containing only a trace of shell while others enclose large, dense shell deposits. At least four Type 2 f eatures contain highly structured deposits of shell and earth, resulting in complex stratified profiles. As noted above, most of these massive pits (at least six of the eight) are associ ated with DP2. At least five of these have evidence for burning near their bases, a fact that, when combined with the ambient shell and general lack of vertebrate fauna throughout DP2 deposits, suggests a primary function related to the steaming or roasting of shellfish. Type 2 features have a horizontal distribution that st retches across the entire area encompassed by the Locus B excavations. Within this area, individual pits frequently overlap each other, often making their recognition difficult and their pr ecise delineation nearly impossible. Features15/16. Features 15 and 16 (Figure 6-45) are two large basin-shaped pits stretching across the entire southw estern half of TU4. These pits were first recognized as a single continuous concentration of shell in the floor of th e test unit at approximately 110 cmbd. This concentration was later dete rmined, however, to include two distinct pits. The features appear in stark contrast to the virtually shell-free sand surrounding them. Examination of the south and east prof iles of TU4 reveals that the pits probably originate from a surface just below the base of the DP3 shell, perhaps at around 70-80 cmbd. This results in estimated maximum depths of 75 cm for Feature 15 and 80 cm for Feature 16. Horizontally, the visible portions of both pits extend for over one meter in diameter, although their actual dimensions ar e obscured by their overlapping margins and the limited perspective provided by the 1 x 2-m test unit. If the upper boundary of the features proposed above is correct, then the fill from both pits can be said to exhibit some level of stratification. At least three distinct filling episodes are evident in both features. These can be seen in the cross-sections of the features shown in the south and east prof iles of TU4 (see Figure 6-16 above). The lowermost, and presumably earliest, fill in Feature 15 consists of brown (10YR4/3) fine sand containing very dense, mostly burned sh ell, ash, and charcoal. Shell constituents include mostly whole and crushed Viviparus with pockets of whole burned bivalve. Portions of this stratum are solidly conc reted and had to be broken up with a pick hammer during excavation of the feature. The bottom stratum of fill in Feature 16 is very similar to that in Feature 15 but wi th the addition of whole and broken Pomacea Plain fiber-tempered pottery was found within the shel l of both features. During excavation of the lower portion of these pits, a smell sim ilar to burned rubber was noted. Above this basal stratum, the stratigraphic sequence of the two pits is virtually identical. In each case, a 25 to 45-cm thick layer of dark yello wish brown (10YR3/4) sand with only sparse whole Viviparus was deposited directly on top of the dense shell. In Feature 16, this layer is penetrated by a possible animal burrow and multiple mineralized roots. Following this, roughly 25 cm of dense whole Viviparus in a very dark grayish brown (10YR3/2) sand was laid down, thus completing the fill sequence. Relatively abundant vertebrate fauna is present with in this upper layer. The fact that this sequen ce of discrete depositional acts was repeated in two dis tinct pits dug at different times suggests intentional structured deposition rather than the haphazard disposal of everyday refuse.

PAGE 275

264 St. Johns Archaeologi cal Field School 2007-2010: Silver Glen Run Figure 6-45. Drawing and photograph of plan view of the basal strata of Features 15 and 16, TU4, 8LA1W. (Note: photograph not to scale.) As noted above, these features were not recognized until their dense basal shell was encountered at approximately 110 cmbd. Consequently, this stratum alone was preserved for fine screening. In each case, sta ndard feature protocol was followed so that half of the fill was 1/8-inch water screened while the remaining half was removed in bulk for flotation analysis. The abundance of shell in the bottom of Features 15 and 16, (including substantial amount s of the larger, meatier Pomacea and bivalve), along with

PAGE 276

Silver Glen Run, Locus B (8LA1-West) 265 the evidence for burning, indicate that these pits were likely utilized as shellfish roasting facilities. Their extraordinar y size and elaborate fill sequences, however, suggest a significance that extends beyond small-scale s ubsistence economics. Charcoal from the base of Feature 15 returned an AMS radiocarbon assay of 3830 40 rcybp (4410-4100 cal BP). Feature 36. Another probable DP2 roasting pit, Feature 36 (Figures 6-46 and 647) is an extremely broad and deep basin lo cated in the northeast co rner of TU40, a part of the 2009 block. The pit appears to be roughl y oval in shape, measuring more than 120 Figure 6-46. Drawing of the plan view and cross-section of Feature 36 from TU40, 8LA1W. Figure 6-47. Photographs showing a) the excavated plan view, and b) the excavated cross-section of Feature 36 from TU40, 8LA1W.

PAGE 277

266 St. Johns Archaeologi cal Field School 2007-2010: Silver Glen Run cm long and at least 80 cm wide, although its actual dimensions are obscured by the test unit boundaries. It was initially recognized as a feature at an absolute elevation of 14.528 m, although its actual upper boundary seems to be slightly higher than this. In crosssection Feature 36 appears as a roughly 90-cm deep strai ght-sided and flat-bottomed basin. In the 2009 blocks east profile (see Figures 6-23 and 6-24 above), the top of the pit can be seen to flare outward, significan tly expanding its maximu m upper diameter. Feature 36 fill consists primarily of da rk brown (10YR3/3) sand with moderate density of Viviparus shell that diminishes with depth. Whole and broken bivalve is also present throughout with the highest density occu rring near its base. The pit is largely bereft of material culture with no significan t vertebrate fauna and only a trace of plain fiber-tempered pottery. The basal portion of the western half of the pit consists of a very distinct stratum of heavily oxidized bright orange (5YR4/6) sand. This thermally altered sand extends for 120+ cm across the entire leng th of the feature and measures up to 15 cm thick in some places. A heavy concentration of charco al was found associated with the oxidized sand in th e northeast quadrant of the pit. Upon recognition, Feature 36 was bisected along a southwest-northeast transect. As its southeastern margin was difficult to delineate in plan view, excavation of this half of the feature was extended beyond the known boundary of the pit in order to expose a clear and complete cross-section. Fill from th is half was 1/8-inch water screened. The northwest half of the feature was carefully excavated only up to the boundary of the pit. Bulk samples were removed from this secti on for flotation analysis and the remainder was -inch screened. A charcoal sample fr om the basal bulk returned an AMS assay of 3590 40 rcybp (3980-3830 cal BP). Feature 41 Feature 41 is a pit of unknown dime nsions located in the northwest corner of TU39, within the 2009 block. Unfortun ately, it was not rec ognized as a feature until bright orange oxidized sand and char coal was encountered near its base (at approximately 13.54 m in absolute elevation), so little information regarding its size and morphology were documented. The depth and stratigraphic position of the pits base, nonetheless, suggest a large and extremely deep feature in line with other Locus B Orange Period roasting pits belonging to DP 2. Along the western half of the blocks north profile, a thin arcuate stratum of shell is visible th at may be an upper stratum of Feature 41. The broad, gently sloping basin-like configuration of this shell layer and the similarity of TU41s basal deposits to Featur e 36 in terms of depth and composition are the primary criteria on which Feature 41 was tenuously classified as a Type 2 rather than a Type 4 feature. The pits basal portion wa s removed in bulk for flotation analysis. Feature 48 Feature 48 is an extremely broad, but relatively shallow, preceramic Type 2 feature that stretches across the boundary be tween TU43 and TU44 in the 2010 block. This massive pit originates fr om a hardened clay-like DP1 surface at approximately 86 cmbd that is lined in s pots by thin lenses of dense bivalve, Pomacea, and relatively abundant vertebrate fauna. Feature 48 was initially recognized as an anomalous pocket of sand and shell within this surface. Because of its unusual size, it was at first unclear whether the anomaly was a feature or simp ly an intersecting stratum.

PAGE 278

Silver Glen Run, Locus B (8LA1-West) 267 Feature 6-48. Drawings and photograph show ing the plan view and excavated crosssection of Feature 48 from TU43 and TU44, 8L A1W. (Note: photograph not to scale.) It measures 230 cm in length and at least 135 cm in width, although its actual maximum width is unknown as the feature intersects the western margin of the excavation block. In cross-sect ion, however, the edges of th e pit can be seen to slope regularly down to a maximum depth of 42 cm, forming an expansive shallow basin. This overall shape contrasts significantly with the deep and often straight-sided Type 2 features associated with DP2. Excavation revealed that most of Feat ure 48 is filled with medium brown (10YR4/3) sand with occasional whole shell. The bottom of the pit however, is lined with dense whole and crushed bivalve and Pomacea shell similar to that found lying on

PAGE 279

268 St. Johns Archaeologi cal Field School 2007-2010: Silver Glen Run the surface from which the pit descends. A large concreted mass of this shell and sand was recovered near the center of the pit, at its deepest point. A moderate amount of vertebrate fauna and frequent charcoal we re the only other material culture observed during excavation. The northern half of the pit has been penetrated by a number of mineralized and live roots. In addition, the eastern edge of Feature 48 appears to have been intersected by a small cylindrical pit (F eature 50), which emanates from the same surface. The evidence for burning and the larg e amount of shell in the bottom of the pit suggest that this feature too may have been utilized in cooking or shellfish processing activities. Already crosscut by the western wall of the block, Feature 48 was bisected again along the east-west transect formed by the boundary between TU43 and TU44, and the south half was removed in order to obtain a perpendicular cross-section. Samples were taken from multiple levels and sections of the feature for flotation analysis and 1/8-inch water screening. The remainder of the feat ure fill was -inch sc reened. Charcoal obtained from the base of the pit returned an AMS radiocarbon assay of 4230 40 rcybp (4860-4650 cal BP), which dates it to near the end of the preceramic Thornhill Lake Phase. Feature 52. Feature 52 is a deep basin-shaped pit located along the western edge of TU46. Although the top of this pit sits near the surface at approximately 10 cmbd, it was largely concealed, by the numerous natu ral and modern distur bances that have affected the upper strata of TU46. Conseque ntly, Feature 52 was not recognized as an anthropogenic pit until near its base. In cross-section (see we st profile of TU46 in Figure 6-12) the feature shows up as a large, relativel y deep basin measuring more than 67 cm in length and approximately 65 cm in maximum de pth. The pit is filled with a complex, stratified sequence of deposits that are somewhat obscured by the various disturbances cutting through it. Nevertheless, a pattern consisting of layers of whole and crushed bivalve alternating with layers of whole Viviparus shell is apparent. Excluding a moderate amount of vertebrate fauna, the feature is devoid of any other material culture that might provide a clue to its function. De spite the lack of ceramics within the Feature 52 fill, its stratigraphic origin near the top of the TU46 profile indicates that it probably dates to the St. Johns period, given the relative abundance of spiculate-tempered ceramic sherds recovered from this level of the test unit. Because the f eature was recognized so late, only the basal portion was available fo r sampling. This entire intact section was removed in bulk for flotation analysis. Feature 54/55 Features 54 and 55 (Figure 6-49) are located in the southwest half of TU57. Originally these were assigned se parate feature number s because pit crosssections were visible in both th e south and west profiles of th e test unit with an apparent gap in the corner. Further inspection of the profiles, however, indicates instead one massive pit (hereafter Feature 54/ 55) covering more than half the test unit. Its center is likely near the midpoint of TU57 and its southw est margin angles up toward the corner of the unit. The feature appears roughly circular in shape and based on the test unit profiles exhibits a maximum diameter of more than 230 cm. The pit itself descends down to a maximum depth of approximately 30-40 cm from a buried A-Horizon visible across most

PAGE 280

Silver Glen Run, Locus B (8LA1-West) 269 Figure 6-49. Photograph showing the plan view of Feature 54/55 at 100 cmbd in TU57, 8LA1W, facing south. of the unit. An additional 30-40 cm of organic leaching extends below this point. Like all other pits associated with DP2, Feature 54/55 was covered with a thick, dense stratum of mostly whole Viviparus and occasional Pomacea shell (DP3). The development of organic soil at the top of the feature suggest s that a substantial amount of time elapsed before the overlying shell was deposited. The fill from Feature 54/55 is co mposed primarily of dense whole Viviparus shell, although smaller amounts of Pomacea and broken bivalve shell are also present. These shell constituents are mixed with dark brow n (10YR3/3) medium sand. This feature is unusual among DP2 pits in that it contains a significant amount of vertebrate fauna, as well as a variety of other cultural materi als including a lithic drill, a bone pin, and frequent plain fiber-tempered ceramics. Th ese materials, along with the shallow crosssection of the pit and the lack of obvious burni ng in the bottom, may re flect an alternative function with regard to other Type 2 features associated w ith DP2, perhaps one related to day-to-day domestic activities. If so, this pit may be the best evidence so far for the residential occupation of Locus B during the Orange period. Bulk samples were collected from vari ous locations through out Feature 54/55 for flotation analysis. The remain ing portions were 1/8-inch water screened or -inch

PAGE 281

270 St. Johns Archaeologi cal Field School 2007-2010: Silver Glen Run screened. An AMS assay from a charcoal sa mple recovered from near the base of the feature provided an age estimate of 3680 40 rcybp (4140-4120 cal BP). Type 3 Features Only one small cylindrical Type 3 feature (Feature 50) has been identified within Locus B deposits, although another probable featur e of this type is visible in the south profile of TU14 but was not recognized as such in the field. Despite being the only representative example, Feature 50 was desi gnated as its own type because, while it is similar in shape to some Type 4 features, it s scale is so vastly different that the two cannot be assumed to have any functional simila rity. It is possible that additional small cylindrical pits were utilized within excavat ed areas of Locus B but were obliterated by the large-scale DP2 pit digging that occu rred during the Orange period. Feature 50 Feature 50 (Figure 650) is a small cylindrical pit that straddles the boundary between TU43 and TU44 within the preceramic deposits of the 2010 block. First encountered at an elevation of 96 cmbd, the pit is almost perfectly circular, measuring 45 cm long and 43 cm wide. It has straight sides and a rounded bottom, exhibiting a maximum depth of 19 cm. Featur e 50 originates from the same organically enriched buried surface from which Features 48 and 49 descend. It is positioned at the westernmost edge of Feature 48 and likely inte rsects the larger pit to some extent. Figure 6-50. Drawings and photograph showing th e plan view and excavated cross-section of Feature 50 from TU43 and TU44, 8LA1W. (Note: photograph not to scale.)

PAGE 282

Silver Glen Run, Locus B (8LA1-West) 271 In the field, Feature 50 was bisected by the east-west line formed by the test unit boundary. The southern half of the feature wa s excavated first in order to expose the pits cross-section. Bulk samples were ta ken from both sides of the feature and the remaining fill was processed through -inch screen. Pit fill is composed virtually exclusively of dense whole Viviparus shell in very dark grayish brown (10YR3/2) sand. The shell was concreted in some places, especial ly near the center of the features base. Although no artifacts were observed during excavation, the pits stratigraphic position and its placement relative to ot her features securely implicat e it as a part of DP1. Few clues were discovered regarding the features function, but it c ould easily have served as a small storage pit. A small charcoal sa mple from the feature returned an AMS radiocarbon assay of 4180 40 rcybp (4840-4580 cal BP), positioning it right on the eve of ceramic use in the region. Type 4 Features A total of 6 pits at Locus B were de signated Type 4 features and another two probable features of this type were subsequently identified in test unit profiles. These features exhibit a cylindrical shape similar to the Type 3 pit but are many times larger, with maximum depths and diameters often exce eding one meter. They do share a number of characteristics with Type 2 pits, the othe r variety of extremely large DP2 features. First, all six Type 4 pits are associated with DP2 and most contain some quantity of Orange plain fiber-tempered pottery but a conspi cuous absence of other material culture. Their contents also vary widely in terms of the quantity and pattern of their shell deposits with some being filled completely with shell from a single species while others contain only trace amounts of various types. Like Ty pe 2 features, Type 4 pits are distributed broadly across Locus B with examples occu rring in both the 2009 and 2010 excavation blocks, as well as in the north-south transect of test units to the east. They are also arranged tightly together and frequently overlap with other pit features. Finally, at least three of the six Type 4 featur es exhibit evidence for burning at their bottoms in the form of orange oxidized sand and charcoal. Togeth er, these parallels suggest similar, or at least interrelated, functions betw een Type 4 and Type 2 features. Feature 26 Feature 26 (Figures 6-51 and 6-52), the first Type 4 feature encountered at Locus B, intercepts the southw est corner of TU22. It was first recognized as a subtle dark brown (10YR3/3) soil stain with low density shell at approximately 126 cmbd, but its actual top is likely higher than this. Although its outline was difficult to make out in plan, it appears to be roughly circular with a diameter of ca. 60 cm. In crosssection, the feature exhibits very slightly in ward sloping sides and a flat bottom, forming a large cylinder more than 65 cm in depth. Th e color of the pits fill grades to a dark yellowish brown (10YR3/4) near its base. The feature runs into the walls of the test unit along its western and southern margins and intersects a similar cylin drical pit (Feature 27) on the east. Feature 26 was bisected along an east-west transect. Two bulk samples were taken from the south half of the pit (one from the top and one from the bottom), while the remaining feature fill was 1/8-inch water scre ened. The fill includes a low to moderate

PAGE 283

272 St. Johns Archaeologi cal Field School 2007-2010: Silver Glen Run Figure 6-51. Drawings of the plan view and excavated cross-section of Feature 26 from TU22, 8LA1W. Figure 6-52. Photograph of the excavated cross-section of Feature 26 from TU22, 8LA1W.

PAGE 284

Silver Glen Run, Locus B (8LA1-West) 273 density of Viviparus shell and occasional bivalve and Pomacea fragments. Beyond this, sparse vertebrate fauna and a small amount of charcoal are the only other cultural materials that were recovered. A charcoal sample from near the bottom of Feature 26 yielded an AMS radiocarbon assay of 3970 40 rcybp (45204300 cal BP), making this the earliest of the large DP 2 pit features yet dated. Feature 27 Feature 27 (Figure 6-53 ) is a large cylindrical pit feature similar to and actually intersecting Feature 26 within TU22. The outline of most of the pit and its Figure 6-53. Drawings and photograph showing the plan view and excavated cross-sections of Feature 27 from TU22, 8LA1W. (Note: photograph not to scale.)

PAGE 285

274 St. Johns Archaeologi cal Field School 2007-2010: Silver Glen Run stratigraphic origin could not be defined due to the pervasive pit-digging that has occurred within the area encompassed by the test unit. Only the east half could be isolated and this was possible only after sectioning the pit along both north-south and east-west axes and carving the intact portion of the feature into multiple pedestaled sections, each at a different level. In th is way, the eastern edge of the feature was recognizable as a dark brown (10YR3/3) so il stain with diffuse margins surrounded by yellowish brown (10YR5/6) medium sand. Feature 27 appears to be similar in both size and shape to Feature 26. Based on the visible portion of the feature, it is roughl y circular to ovoid in outline and exhibits a diameter of greater than one meter. The pit bottoms out at approximately 50 cm below the point at which it was recogni zed but its actual depth may have been as much as twice this. It is flat-bottomed and appears to have steeply sloping sides similar to those noted for Feature 26. Shell was found scattered throughout the pit and included a low to moderate density of whole Viviparus shell along with occasional bivalve and Pomacea A small amount of fiber-tempered pottery was also recovered but no vertebrate fauna was observed. Bulk samples were taken from the top and bottom of the feature for flotation. Feature 33. Feature 33 (Figure 6-54) is a Type 4 feature that straddles the boundary between TU39 and TU41, along the western edge of the 2009 block. The visible eastern half of the feature forms a semicircle with a maximum diameter of approximately one meter and extends 66 cm into the block. It was initially recognized as a discrete concentration of shell at an absolute elevation of 14.755 m but appears in profile to originate from a buried surface, pe rhaps 30 cm above this point. This is the same Orange period surface from which several additional large DP2 pits were dug. The very top of Feature 33 may have been impacted somewhat by the lower extension of a burned root disturban ce (originally designated Feature 29). The feature was bisected al ong an east-west axis running perpendicular to the wall of the block, and the southern half was excavated first. The resulting cross-section reveals a deep, straight-sided, cylindrical pit with four distinct fill strata. The uppermost stratum consists of dense whole Viviparus shell in dark gr ayish brown sand. It is virtually identical to the overlying DP3 shell and may reflect the portion of the pi t that was open at the time that this shell was deposited. Undernea th this is a thin layer of dense and highly concreted whole Viviparus and bivalve shell with occasional whole Pomacea intermixed. This concreted layer sits atop a 25-35-cm stratum of very dark grayish brown sand containing moderate density of whole Viviparus Most of the pit bottom is filled with shell-free medium brown sand. In the south half, this sand is underlain by a thin layer of bright orange oxidized sand and frequent ch arcoal, indicating yet another massive DP2 pit with evidence for thermal alteration. A modest quantity of fiber-tempered plain pottery and a trace amount of vertebrate fauna were found scattered throughout the feature. As noted above with reference to Type 2 features, the stratification present within Feature 33 indicates multiple pitfilling episodes and provides convincing evidence for intentional, structured deposit ional practices by Late Archaic peoples at Locus B.

PAGE 286

Silver Glen Run, Locus B (8LA1-West) 275 Feature 6-54. Drawing showing the plan and cross-section of Feature 33 from TU39 and TU41, 8LA1W. Feature 42 Feature 42 is a large pit feature located in TU39 and TU40 that went undetected until near its base due to the pervas ive overlapping pits that exist in this area of Locus B. It was finally recognized as a discrete feature at an absolute elevation of 13.886 m when a roughly circul ar pocket of orange oxidized sand and charcoal was encountered. These heat altered deposits ar e interspersed with dark brown (10YR3/3) sand across an area measuring approximately one meter in diameter. This has been inferred as a minimum diameter for the pit, although its rim may have been significantly broader. Given the close proxi mity of the features bases, it is likely that Feature 42 intersected Feature 38 at some point near th eir tops. Feature 42 may also have been truncated somewhat on its western edge by additional pit digging. One small fibertempered plain sherd was found among the burned deposits at the bottom of the pit, dating this feature to the Or ange period and likely confirmi ng its association with DP2 activities. Feature 45. Feature 45 (Figure 655) is a large, roughly cylindrical pit located within TU43 and TU45 at the northern edge of the 2010 block. The feature was first recognized in TU43 at approximately 30 cmbd as a definable concentration of particularly dense whole Viviparus shell. Soon after excava tion of the feature began, a relatively large portion of a fiber-tempered incised vessel was encountered protruding out

PAGE 287

276 St. Johns Archaeologi cal Field School 2007-2010: Silver Glen Run Figure 6-55. Drawings of the plan view a nd excavated cross-section of Feature 45 from TU43 and TU45, 8LA1W. of the north wall of the test unit within the upper portion of the pit. As explained earlier in this chapter, TU45, a small L-shaped unit, was then tacked onto the northeast corner of the block to expose the northern half of the feature and make retrieval of the ceramic sherd possible. When the complete top of the feature was exposed, a roughly ovoidshaped pit was revealed measuring 71 cm x 58 cm. Using the line of bisection formed by the test unit boundary, the southern half of the feature was first excavated in its entirety. This exposed an unusual profile with clear compositional differences between the eastern and western si des of the feature. The western half contains light brownish gray (10YR6/2) sand and ash with a moderate density of Viviparus and crushed bivalve shell. Near its bottom, this side of the feature is highly concreted and may contain a concentration of mineralized roots. The eastern half of the feature profile shows a disc rete deposit of dense shell (whole Viviparus with occasional bivalve and Pomacea ) and brown (10YR5/3) sand a pparently intruding into this ashy fill. It thus app ears that Feature 45 may actually include two cylindrical pits, with one having been dug into the other. The larger and earlier pit exhi bits roughly straight vertical walls and measures 70 cm deep at its lowest point. The intrusive pit, on the other hand, displays a similar morphology but tapers slightly at the bottom and is approximately 50 cm deep. The primary func tions are not known for either pit, although the evidence for burning in the earlier one and the corresponding lack of it in the later

PAGE 288

Silver Glen Run, Locus B (8LA1-West) 277 may indicate two different uses. The intrus ive pit is unusual among Locus B features in containing incised fiber-tempered ceramic sher ds, perhaps indicating its association with DP3. Bulk samples were taken from both the no rthern and southern ha lves of the feature for flotation analysis. The remaining featur e fill was processed through a combination of -inch dry screening and 1/8-inch water screening. Feature 51. Feature 51 (Figure 6-56) is an ex tremely large cylindrical pit located in the northwest corner of TU43 in the 2010 block. Although it appears that only about a quarter-section of the feature was exposed, this portion extends more than a meter out from the corner of the test unit. The pit originates from the paleosurface upon which the DP3 shell was initially deposited. The fill within Feature 51 is virtuall y indistinguishable from the overlying shell, consisting of very high density whole Viviparus burned crushed bivalve, and infrequent Pomacea in dark grayish brown (10Y R4/2) sand. In fact, the feature was first recognized as an apparent dip in this sh ell mixture into the underlying, relatively sand-rich stratum. This is perhap s evidence that Feature 51 was still an open pit at the time that the DP3 shell was deposited and was infilled during the course of that event. In the cross-sections provided by the north and west walls of the test unit (see test unit profiles in Figures 6-30 a nd 6-32) Feature 51 can be s een to drop down vertically Figure 6-56. Drawing of the plan view and photogr aph of the excavated cross-sections of Feature 51 from TU43, 8LA1W.

PAGE 289

278 St. Johns Archaeologi cal Field School 2007-2010: Silver Glen Run before rounding off at its base, achievi ng a maximum depth of 92 cm. The exposed portion of the feature appears unimpacted by any of the other f eatures within the excavation block. There is, however, a sizab le pocket of charcoal and burned shell observable in the west profile near the top of the feature, probably a result of a tree root that smoldered in place. Bulk flotation an d 1/8-inch water screen samples were taken strategically from different s ections of the feature while the remaining fill was -inch screened. Type 5 Features Type 5 features from Locus B are define d above as cone shaped pits with broad openings and tapered profiles. Only one Type 5 feature (Feature 38) has thus far been identified at Locus B; however, like Feature 50, the size and morphology displayed by this pit is distinct enough to warrant its own type designation. Feature 38. Feature 38 (Figure 6-57; see also no rth profile of block in Figures 619 and 6-20) straddles the boundary between TU39 and TU40 along the northern edge of the 2009 excavation block. Although only partially exposed, the top of the pit appears roughly circular in shape with an approximate diameter of 120 cm, making it comparable to the large DP2 pits in size. While Feature 38 clearly cuts through the surface on which the DP3 shell was deposited, it is unclear whet her this surface is the origin point from which the pit was dug or whether it originates from some higher point. This ambiguity makes it difficult to determine stratigraphically whether the pit is associated with DP2 or DP3 activities. From this level, the pit pe netrates at least 94 cm into underlying deposits, bottoming out at an absolute elevation of 13.764 m. It narrows with depth, coming almost to a point at its base. As can be seen in the blocks north profile, this feature was dug through at least one and perhaps two large preexisting basin shaped pits. These were not recognized during excavation but were revealed through subsequent profile examination. In addition to its unusual morphology, Featur e 38s fill also exhibits some unique characteristics among Locus B pits. Like a few of the large DP2 pits, Feature 38 is stratified, containing at least th ree distinct layered deposits. The uppermost layer consists of very dark grayish brown (10YR3/2) sand with moderately dense whole and crushed Viviparus This sits atop a 15-20 cm thick laye r of very dense crus hed bivalve and whole Viviparus mixed with somewhat lighter (10YR4/2) sa nd. The base of the pit is filled with dense and highly concreted whole Viviparus and sand. These differe nt strata are clear and distinct in cross-section and the contacts between them are sharp suggesting that they resulted from three planned and discrete filling episodes rather than a hodgepodge of accumulated domestic trash. Small flakes of charcoal recovered from the pits bottom and middle strata yielded respective AM S radiocarbon assays of 3670 40 rcybp (41403890 cal BP) and 3590 40 rcybp (3980-3830 cal BP). Because these estimates overlap within the 2-sigma range, the strata must be considered contemporaneous at the level of precision offered by radiocarbon dating methods The actual temporality of their deposition, however, be it on the order of hours, days, or even years, is unknown.

PAGE 290

Silver Glen Run, Locus B (8LA1-West) 279 Figure 6-57. Drawings and photograph showing the plan view and excavated cross-section of Feature 38 from TU39 and TU40, 8LA1W. (Note: photograph not to scale.) Regardless, Feature 38 provides yet further eviden ce that the infilling of at least some of Locus Bs abnormally large pits was an intentional, structured affair. Feature 38 is also one of only two Locus B pit features to c ontain incised fibertempered pottery and the only one with Ti ck Island style incisions and punctations. While this ceramic variety is rare in feature contexts, it occurs relatively frequently within the expansive shell cap constituting DP3. Given Feature 38s somewhat uncertain stratigraphic origin in relation to the la rge DP2 roasting pits and relatively late radiocarbon dates, it is possible that this feature is either one of the last DP2 pits to be dug and utilized at Locus B or th at it reflects one of the earl iest events associated with

PAGE 291

280 St. Johns Archaeologi cal Field School 2007-2010: Silver Glen Run DP3 activities. The primary function of the pi t remains unclear, as it displays an unusual morphology and does not exhibit the obvious ev idence of burning ob servable in other large Orange period pits at Locus B. Type 6 Features Two Type 6 features have been recorded at Locus B. These are defined as discrete, isolated pockets of shell located in the otherwise sh ell-free sand underlying Locus Bs large, expansive cultural deposits. These features are assumed to mark pits or perhaps the basal portions of pits, but this interpretation is tenuous due to the fact that neither of the examples appears to be connect ed to an apparent paleosurface or other cultural deposits. It is also possible that these shell concentrations are simply animal burrows that were filled in with overlying shell, although no burrow passages or collapses have been identified connecting them to upper shell deposits. Both Type 6 features are located near the center of Locus B and were en countered in 1 x 2-m test units making up the north-south transect bisecting this part of the site. If cultural, they are almost certainly associated with DP1 and may constitu te yet another type of domestic feature associated with the late preceramic Thornhi ll Lake Phase habitation of Locus B. Feature 17 Feature 17 (Figure 6-58) is an isolated concentration of concreted shell and sand extending out of the wester n wall of TU12. The exposed portion of the feature is amorphous in shape and measur es 47 cm in length, although its complete dimensions and configuration are unknown. It is composed of whole and crushed Viviparus, Pomacea and bivalve mixed with brown (7.5YR5/4) medium sand. In profile, the feature shows up as a 15-cm th ick pocket of shell surrounded by virtually shell-free sand. It is positioned approximately 25-30 cm beneath a shell-rich stratum but there is no indication in the profile that the f eature originates from or is connected in any way to this shell. No artifacts were recovered from the feature and its function remains unclear, although it may be a small pit originating from a deeply buried surface imperceptible in the profile of the test unit. Feature 25. Similar in many ways to Feature 17, Feature 25 is an isolated pocket of concreted shell jutting out of the north wall of TU21. The feature appears roughly circular or ovoid in shape and exhibits a maximum exposed diameter of 38 cm and a maximum thickness of 28 cm. It is composed primarily of concreted whole Viviparus shell but Pomacea is also frequent, especially along its base. Like Feature 17, this shell concentration is surrounded on a ll sides by shell-free sand; howe ver, in this case, a dense shell stratum is located only 5-10 cm above it. There is, nonetheless, nothing connecting the feature to the overlying shell and its function is unknown. Non-cultural and Modern Features Nine anomalies were encountered during Locus B excavations that were initially designated cultural features and assigned feature numbers but were subsequently downgraded. Seven of these were eventually recognized as natural disturbances, while two were determined to have resulted from recent modern activities. Features 2, 3, 5, and

PAGE 292

Silver Glen Run, Locus B (8LA1-West) 281 Figure 6-58. Drawing and photograph of the plan view of Feature 17 from TU39, 8LA1W. (Note: photograph not to scale.) 13 were all documented during the first year of Locus B excavation within TU4. They were originally recorded as possible post mo lds but, as experience with these phenomena accumulated, were ultimately determined to be mineralized root casts whose size and vertical orientation mimic those usually disp layed by architectural supports. Similarly, Features 29, 30, and 31, were initially reco rded as potential cooking hearths. These features all display a similar morphology and structure consisting of a gray ovoid deposit of burned shell, ash, and charcoal ringed by orange heat-oxidized shell (see Figure 6-60).

PAGE 293

282 St. Johns Archaeologi cal Field School 2007-2010: Silver Glen Run Feature 6-59. Drawing of the plan view of Feature 25 in TU21, 8LA1W. Figure 6-60. Photographs showing the a) plan view, and b) excavated cross section of Feature 29 (downgraded) in TU39, 8LA1W. In profile, however, these burned anomalies tended to taper to a point and turn unpredictably, suggesting that they are actually a result of tree roots that penetrated shell deposits and burned in place. In fact, Featur e 29 actually had a large charred root running through its bottom, further solidifying this secondary inte rpretation. In 2009, an unusual subrectangular feature (Feature 39) with sharp margins that connect at right angles was encountered in TU39 but was quickly recognized as a shovel test pit dug by field school students in 2007. And finally, excavation of TU57s upper deposits revealed an almost perfectly circular dark soil st ain that persisted for approxima tely 30 cm before rounding off at its base. The features sharp margins, along with the large am ount of apparently recent charcoal that it contains, led to its te ntative interpretation as a modern disturbance of some kind, although its pr ecise source is unknown.

PAGE 294

Silver Glen Run, Locus B (8LA1-West) 283 Summary and Discussion of Locus B Feature Assemblage Excavations at Locus B have uncovered a diverse assemblage of pit features that can be categorized into si x basic types based on size and overall morphology. The horizontal and vertical distribution of these pit types has in part led to the identification of three distinct patterns of shell deposition at Locus B (discussed in the previous section), each of which corresponds to a fundamentally different use of this space during a particular time in the sites occupational history. The specific characteristics of individual pits and pit types offer clues as to the nature of these contrasting depositional patterns and their relationship to coterminous places throughout the region. DP1 includes a wide variety of features including Type 1 small basins, Type 2 large basins, Type 3 small cylinders, and Type 6 isolated shell pockets. In all likelihood, this diversity of feature shapes and sizes reflects the diverse functions of pits involved in the many activities associated with sustained everyday living during the late preceramic period at Locus B. This interpretation is supported by the relative abundance of diverse tool types and vertebrate faunal remains with in and surrounding DP1 pi ts. All of the DP1 pits with discernable origin points descend fr om a crushed shell surface near the base of Locus Bs shell deposits. Given the stacked sequence of such surfaces found at the base of TU46, a series of successive preceramic occupations are most likely responsible for these features. Although prec ise functions are difficult to assign, the shapes, sizes, and contents of the features sugge st their use in cooking and st orage activities related to relatively small-scale domestic food producti on. The often heat-altered and concreted shell within most DP1 pits poi nts to their probable role in the processing of shellfish. DP1 pits from Locus B are consistent in thes e respects with features associated with the slightly earlier Mount Taylor habitation co mponent at 8LA1Ws Locus A (Chapter 5 of this report), as well as those from roughly coterminous deposits at the nearby sites of Hontoon Island North (8VO202) (Sassaman et al. 2005), Blue Spring Midden B (8VO43) (Sassaman 2003b), and Thornhill Lake (8VO60) (Endonino 2010). These preceramic features are restricted to the s outheastern quadrant of Locus B. The most striking contrast between DP1 and DP2 involves the scale and frequency of pit digging across Locus B. Th is transition entailed a shift from the everyday use of a relatively small number of highly diverse pits to the repeated digging and infilling of countless remarkably large fe atures across an expansive area of 8LA1W. The DP2 feature assemblage, while including a handful of Type 1 shallow basins, is dominated by massive Type 2 basins and Type 4 cylinders, several of which exceed 1 m in diameter and/or depth. Calculating precis e diameters for most of these features is complicated because their size virtually ensu res that they will intersect test unit walls unless large blocks are excavated. Nevertheless for two of the pits that appear to most closely approach true cylinders, Features 51 and 36, respective volumes can be estimated using the formula to be 1500+ L and 800+ L, fa r beyond what would be expected if these were intended simply to address the short-term subsistence requirements of small kin-based groups.

PAGE 295

284 St. Johns Archaeologi cal Field School 2007-2010: Silver Glen Run Excavations at Locus B have currently ex tended to a sufficiently large area that some projections can be made as to the overall scale of pit digging and shellfish processing that took place there. Across the area highlighted in gray in Figure 6-61, 31 m2 have been excavated within which 14 large DP2 pits were documented. Assuming that pits are present in the intervening areas between excavation units, projecting this pit density across the entire area encompassed by their known distribu tion results in an estimate of more than 310 massive pits at Locu s B. Due to the fact that some pits could not be discretely defined because of their ov erlapping distribution and that these features are likely to extend beyond current excavat ion boundaries, this es timate should be considered conservative. The contents of the pits may offer some cl ues as to the nature of the activities in which they were involved. As noted above, th e bases of many, although not all, DP2 pits are marked by heat oxidized sand, charcoal, an d lenses of burned, concreted bivalve. This, along with the conspicuous absence of ve rtebrate fauna and artifacts aside from a small amount of fiber-tempered pottery, sugge st a specialized function for these massive features related to sh ellfish processing. Ag ain, the scale of the pi ts, and correspondingly the quantity of shellfish they could have pr oduced, is out of proportion with the everyday domestic needs of a small group. If looked at in the broader context of the Silver Glen Run complex, the Locus B pits may instead be speculated to have been a means for quickly producing large amounts of food needed for consumption at the large scale social gatherings hypothesized to have taken place at 8LA1Es U-shaped monument, less than half a kilometer to the east. The fact that new pits were frequently dug through preexisting ones does suggest intermittent pulses of intense pit-centered activities rather than long-term sustained o ccupation during this time. Figure 6-61. Map showing test units containing T ype 2 and Type 4 pits associated with DP2 and the bounded area used to project the total number of these pits at Locus B. 686.84 m2

PAGE 296

Silver Glen Run, Locus B (8LA1-West) 285 Subsequently, the pits were infilled with various combina tions of sand and shell, a process that at least occasionally involved multiple structured and intentional depositional acts. The resulting elaborately stratified f eature cross-sections provide strong evidence that although many of these features may have begun as roasting pits their significance extended beyond mere practical f unctionality. In fact, as time went by and pit excavation persisted, the encountering of old infilled pits must have become the expected outcome of, and perhaps even added motivation for, continued digging. Under these circumstances, sequences of pit fill may have come to be viewed by the Orange period inhabitants of Locus B similarly to the laye ring of shell above ground to form mounds, as a historical practice through which the past was accessed and the future could be anticipated ( sensu Sassaman 2010, 2012). DP2 thus signa ls a major intensification of shellfish production at Locus B and perhaps a fundamental rework ing of the way in which people accessed and related to their past. DP3, for the most part at least, signals th e end of large-scale pit digging at Locus B. Its historical connection to DP2, however, is unquestionable given that the distribution of DP3 shell seems to have been mapped ont o the area containing DP2 pits. In some instances, it even appears to have filled in large open pits and evened out the Locus B surface. As noted above, shell capping ev ents have been observed covering Mount Taylor domestic occupations and have been in terpreted as signaling the death of these localities as places of inhab itation (Sassaman 2010:72). In this case, the capping of the DP2 pits may have similarly marked the death of these features or of the entire area as a place of ritualized shell processing and deposition. ARTIFACT ASSEMBLAGE A total of 3704 artifacts were recovered fr om level excavations of Locus B test units during the 2007-2010 field seasons. This number does not take account material culture from features whose samples were not completely analyzed as of the writing of this report, nor does it include unmodified vertebrate fauna or marine shell. A large majority of all artifact types were recovere d from Late Archaic Thornhill Lake Phase and Orange period contexts, although later preh istoric and historic objects are also represented. Results of the preliminary anal ysis of the Locus B ceramic assemblage and descriptions of all other artifact classes are provided below. A summary of all artifacts recovered by test unit is shown in Table 6-25. Pottery Locus B excavations produced a total assemblage of 889 ceramic sherds, excluding crumb sherds. Of these, 603 have been classified as Or ange, 281 as St. Johns, and five as generic sand-tempered potter y. While Orange sherds are found throughout the strata comprising DP2 and DP3 and in all Locus B test units, the other two types are largely restricted to the plow zone in this area 8LA1W. Consequently, their assemblages are highly fragmented and are assumed to have been displaced from their original depositional contexts in most cases. As a result, the pottery analysis that follows is

PAGE 297

286 St. Johns Archaeologi cal Field School 2007-2010: Silver Glen Run Table 6-25. Summary of Artifacts Recovere d by Test Unit from Locus B, 8LA1W.

PAGE 298

Silver Glen Run, Locus B (8LA1-West) 287 focused largely on the Orange pottery with brief descriptions offered for the St. Johns and sand-tempered varieties. Orange Fiber-Tempered Pottery. Orange pottery, a low-fired earthenware defined by its distinctive temper consisting of Spanish moss and possibly palmetto fibers (Brain and Peterson 1971; Simpkins and Allard 1986), is the earliest pottery technology in Florida. First appearing along the states northern Atlant ic coast by at least 4200 rcybp (Russo 1993), it rapidly spread inland to the nearby St. Johns River Valley most likely via preexisting exchange netw orks (Sassaman 2003a, 2004). Ora nge pottery is frequently divided into three basic varieties based primarily on surface treatment (Griffin 1945; Bullen 1972; Milanich 1994): 1) Orange Incised, exhibiting primarily straight rectilinear incisions and occasional tick marks; 2) Tick Island Incised displayi ng curvilinear spiral shaped incisions and round punctations; and 3) Orange Plain. Research over the past decade suggests that Orange Incised pottery occurs in the highest proportions at shell rings along the Atlantic Coast and at larg e shell mounds within the St. Johns River Valley, while Orange Plain vessels are most frequent at smaller non-mounded sites in both areas (Sassaman 2004; Saunders 2004a, 2004b). Little is known about the spatial distribution of Tick Isla nd Incised pottery other than it appe ars to be largely restricted to a relatively small area within the Middle St. Johns River Valley (Griffin 1945). In the middle part of the last century, Bullen (1955, 1972) argued that formal and stylistic differences in Orange pottery was primarily chronological and devised a ceramic chronology with five distinct periods based on this variation. The general trends noted by Bullen were a progression from plain pottery in early Orange times to decorated sherds in later periods, greater diversity in lip form through time, increased vessel size and wall thickness through time, and a movement from exclusively fiber-tempered pots to ones also containing sand and sponge spicules. Recent research, however (Cordell 2004; Sassaman 2003a; Saunders 2004a), effectively de monstrates that th is entire range of variation actually overlaps temporally and should instead be evaluated with an eye toward concurrent functional and/or ethnic diversity. Vessel Lots A vessel unit analysis was conduc ted for Locus Bs fiber-tempered pottery. Sherds were separated into vessel lots using a variety of criteria including surface treatment, temper, rim form and thickness, orifice diameter, and basic sherd shape (e.g., straight or curved). Provenience was not taken into account in assigning sherds to vessel lots, as all excavated proveniences at Locus B are within 50 m of each other and could conceivably contai n sherds from a single vessel. Based on these considerations, a minimum of 98 fiber-tempered vessels are repr esented in the Locus B assemblage. These vessels were evaluated with regard to a number of stylistic and technological variables. Basic morphological and metric data associated with each of these variables are listed in Table 6-26. Surface Treatment Of the 98 vessels, 50 (51.0 percent) are plain, 33 (33.7 percent) have rectilinear incisions associat ed with the Orange Incised variety, 14 (14.3 percent) exhibit Tick Island style curvilinea r incisions and/or punctations, and one vessel (1.0 percent) has a surface too eroded to de termine surface treatment. Examples showing

PAGE 299

288 St. Johns Archaeologi cal Field School 2007-2010: Silver Glen Run Table 6-26. Data on Orange Fiber-Tempere d Vessel Lots from Locus B, 8LA1W. V # # Sherds Temper1 Surface Treatment2 Lip Form3 Lip Thk. (mm) Rim Form4 Rim Thk. (mm) Orifice Diam. (cm) 1 5 FAS INP RD 5.1 ST 9.3 28 2 17 FAS INR XF 3.9 22 3 2 FAN INC 4 7 FAS INP RD 7.6 5 2 FMS PL RD 4.3 6 2 FAS INC 7 1 FAS INR 8 1 FAS/S INR 9 1 FAS/S INR 10 4 FAS INR RE 3.0 ST 11 1 FAS PUN 12 1 FAS/S INR 13 1 FAS INR 14 1 FAS/S INR RD 4.8 IN 6.6 15 2 FAN INR RE 8.6 26 16 14 FAS INP RE 8.7 IN 8.9 30 17 1 FAN/S INR RE 6.4 14 18 2 FAS INR RE 6.4 19 1 FAS INR RI 6.7 ST 7.5 20 1 FAS/S INR RE 21 3 FAS INR RD 4.5 ST 24 22 1 FAS INP RD 6.0 ST 8.9 23 1 FAS INP XF/D 17.0 ST 8.8 30 24 1 FAS INC 25 3 FAS/S PL RE/T 6.6 ST 5.3 26 17 FAS/S PL RE 5.2 ST 8.3 27 1 FAS PL FI 6.4 ST 8.0 28 2 FAS PL RD 4.6 29 1 FAS PL XF 4.6 30 2 FAS PL FI 3.8 ST 7.0 31 2 FAS PL BE 4.0 IN 9.0 32 1 FAS PL PR 2.2 33 5 FAS PL XF 5.2 ST 6.2 34 34 1 FAS PL RE 2.9 8 35 1 FAS PL PR 2.1 ST 36 1 FAS PL RE 5.2 ST 37 1 FAS PL RI 4.6 EX 6 38 1 FAS PL BE 3.9 39 1 FAS PL BE 5.5 40 2 FAS PL BI 4.9 ST 41 1 FAS PL RE 3.8 42 1 FAS/S PL RE 3.6 6 43 2 FAS/S PL RE 5.0 44 1 FAS PL RD 5.7 16 45 2 FAS/S PL FI 4.7 IN 9.7 46 1 FAS/S PL RE 4.2 IN 47 1 FAS PL RE 4.2 48 1 FAS PL BE/T 7.8 49 1 FAS PL XF/T 7.5 ST 5.9 50 1 FAS/S PL FD/T 51 1 FAS PL XF 4.5 52 1 FAS PL RE 3.3 IN 53 2 FAS PL FI 4.7 ST 7.4 20 54 2 FAS PL FI 5.7 IN 9.8 36 55 1 FAS PL BV 4.6 IN 56 1 FAS PL FI/T 6.0 (continued on next page)

PAGE 300

Silver Glen Run, Locus B (8LA1-West) 289 Table 6-26. (continued) V # # Sherds Temper Surface Treatment Lip Form Lip Thk. (mm) Rim Form Rim Thk. (mm) Orifice Diam. (cm) 57 3 FAS/S PL RD 5.6 ST 7.3 16 58 1 FAS/S PL RE 8.0 59 2 FAS PL FI 2.5 ST 3.9 10 60 1 FAS PL RD 5.3 ST 7.3 61 1 FAS PL FI/T 9.1 IN 6.1 62 1 FAS PL BE 6.4 63 4 FAS PL PR 3.9 IN 7.2 64 4 FAS/S PL RE 4.9 65 3 FAS/S PL XF 6.0 20 66 2 FAS PL XF 5.3 12 67 1 FAS PL RD 4.8 68 2 FAS/S PL RD 4.3 69 2 FAS/S PL RE 5.7 70 1 FAS INR RD 6.0 IN 18 71 1 FAS INR 72 1 FAN/S INC 73 3 FAS INC 74 1 FMS/S PUN 75 1 FAS/S INR 76 1 FAS INR XF 6.2 ST 77 3 FAS PL BE 6.0 ST 6.7 78 1 FAS/S PL FE/T 17.9 79 2 FAS/S INR XF 4.4 ST 80 1 FAS/S INR BE 4.9 ST 8.7 81 2 FAS INR 82 7 FAS INR XF 7.5 ST 7.8 16 83 1 FAS INR PR IN 8.1 84 1 FAS INR RE ST 85 1 FAS/S INR 86 1 FAS/S INR 87 1 FAS INR XF 5.4 88 3 FAS/S INC RD 7.5 IN 24 89 1 FAS INR 90 1 FAS/S INR RD 5.3 ST 10 91 3 FAS INC 92 4 FAS/S INR FI/D 10.2 IN 7.6 93 2 FAS/S ER RD 6.5 ST 94 2 FAS INR 95 1 FAS PL RE 3.3 ST 96 2 FAS PL RE 3.4 97 4 FAS INR RD 7.6 IN 7.1 98 17 FAS/S INR BE 9.1 ST 11.4 30 1Temper categories: FAN fiber abundant, no visible aplastics; FAS fiber abundant, visible aplastics; FMS fiber trace, visible apla stics; /S suffix indicating presence of sponge spicules 2Surface Treatment categories: INR incised rectilinear; INC incised curvilinear; INP incised and punctuated; PUN punctuated; PL plain; ER eroded 3Lip Form categories: XF flat; RD rounded; RI rounde d interior; RE rounded exterior; PR tapered; BV beveled; BI beveled interior; BE beveled exterior; FI flanged interior; FE flanged exterior; /T suffix added if thickened; /D suffix added if decorated 4Rim Form categories: ST straight; IN incurvate; EX excurvate

PAGE 301

290 St. Johns Archaeologi cal Field School 2007-2010: Silver Glen Run Figure 6-62. Examples of Orange Incised fiber-tempered vessels from Locus B, 8LA1W.

PAGE 302

Silver Glen Run, Locus B (8LA1-West) 291 Figure 6-63. Examples of: a-i) Tick Island Incised vessels; and j) a fiber-tempered vessel with an unusual motif featuring elongated ticks.

PAGE 303

292 St. Johns Archaeologi cal Field School 2007-2010: Silver Glen Run the range of variation within the incised types are provided in Fi gures 6-62 and 6-63. Locus Bs Orange Incised vessels display a diversity of motifs, a majority of which incorporate some combination of oblique and/or horizontal lines. Zoning is apparent in only a few examples, although the small vessel portions available in most cases may obscure larger zoned patterns. Two vessels have cross-hatched designs and seven have nested chevrons, with one of these, vessel 98 (Figure 6-62h), displaying a pattern of upward and downward pointing chevrons that cross each other on their margins. Tick Island motifs from Locus B are less variable and consist primarily of large spiral-shaped incisions surrounded by punctations. On all four of the Tick Island vessels for which a substantial portion of the rim is present, two straight horizontal in cisions line the rim. Two of the Tick Island vessels also have paired vertical lines that divide their surfaces into distinct zones. Vessel 9 (Figure 6-63, c) is the only Locu s B vessel that shows curvilinear incisions in th e absence of punctations. In all but two cases, surface decorations are restricted to the exterior su rfaces of vessels, the exceptions being Vessels 23 and 92, which both have incised lips. Execution of the incisions is also variable with some exhibiti ng perfectly straight, evenly spaced lines while others are highl y irregular. With only a few exceptions, both the rectilinear and curvilinear designs on Locus B pots consist of broad deep lines, apparently incised with a flat-tipped or r ounded stylus. Many incisions exceed 2.5 cm in width. Punctation diameters vary from 2-4.3 mm. The narrow ri dges of clay built up on the edges of most incisions and punctations su ggest that they were applied while the clay was still wet. Temper A small fresh break was made on sher ds from each vessel lot in order to evaluate the constituents of their added temper(s). Sherds were examined under a stereoscopic microscope at 40X magnification. All but 2 of the 98 vessel lots were determined to contain abundant fiber as well as visible ap lastic inclusions, primarily quartz sand. Of these, 32 (32.7 percent) also contain freshwater sponge spicules, the defining characteristic of St. Johns type po ttery but also a common constituent of some Orange assemblages in th e Middle St. Johns River Va lley (Cordell 2004; Sassaman 2003a). At this point, it is still unclear the ex tent to which spicules were intentionally added to clay as a temper or were simply natural constituents of local clays (Borremans and Shaak 1986; Cordell 2004; Cordell and Koski 2003; Rolland and Bond 2003). Interestingly, unlike other sites in the re gion, spiculate pastes within the Locus B assemblage do not appear correlate with surface treatment (see Cordell 2004; Sassaman 2003b), as at least 14 (43.8 percent) of the vessels with spicules are plain. Of the remaining vessels with spiculate pastes, 14 (43.8 percent) have rectil inear incisions, three (9.4 percent) have curvilinear incisions an d/or punctations, and one (3.1 percent) is eroded. Morphology and Size Morphological and metric data were gathered for each vessel with regard to a number of variables including li p form, lip thickness, rim form, rim thickness, orifice diameter, and basic vess el shape (round or re ctangular). Lip and rim form were coded using the system deve loped by Sassaman (1993) for fiber-tempered pottery from Georgia and more recently utilized in the analys is of Orange pottery from

PAGE 304

Silver Glen Run, Locus B (8LA1-West) 293 Blue Spring Midden B (Sassama n 2003b). Rim thickness was measured at a distance of 3 cm from the lip. Orifice diameter was estim ated using a standardized rim chart for all vessels with five percent or more of the rim present. In addition, rim profiles were drawn for all vessels with at least 3 cm of the rim present (see Figures 6-64 and 6-65). Figure 6-64. Profiles of rim portions of select Orange Plain vessels from Locus B, 8LA1W.

PAGE 305

294 St. Johns Archaeologi cal Field School 2007-2010: Silver Glen Run Figure 6-65. Profiles of rim portions of select Orange Incised and Tick Island Incised vessels from Locus B, 8LA1W. A broad diversity of lip forms are repr esented among Locus B Orange vessel lots, likely bespeaking the lack of standardization present during this period that was also

PAGE 306

Silver Glen Run, Locus B (8LA1-West) 295 noted by Sassaman (2003b:115) at Blue Spring Midden B. A majority of rims are straight and outward sloping, resulting in broad open vessel profiles, although a handful of rims are incurvate near the lip and must have had somewhat restricted openings. Rim thickness is also highly variable, ranging from 3.9-11.4 mm, as is estimated orifice diameter with a range of 6-36 cm. There is a strong positive co rrelation between rim thickness and vessel size as estimated by orif ice diameter (Pearson correlation coefficient = 0.642), suggesting that thicker walls may have been necessary to prevent larger vessels from collapsing in on themselves during th e manufacturing process (Espenshade 1983). There also appears to be some connection be tween vessel size and surface treatment, as the mean diameters are 16.73 cm for plain vess el lots and 22.67 cm for incised vessels, although this difference is not st atistically significant (t va lue = -1.644; probability = 0.115). Only a few different vessel types can be inferred from the Locus B assemblage. A large majority of vessel lots for which vesse l shape can be surmised appear to be round or oval bowls with straight or slightly incurving rims and flat bottoms. Two of the largest reconstructible vessel portions from Locus B bo th conform to this description. Vessel 26 (Figure 6-66) is an Orange Pl ain roughly circular bowl with thin walls that round into a flat base. It measures 11.3 cm in height and has an 8.3-mm thick rim gauged at 3 cm below its flattened lip. Its ex terior surface is extremely fria ble and it shows significant thermal attrition on its base. Vessel 16 (Figur e 6-67) is a large, elaborately decorated Tick Island Incised bowl with thick, incurving walls and a lug handle projecting out of its rim. The vessels rim measures 8.9 cm in thickness and it has an estimated orifice diameter of 30 cm. Rim sherds from six ve ssel lots show no apparent horizontal or vertical curvature and probably come from rectangular vessels. Of these, four are plain, one has rectilinear incisi ons, and one is eroded. Chronology The stratigraphic distribution of the different types of fibertempered pottery at Locus B provides important clues as the chronol ogical relationships between them. These relationships, while repe ated within virtually all excavated Locus B contexts, are perhaps most clearly visible in the pottery data from the 2009 excavation block. Figure 6-68 is a backplot showing th e vertical locations of all piece-plotted fibertempered sherds in front of the north pr ofile of the 2009 block. What this figure demonstrates is that the lowermost and hence earliest pot sherds at Locus B are primarily plain, while decorated sherds (including both Orange Incise d and Tick Island incised) arrive on the scene relatively late and are asso ciated with the DP3 shell layer capping the areas massive pits. The only exceptions to this pattern are a few fiber-tempered incised sherds visible near the center of the profile in Figure 6-68 that were all recovered from Feature 38. This feature is ex ceptional in that it is the onl y large Locus B pit to contain incised pottery, including Tick Island Incised sherds. This pattern is corroborated by extensive radiocarbon data th at date pits containing plai n pottery to between 3970 40 and 3590 40 rcybp and the one pit containing incised pottery to the very end of this range at 3590 40 rcybp (see complete radioc arbon data in Appendix B of this report). This sequence of plain fiber-tempered pottery followed by incised vessels with Tick Island designs is the same one observed by Bullen (1955, 1976; Bullen and

PAGE 307

296 St. Johns Archaeologi cal Field School 2007-2010: Silver Glen Run Figure 6-66. Partially reconstructed Orange Plain vessel from Locus B, 8LA1W. Figure 6-67. Partially reconstructed Tick Island Incised vessel from Locus B, 8LA1W.

PAGE 308

Silver Glen Run, Locus B (8LA1-West) 297 Figure 6-68. Backplot showing vertical stratigra phic distribution of Orange/Tick Island Incised vessels and Orange Plain vessels in the 2009 excavation block at Locus B, 8LA1W. Stoltman 1972) at the Bluffton (8VO22) and Pa lmer (8SO2A) sites that largely provided the basis for his Orange 1 and Orange 2 subperiods. In Bullens chronology, these pottery types were thought to be characteristic of the Orange Periods earliest manifestations. He argued that these were subsequently replaced by pots with more elaborate straight-line motifs in the Orange 3 subperiod, whic h were in turn succeeded by vessels with simpler designs and chalky pa stes by Orange 4. As noted above, this chronology has been largely invalidated by subs equent research that has demonstrated contemporaneity between these different st yles of Orange pottery (Cordell 2004; Sassaman 2003a; Saunders 2004a). Data from Lo cus B further corroborates this research by demonstrating the use of Orange 1 and Or ange 2 vessels during a period hundreds of years later than dates obtained by Sassaman ( 2003a) for Orange 3 and Orange 4 vessels at the Mouth of Silver Glen Run (8LA1E) and other sites in the regi on. Importantly, Locus B research has also resulted in the first s ecurely dated context containing Tick Island Incised pottery. The date of 3590 40 rcybp (3980-3830 cal BP) for F eature 38 contrasts with Bullens chronology by firmly placing this relatively rare variety of fiber-tempered pottery late in the Orange Period, centuries after the initial appearance of the classic Orange Incised variety. Comparisons to Orange In cised Pottery from 8LA1E In 2009, a technofunctional analysis was conducted of 146 vessel lots co llected from the north ridge of the 8LA1E mound as a part of a class project (Gilmor e 2009). Radiocarbon assays obtained from soot on sherds from this massive U-shaped monument range from 4070 40 to 3680 60 rcybp (Sassaman 2003a) and thus overlap with Orange Period dates from Locus B, indicating that use of the respective fiber-t empered pottery assemblages from the two areas was coeval. With this in mind, there are a number of interesting stylistic and technological differences appare nt between the two contexts.

PAGE 309

298 St. Johns Archaeologi cal Field School 2007-2010: Silver Glen Run As at Locus B, a majority of the 8LA1E pots for which vessel type can be inferred are shallow open bowls, perhaps best suited as serving containers. However, whereas just under half (49 percent) of the vessel lo ts from Locus B have some kind of surface decoration, more than 75 percent of those an alyzed from 8LA1E are decorated. This pattern parallels the one noted by Saunders (2004a) between shell rings and contemporary non-mounded sites along the Florid as Atlantic coast but occurs instead as an intrasite arrangement at Silver Glen Run. Comparing Locus Bs Orange Plain vessels to the incised pots from the mound also reveals a number of additional differences beyond those related to surface treatment. First, a large majority of north ridge vessel lo ts have chalky, spicul ate-rich pastes, while spicules were observed in less than one third of Locus Bs plain vessel lots. These plain vessels are also smaller and thinner-walled than their incised counterparts from the mound with an average orifice diameter of 16.7 cm and rim thickness of 7.2 mm compared to 25.8 cm and 10.0 mm for the 8LA1E pots. Other differences relate to the manner in which the pots were actually utilize d. While heat attrition is observable on the exteriors of vessels from both assemblages, only vessels from 8LA1E have soot preserved on their surfaces, perhaps indicating that they were suspended over open flames while Locus B pots were placed direct ly onto smoldering coals. In addition, while six 8LA1E vessels have holes drilled into them for mending, no Locus B vessels show evidence for the repair of broken vessels. Interestingly, the decorate d vessels from Locus B bear more technological and stylistic resemblance to the plain pots from the same location than they do the Orange Incised vessels from the 8LA1E mound. First, while the rectilinea r motifs between the two areas are similar, a much higher proporti on (29.8 percent as compared to 7.1 percent) of Locus B vessels exhibit the curvilinear incisions and/or punctations characteristic of Tick Island style potte ry. Importantly, the rectilinear incisions on Locus B pots, which form similar overall motifs to the Orange Incised vessels from 8LA1E, more closely resemble the thick and sometimes irregular Tick Island incisions than they do the mostly thin, precise lines on their counterparts from the 8LA1E shell mound. The 8LA1E designs also lack the thin ridges of clay lining the margins of incisions, perhaps indicating that their lines were applied at a late r stage in the drying process. In terms of size and thickness, the decorated Locus B vessel s are actually intermediate between that areas plain pottery and the incised pots from the mound with an average rim thickness of 8.4 mm and orifice diam eter of 22.1 cm. Given that these intrasite differences in fiber-tempered pottery can no longer be attributed to simple chronological su ccession, other factors including functional specialization and/or ethnic di versity must be considered (S assaman 2003a). Differences between Locus B pottery and that from the mound at 8LA1E may be explained in terms of the social scale of the respective events orchestrated at these two areas. The relative size of vessels is commonly used to deduce th e size of the groups being served (e.g., Blitz 1993; Mills 1989) and can also inform on the physical distance between the pots and the people intended to view them (Mills 2007). Moreover, stylistic elements of pots, including surface decorations, often encode messages regarding group affiliation

PAGE 310

Silver Glen Run, Locus B (8LA1-West) 299 (Pikirayi 2007; Weissner 1983; Wobst 1977) or status (Hayden 1995; Russo 2004; Russo et al. 2002). These messages are recognized by both sender and receiver and become increasingly important as interaction occurs between groups that are more socially distant (Wobst 2007). Although the relationship betw een style and identity are likely more complex than generally portrayed (Gossela in 1998; Hegmon 1992; Stark et al. 2000), high frequencies or exaggerated forms of potte ry surface decorations may be indicative of their use in socially diverse contexts. With this in mind, the large, elaborately decorated vessels from 8LA1E suggest relatively largescale and possibly multi-ethnic social events in line with the regional-scale feasts s uggested by some to have occurred at Orange period shell rings (Russo 2004; Russo et al. 2002; Saunders 2004a, 2004b). The smaller and predominantly plain vessels associated with the huge roasti ng pits at Locus B, on the other hand, are probably indica tive of more socially restri cted, although not necessarily less ritually charged, activities. Given their close spatial proximity and contemporaneous use, these two places, in all likelihood, functione d in concert as coordinating parts of the same ritual landscape. The seemingly sudde n appearance of Tick Island Incised vessels late in the sites Orange period history must signal a significant transformation in the social conditions surrounding Locus Bs use, although the ci rcumstances surrounding that shift and its larger scale significan ce are not currently well understood. Post-Archaic Pottery. The remaining pottery is composed entirely of postArchaic varieties with most be longing to the St. Johns sponge -spiculate tradition (Figure 6-69). Not including crumb sherds, a tota l of 281 St. Johns pottery sherds were recovered, including three St. Johns Incise d, 240 St. Johns Plain, 30 St. Johns Check Stamped, and eight eroded. St. Johns Incise d pottery is diagnostic of the St. Johns I subperiod (ca. 2800-1300 cal BP) while St. Johns Check Stamped is characteristic of St. Johns II (ca. 1300-500 cal BP) (Mil anich 1994:247). By far the most numerous variety at Locus B, St. Johns Plain, is found throughout both subperiods. Unfortunately, as noted above, the St. Johns component at Locus B is contained almost entirely within the plow Figure 6-69. Examples of: a) St. Johns Check Stamped; and b) St. Johns Incised sherds from Locus B, 8LA1W.

PAGE 311

300 St. Johns Archaeologi cal Field School 2007-2010: Silver Glen Run zone, resulting in small fragmentary sherds that largely prohibit any meaningful, contextdependent interpretations. The only other pottery type identified is sand-tempered plain, of which 6 undecorated sherds were recovered. These were found in the same stratigraphic contexts as the St. Johns pottery and could date to any post-Archaic period. Flaked Stone A total of 284 flaked stone objects were recovered from le vel excavations at Locus B between 2007 and 2010, including 17 form al tools and preforms (Table 6-27). A large majority of both tools and debitage were found within the preceramic Thornhill Lake Phase component. Flaked stone is especial ly sparse within the sites Orange period deposits and is in fact virtually absent fr om the many large roasting pits constituting DP2, perhaps adding supporting evidence to the hypoth esis that Locus B was a special-use site rather than a place of residence during this time. Hafted Bifaces. Among the formal lithic tools recovered from Locus B, all but two exhibit bifacial flaking. Three of thes e have basal stems presumably related to hafting and can be classified within the br oad Florida Archaic Stemmed (FAS) category. One of these hafted bifaces is a classic Ne wnan type with a broad excurvate blade and long contracting stem (Figure 6-70, k). It is mostly complete except for the tip that was removed by a transverse break and the corner s of the shoulders, which were slightly damaged during excavation. Like most FAS points, this example is asymmetrical and has relatively steep edge angles, probably indicating its use as a sharpened cutting implement rather than a projectile point. Two other Locus B bifaces have similar contracting stems and probably started out as FAS points. On e of these (Figure 6-70, i) exhibits the contracting stem and right-angle d shoulders that are characteri stic of Newnan points. Its blade, however, has undergone repeated latera l sharpening, resulting in a cruciform drill shape probably representing th e tools final stage of reduc tion. The other contractingstem biface appears to be a FAS point that e xperienced a perverse fr acture along its long axis and was subsequently recycled into a sm aller hafted cutting tool. Its stem remains largely intact, while the curv ed broken edge of the blade has been sharpened via the removal of several flakes at a steep angle. The tool has a thick i rregular cross-section, perhaps indicating it was not yet complete at the time of the original break. All three FAS tools were found within the precer amic component at Locus B. The only hafted biface that cannot be cla ssified as a FAS implement is a Kirk Stemmed/Serrated point found near the bottom of TU19 (Figure 6-70, m). It has a square stem, straight shoulders, and slightly incurvate blade margins. The blade exhibits broad flake scars across most of its width and was se rrated through the removal of a series of short, deep flakes along both edges. A transverse break rem oved the tip of the point, but the specimen is otherwise intact. It is co mposed of chert that is heavily patenated, indicating an advanced age. Serrated Kirk Stemmed tools have been interpreted as knives and are thought to date to the Early-Middle Arch aic between 8500 and 7000 cal BP (Shroder 2002:76). The fact that this sp ecimen was recovered from a Late Archaic stratigraphic context at Locus B suggests that it was picked up and perhaps utilized long after its initial manufacture and deposition.

PAGE 312

Silver Glen Run, Locus B (8LA1-West) 301 Table 6-27. Attributes and Metric Characteristic s of Flaked Stone Artifacts Recovered from Test Units at Locus B, 8LA1W. Prov. Fig. 6-70 letter Description Condition Max. Length (mm) Max. Width (mm) Max. Thickness (mm) Wt. (g) TU13C d biface tip 18.9 13.1 5.3 0.8 TU14E k hafted biface missing tip 62.1 40.2 7.9 16.0 TU14H c biface preform tip 24.1 17.8 6.6 2.1 TU19G f biface fragment 21.9 20.8 9.1 3.4 TU19I m hafted biface missing tip 60.7 34.4 8.9 17.7 TU19L biface complete 34.9 30.9 10.7 11.5 TU21F j hafted biface recycled/ complete 41.7 20.0 10.6 5.7 TU21H i hafted biface-drill co mplete 41.8 31.1 7.9 5.4 TU39F.38 a uniface-expanding base microlith complete 16.7 10.0 2.8 0.2 TU40G g biface recycled/ dist al 43.2 29.2 13.9 17.5 TU41H biface fragment 32.2 20.8 11.7 6.9 TU43G biface preform tip 20.2 17.2 7.2 1.2 TU46B biface preform medi al 24.1 18 .4 9.2 4.2 TU46G b uniface-expanding base microlith complete 26.7 13.6 4.3 1.0 TU46I h wedge complete 31.6 29.1 8.6 8.8 TU57B e biface preform dist al 30.1 26 .2 9.25 6.2 TU57B biface preform medi al 22.5 17 .1 5.8 3.2 TU57H l biface drill complete 66.9 42.9 8.3 12.5 Preforms Five flaked stone ob jects from Locus B were classified as preforms because they appear to be at a late stage of reduction but do not exhibit the extent of edge retouching and shaping expected of finished tools. Three of th ese consist of tips or distal portions that appear to have been detached via transverse breaks (Figure 6-70, c-e). The other two examples are medial fragments, one with roughly straight parallel margins and the other with contracting edges that may reflect its original location near the base of the

PAGE 313

302 St. Johns Archaeologi cal Field School 2007-2010: Silver Glen Run Figure 6-70. Select flaked stone artifacts from test units at Locus B, 8LA1W.

PAGE 314

Silver Glen Run, Locus B (8LA1-West) 303 tool. Both of these have thick biconvex cr oss-sections and little evidence of fine edge retouch. Because of their fragmentary conditio n, inferences regarding the intended final form or planned use of Locus B preforms are not possible. Other Bifacial Tools. One of the bifaces from Locus B has been classified as a wedge (Figure 6-70, h). It is roughly circul ar in shape and has steeply retouched edges running along its entire perimete r. The presumed distal end of the tool has a number of small step fractures, apparently a result of re peated battering. The tool was found within a preceramic stratum within TU46 and likely belongs to the Mount Taylor Thornhill Lake Phase component. An almost identical tool was found across the spring run at the Silver Glen Springs site (8MR123) that was interpreted as a possible woodworking implement (Randall et al. 2011:144). Another flaked stone object from Locus B has been classified as a drill (Figure 6-70, l). This bifacial tool has undergone a gr eat amount of lateral edge retouch so that its margins curve inward forming a thin elongated blade. The tool terminates proximally in a straight transverse break, although it is unclea r whether this is it s original base or whether it has been truncated. In its present condition, there is no evidence of a hafting element. The drill is made of chert and is heavily patinated, again suggesting an advanced age. Assigning it a cultural affiliation is difficult, as it was found within preceramic deposits but near the edge of a large Orange period pit (Feature 54/55). Various other bifacially flaked fragment s were also recovered whose irregularity and/or condition prevent classi fication into a formal categor y. One of these (Figure 6-70, g) is a thick, chunky biface that appears to have been recycled from a larger tool. Although small flake scars with feather termin ations are visible on both sides of the biface, its edges exhibit evidence of large ir regular flake removal. The proximal end of the original tool is missing, al though it is impossible to tell whether th is break occurred before or after the tools secondary modifica tions. It is one of the few examples of a lithic tool that can be confidently attributed to the Orange period component at Locus B. Microliths. Microlithic tools have been recove red from a number of prehistoric contexts in Florida, including multiple middle St. Johns River Valley sites such as Lake Monroe Outlet Midden (8VO53) (ACI and JR 2001) and the Thornhill Lake Complex (8VO58-60) (Endonino 2010:292-294). Several exam ples have also recently been unearthed at other locations within the Si lver Glen Run Complex including both the Silver Glen Springs site (8 MR123) (Randall et al. 2011) and the Mouth of Silver Glen Run (8LA1E) (Chapter 3, this re port). Two flaked stone obj ects from Locus B have been classified as microlithic tools (Figure 6-70, a-b). Both of these are whole with maximum respective lengths of 16.7 mm and 26.7 mm. In shape, they conform closely to the category of expanded base microliths, as described by Randall et al. (2011:155) based on specimens from 8MR123. They have broad flat bases that contract rapidly into elongated distal shafts and terminate in rela tively dull points. Both Locus B examples appear to have been manufactured from bifacial thinning flakes and were formed via unifacial flaking at an extremely high angle (c a. 90 degree) along the length of the shaft. While Randall et al. (2011:151) suggest a possi ble connection between this tool type and

PAGE 315

304 St. Johns Archaeologi cal Field School 2007-2010: Silver Glen Run the production of marine shell beads, their true function(s) remains poorly understood and could include drilling, pe rforating, and/or incising. Bone Tools and Ornaments Thirty modified bone objects were recove red from test unit excavations at Locus B (Table 6-28). An additional bone tool was recovered from Feature 54/55 in TU57 and is included in Table 6-28 but was not counted in the inventory presen ted in Table 6-25. Unlike the flaked stone tools, which were mostly found within Locus Bs preceramic DP1 component, modified bone appears to be less patterned stratigraphically and was found within Thornhill Lake and Orange depos its, as well as within the upper St. Johns Period component. All modified bone was classified according to the categories outlined by Sassaman et al. (2011) based on the assemblage from Salt Springs (8MR2322). Identified artifact classes include bone pins bone awls, bone splinters, and miscellaneous cut bone and antler. Representative examples of each class are pictured in Figure 6-71. Measurements of each specimen were included in Table 6-28 regardless of condition. Bone pins are defined as highly polishe d items retaining lit tle of the bones cortical surface or medullary cavity and ge nerally exhibiting roughly circular crosssections (Sassaman et al. 2011: 57). Two bone pin fragments were recovered from Locus B, one from the preceramic component of TU 22 and the other from near the base of ceramic-bearing deposits of TU44. The ex ample from TU22 (Figure 6-71, a) is a proximal fragment of a pin that is well-po lished and has a round cross-section. Its proximal end tapers almost to a point and feat ures five thin parall el lines incised around its circumference, resulting in a rattlesnake-like appearance. It is the only decorated bone object from Locus B. The other specimen in this category (Figure 6-71, b) is a small medial fragment from a thinner, undecorated pin. Bone awls are by far the most common modified bone objects recovered from Locus B test units, comprising 76.7 percent of the assemblage. Awls are here defined as pointed tools retaining some element(s) of original bone morphology, usually a portion of the medullary cavity (Sassaman et al. 2011:60). A majority of the Locus B awls have relatively thick, almost triangular cross-sec tions and comparatively blunt working ends (see examples in Figure 6-71, ck). At least six awls, howev er, have somewhat thinner flattened cross-sections, are more highly polished, and exhibit exceedingly sharp and symmetrical pointed ends (Figure 6-71, l-q) Other modified bone types include one bone splinter that has been sharpened to a poi nt on one end (Figure 671, r) along with a deer metapodial fragment and antler tine, both of which were cut straight across perpendicular to their long axis (Figure 6-71, s and t). Apparent damage to the point of the tine suggests that the antler may have been used as a flaking tool. Two tubular bone beads were also recovered from Locus B. One of these (Figure 6-71, u) measures 20.1 mm long, has a diameter of 5.4 mm, and appears to have been made by scoring and snapping off the ends of a small mammal or bi rd long bone. It is similar in both size and morphology to be ads recovered from Gr oves Orange Midden (8VO2601) in nearby Volusia County (Wheeler and McGee 1994). Th e other bead is

PAGE 316

Silver Glen Run, Locus B (8LA1-West) 305 Table 6-28. Attributes and Metric Characteristics of Bone Artifacts from Test Units at Locus B, 8LA1W. Prov. Fig. 6-71 letter Description Condition Max. Length (mm) Max. Width (mm) Wt. (g) TU4-IV awl distal fragment 29.8 4.5 0.8 TU4-VII awl distal fragment 33.5 9.8 1.1 TU12F l awl distal fragment 45.0 8.2 1.4 TU12F t cut antler 34.3 8.7 1.2 TU12G g awl distal fragment 56.0 11.7 4.5 TU14F u bead complete 20.1 5.4 0.4 TU14G k awl distal fragment 29.3 7.6 1.1 TU21C c awl medial fragment 64.8 11.0 5.3 TU21D awl medial fragment 35.4 10.6 1.8 TU21D awl medial fragment 37.0 9.5 1.6 TU21I v bead complete 15.8 17.0 1.8 TU22O a incised pin proximal end 44.5 4.7 1.0 TU40-PZ awl medial fragment 31.2 7.5 1.6 TU40I awl medial fragment 28.5 7.1 1.2 TU40I e awl missing proximal end 134.5 9.3 8.1 TU40J o awl distal fragment 33.0 6.8 0.7 TU41F d awl missing proximal end 142.2 11.2 10.9 TU41J m awl distal fragment 44.8 8.2 1.4 TU44G awl medial fragment 71.9 9.9 3.9 TU44G awl distal fragment 73.5 10.17 4.5 TU44G b pin medial fragment 25.19 3.6 0.3 TU44H n awl distal fragment 42.3 11.0 2.2 TU44H awl medial fragment 23.4 8.0 0.6 TH44H awl medial fragment 27.33 7.6 0.9 TU44I h awl distal fragment 50.2 9.8 3.5 TU46E s cut bone fragment 46.9 17.3 3.5 TU46G q awl distal fragment 45.7 8.0 1.4 TU46G i awl distal fragment 48.5 12.6 4.1 TU57D r splinter 44.9 8.7 1.4 TU57D p awl distal fragment 63.4 8.3 2.0 TU57-Fea 54/55* q awl complete 83.7 16.9 4.6 *not included in Table 6-25 artifact inventory

PAGE 317

306 St. Johns Archaeologi cal Field School 2007-2010: Silver Glen Run Figure 6-71. Select bone tools and ornaments from Locus B, 8LA1W including: a-b) bone pins; c-q) bone awls; r) bone splinter; s) cut bone; t) cut antler; u-v) bone beads.

PAGE 318

Silver Glen Run, Locus B (8LA1-West) 307 shorter (15.8 mm in length) and much broader ( 17.0 mm in diameter). It is also made of bird bone and has an almost perfectly circular cross-section. Both beads have polished exterior surfaces. Modified Marine Shell A total of 10 modified marine shell objects were recovered from test unit excavations at Locus B, not including items r ecovered from features. Six of these came from secure preceramic contexts while at leas t one of the others is likely preceramic in age but was recovered from near the interface between Thornhill Lake and Orange Period components. Overall, the stratigraphic dist ribution of marine shel l (both modified and unmodified) within Locus B largely mirrors that observed with regard to flaked stone. It is by far most common in preceramic contexts becoming much sparser in the deposits of subsequent periods. Marine shell tool ty pe determinations were made using the classification scheme provided by Wheeler and McGee (1994) based on the Groves Orange Midden assemblage. Representative ex amples of the different types identified at Locus B are shown in Figures 6-72 and 6-73. Most of the modified marine shell objects from Locus B are made from large marine gastropods ( Busycon sp.). Two of these are lightning whelk ( Busycon contrarium ) shells that have been modified for use as cont ainers that have been variously referred to as shell receptacles (Wh eeler and McGee 1994:365) or vessels (Sassaman et al. 2011:61). These vessels, which are common in Mt. Taylor contexts throughout the St. Johns River Valley (e.g., ACI 2001; Sassaman 2003; Sassaman et al. 2011; Wheeler and McGee 1994), are hollowed out via removal of their columnella. On e of the examples from Locus B (Figure 6-73) is complete asid e from a large whole that has been burned through the bottom of the vessel, presumably a result of heating its contents. It measures 256 mm in length, 152 mm in width, and is appr oximately 90 mm tall. The other Locus B shell vessel (Figure 6-72, c) is a broken fragment of the wh elks outer whorl with clear evidence for removal of the columnella. Sassaman et al. (2011:61) point to the relatively small capacities offered by even the largest of these shell vessels in suggesting that they were probably special use containers rath er than everyday cooking utensils, perhaps employed in the preparation of medicines or poisons. The other Busycon sp. tools from Locus B are all cu tting implements. Two tools (including the one pictured in Figure 6-72, e), also fashioned from Busycon contrarium fall into the adze/gouge category. These are tr iangular sections of the lower outer whorl that have been sharpened on their broad end. Generally these tools are beveled on the interior in order to produce a unifacial cutting edge, although the working edges are missing on both of the Locus B examples. Anot her Locus B marine shell implement is a knobbed whelk ( Busocyon carica) cutting edge tool (Figure 6-72, d). It is similar to the Type X tools described by both Goggin ( 1952:115) and Wheeler and McGee (1994:365) in having a beveled bit formed by grinding th e siphonal canal but this specimen lacks the shoulder perforation thought to be necessary for hafting. The only other Locus B cutting tools include two Strombus gigas celt fragments (including Fi gure 6-72, f). Wheeler and McGee (1994:361) group all of these variou s cutting tools within the Mt. Taylor

PAGE 319

308 St. Johns Archaeologi cal Field School 2007-2010: Silver Glen Run Figure 6-72. Select marine shell tools and ornaments from Locus B, 8LA1W including: a-b) disk beads; c) Busycon contarium vessel fragment; d) Busycon carica cutting edge tool; e) Busycon contrarium adze/gouge; f) Strombus gigas celt. Working edges are oriented downward for all cutting tools. woodworking complex, hypothesized to have been geared toward the reduction and carving of wood. Three marine shell objects from Locus B fall into Wheeler and McGees (1994:365) personal adornment complex and have been classified as disk variety beads (Figure 6-72, a-b). All three consist of sma ll flat fragments of marine shell with holes drilled through the middle from one side. The largest was recovered from Level F of TU12, near the interface between late preceramic and early Orange period deposits. It

PAGE 320

Silver Glen Run, Locus B (8LA1-West) 309 Figure 6-73. Burned-out vessel made of lightning whelk ( Busycon contrarium) recovered from TU19 at Locus B, 8LA1W. measures approximately 7.0 mm in diameter and is 2.1 mm thick and has smoothed sides and rounded edges. The others were both f ound in fill from Feature 36, a large DP2 pit dating to the Orange period. They are both smaller and more irregular, with snapped edges that have not been rounded off. Determination of the shellfish taxa utilized for these beads was not possible. DISCUSSION AND CONCLUSIONS Four seasons of stratig raphic testing at Locus B have yielded a tremendous amount of information regarding the extent, st ructure, and archaeological significance of the areas cultural deposits. A combinati on of topographic mapping, auger testing, and a total of 45 m2 of test unit excavations were co nducted between 2007 and 2010. Initially, these investigations were geared toward the elucidation of Late Archaic domestic practices and broad-scale community patternin g. As work has progressed and new data have accumulated, however, a picture has grad ually emerged of Locus B not as a fixed

PAGE 321

310 St. Johns Archaeologi cal Field School 2007-2010: Silver Glen Run and stable village, but rather as a hist orically volatile place whose functions and meanings frequently shifted with changing local and regional conditions. As a result, more recent research has fo cused on investigating the various manners in which Locus B was inhabited prehistorically, along with the events and proce sses through which the place was transformed. Information gleaned from auger tests a nd test unit excavations indicates that Locus Bs modern surface topography is larg ely a result of anthropogenic processes involving the deposition of she ll and sand. Radiocarbon data s uggest that most of this landscape modification occurred during the Late Archaic (ca. 5740-3830 cal BP) when the site underwent at least three successive and fundamental ly distinct patterns of deposition. The earliest of these, Depositional Pattern 1 (DP1), was undertaken during the late preceramic Thornhill Lake Phase. Center ed on an area in the southeastern part of Locus B, DP1 resulted from a series of small-scale domestic settlements and abandonments. These settlements are eviden ced first by the stacked sequence of thin, horizontal lenses of crushed shell and intervening layers of dark organic sand most clearly visible near the bottom of TU46. In this te st unit, as well as the 2010 block, DP1 deposits are composed primarily of bivalve and Pomacea shell containing a broad range of artifact types and debitage generally associ ated with the activities of everyday living. A number of pit features exhibiting various sizes, morphologies, and fills also support the interpretation of DP1 as resulting prim arily from routine dom estic practices. The composition and structure of the DP1 deposits at Locus B are comparable in most ways to other late preceramic Archai c sites in the Middle St. Johns Valley. The lithic, shell, and bone artifact assemblages asso ciated with DP1 are all entirely consistent with existing outlines of Mount Taylor period technology (e.g., Randall 2010; Wheeler and McGee 1994; Wheeler et al. 2000). Similarly, the range of pit features and sequence of stacked shell-lined surfaces from Locu s B are also largely duplicated at other coterminous sites in the region (e.g., E ndonino 2010; Sassaman 2003b, Chapter 3 of this report; Sassaman et al. 2005). Like Locus B, these other sites have been interpreted as places of residence during the Mount Taylor and/or Thornhill Lake Phases. By ca. 4600 cal BP, Locus B appears to have been abandoned or perhaps utilized only sparingly for a time, as evidenced by a gap in the radiocarbon data and the presence of a persistent and extensive buried A-horiz on positioned directly above the most recent DP1 deposits. Following this period of disuse, Locus B was transformed from a place of sustained residence to a plac e of periodic, intensive acti vity. Depositional Pattern 2 (DP2) began during the Orange period by ca. 4500 cal BP and i nvolved the excavation and infilling of hundreds of massive pits across an expansive swath of Locus B. In terms of scale, these DP2 pits far exceed any that either preceded or followed them within the Silver Glen Run complex. Many of them ove rlap each other, apparently having been dug one on top of another as a part of intermittent pulses of intense activity. The charcoal, oxidized sand, and concreted bivalve shell found lining the bases of several pits suggest that they may have served as large-scale mu ssel processing facilities, geared toward the rapid production of large amount s of food. While the pits size and number seem to preclude a domestic subsistence-focused explanation, Locus Bs proximity to the

PAGE 322

Silver Glen Run, Locus B (8LA1-West) 311 contemporaneously utilized U-shaped shell mound at 8LA1 E suggests the interesting possibility that the processed shellfish ma y have been consumed at the large-scale ritualized gatherings hypothesized to ha ve taken place there. Regardless of where the shellfish ended up, however, the significance of Locus Bs DP2 pits undoubtedly extended beyond their practical functional ity as processing tools. This is made abundantly clear by th e complex, highly stru ctured deposits though which the pits were infilled. Virtually none of the DP2 pits contain a substantial quantity of artifacts or vertebrate fauna that would indicate their use as refuse containers. In fact, DP2 deposits are largely devoid of most classes of materi al culture save for small amounts of plain fiber-tempered pottery. Instead, pits are filled with various combinations of shell and/or earth that, in several cases, form elaborate stratified sequences of deposition reminiscent of those composing the countless shell mounds found throughout the region. It is possible that Locus Bs infilled DP2 pits were essentially inverted, subterranean versions of these above-ground monuments. If so, their layered deposits were likely a means of inscri bing particular histories into the landscape. While unlike traditional mounds these deposits were at least initia lly hidden from view, their frequently overlapping distribution means that earlier DP2 pits are likely to have been cut into and exposed by later ones, a fact of which the Locus B pit diggers themselves must have been aware. Whereas DP1 at Locus B is replicated at a number of coeval sites throughout the region, DP2 is virtually unique among known Orange period sites, although the data available for comparison is relatively limite d. The best information regarding Orange period domestic practices co mes from Blue Spring Midden B (8VO43) (Sassaman 2003b) in the St. Johns valley and from Summe r Haven (8SJ46) (Janus Research 1995) on the adjacent Atlantic coast. In both cases, excavations revealed relatively artifact-rich deposits, abundant vertebrate fa una, and diverse assemblages of small domestic features, all of which contrasts with th e dearth of material culture and the specialized, hypertrophic pits found at Locus B. A more comparable depositional history may actually have occurred at the Bluffton site (8VO22/23) approx imately 25 km to the south of Silver Glen Run. There, Bullen (1955:3) describes a large cooking hearth some 16 feet across that was characterized by mussel shells that were cemented together by heat and a base that displayed a pink area, eight inches across, which contained nineteen lumps of red ochrecemented sand. This hearth was overlain by a thick layer of r elatively clean, loose shells. It is likely that what Bullen actua lly observed was the burn ed base(s) of one or more large Orange period pits and that the r ed ochre was in reality heat-oxidized sand. His recognition of individual pits may have been hampered by the relatively restricted view offered by the narrow trench excavation that he was overseeing. The data from Bluffton suggest, not surprisingly, that the depositional practices at Locus B were not unique or isolated but were in stead implicated in broader, regional scale processes of shell-centered history-making during the Late Archaic. While the precise events tr iggering the transition from DP1 to DP2 remain poorly understood, it is clear that this shift took place in a context of regional-scale spatial and social transformation. Although traditional ar chaeological consensus has long been that,

PAGE 323

312 St. Johns Archaeologi cal Field School 2007-2010: Silver Glen Run beyond the introduction of pottery technology, the onset of th e Orange Period came with few, if any, significant changes in the lif eways of Floridas Archaic hunter-gatherers (e.g., Milanich 1994:86), that position is no lo nger tenable given cu rrent archaeological data. In terms of material cu lture, the beginning of the Orange period in the St. Johns is characterized by a marked reduction in the number nonlocal objects, including bannerstones, marine shell, and chipped stone artifacts, when compared to the preceding Thornhill Lake Phase (Randall 201 0). It is possible that pottery replaced some of these other materials in maintaining extralocal relatio nships but even if this were the case, the size of the exchange network appears to have been constricted to a large degree. The scant evidence we have for Orange period settlement layout indicates a transition to circular, or perhaps semicircular (e.g., Randall et al. 2011; Sassaman 2003b) arrangements of houses, a pattern that stands in stark contrast to the mostly linear settlements of the prior period. It has been su ggested that this shift may be related to an influx of people and ring-centered ideologies from the Atlantic coast into the interior St. Johns region (Sassaman 2012). Perhaps the most drastic transformation, however, involved an apparent shift in historical consciousness from Mount Taylor times when past and present existed side-byside, to the Orange period when the past was kept at a dist ance from everyday life. One area in which this is most clearly manifested is that of mortuary practices. There is currently no indication that the long-lived Mt. Taylor tradition of burying the dead in shell or sand mounds near settlements continue d into the Orange Peri od. In fact, virtually no Orange burials have been encountered in the Middle St. Johns region (in either domestic or ceremonial contexts), suggesting that whatever Orange people did with their dead, it involved separating thei r remains from contexts of everyday living. In addition, whereas Mt. Taylor communities repeatedly se ttled in the same locations, constructing conspicuous material histories in the form of tell-like mounds of debris, the few known Orange settlements in this area appear sca ttered and relatively ephemeral. And while Orange components are sometimes found within a few tens of meters of Mt. Taylor mounds, the mounds themselves appear to have been actively avoided in all but four known cases, one of these be ing the Silver Glen R un complex (Randall 2010). Based on their huge quantities of large elaborately decorated ceramics and massive deposits of shell, these four Or ange period mound centers are thought by many to have served as regional gathering places during the Orange Period where ritual feasts and other ceremonies were conducted. At Silv er Glen Run, and at least two of the other sites, Orange mounds were constructed dire ctly atop preexisting Mt Taylor mortuaries (Aten 1999; Randall 2010). Thus, in direct contrast to Orange settlements, where the past was intentionally avoided, practices in these specialized ceremonial locations seem to have been geared explicitly toward drawing on the power of the past, probably as a source of ritual legitimacy. In this context, by exposing older deposits, the excavation of Locus Bs DP2 pits would have provided yet another means by which the past could be accessed and exploited at Silver Glen Run. By infilling pits in structured, meaningful ways, particular histories were literally constructed by actors cognizant of the fact that they would eventually be uncovered by future digging.

PAGE 324

Silver Glen Run, Locus B (8LA1-West) 313 Shortly following the cessation of large-s cale pit digging at Locus B, a massive amount of largely undifferentiated, whole Viviparus shell was deposited across the entire area encompassed by the DP2 pits. This mantle of shell (cons tituting Depositional Pattern 3 [DP3]) contains Tick Island Incise d pottery, a style that contrasts dramatically with those that preceded it, and little else. It appears to have been laid down in a single depositional event in some areas, while in othe rs it is broken up by th in lenses of crushed bivalve shell, perhaps indicating a multi-staged depositional process. In either case, the paucity of artifacts, ve rtebrate fauna, and crushed shell point to the inte ntional, rapid emplacement of shell. DP3 completely obscured any evidence of the underlying DP2 features, transforming the Locus B landscape from what must have been an unusually rough and pocked surface into a smooth and unremarkable one. As noted above, the practice of capping pl aces in clean shell at the end of their use lives as a symbol of renewal or tr ansition was a common practice during the preceramic Mount Taylor period. DP3 may simp ly be an Orange period manifestation of this long-lived regional tradition. Alterna tively, multiple lines of evidence, including Locus Bs position back away from the spring run and its elaborate, yet buried mounds of shell, suggest that the ritualized practices associated with DP2 may have intentionally been kept hidden from view. It is possibl e then that during the Orange Period Locus B served as a secluded back-region ( sensu Giddens 1984) where relatively socially restricted rites were conducted in conjuncti on with the larger-s cale, more inclusive activities carried out at the nearby mound. This would help to explain the contrasts in pottery style observed between Locus B and th e north ridge of the mound at 8LA1E. In this scenario, DP3 may actually have been a final step in e fforts to conceal all material traces of DP2 pit digging. Regardless of its specific m eaning, though, DP3 clearly marks yet another major transformation in the Late Archaic use of Locus B. At some point subsequent to DP3, Locus B was once again occupied in a materially conspicuous mannerthis time by people utilizing spiculate-tempered St. Johns pottery. This compone nt has unfortunately been heavily disturbed by modern activities including historic plowing and gopher tortoise burrowing. Consequently, aside from confirming its existence at Locus B, excavations have revealed little about the nature of this areas St. Johns occupation or its historical relationship to underlying deposits. Four seasons of excavations at Locus B ha ve thus shown it to be a significant, and possibly unique, archaeological resource. The data gathered during these investigations are vital not only for illuminating the Late Archaic occupati on of the Silver Glen Run complex but also for achieving a better und erstanding of the larger regional-scale historical processes and even ts that shaped this dynamic period of Floridas past. Ultimately, however, research at Locus B is an ongoing process and important questions remain unanswered. It is still not clear wh at the range of actual practices was that contributed to each of the depositional pattern s discussed above. In addition, what were the events that triggered the seemingly rapid and sweeping transitions between these patterns of use? And finall y, what was Locus Bs actual role in the creation and perpetuation of larger scale social networ ks and historical transformations that

PAGE 325

314 St. Johns Archaeologi cal Field School 2007-2010: Silver Glen Run characterize the Late Archaic in the surrounding region? Th e investigations outlined in this chapter provide a sound basis for delving de eper into these and other issues. Future research, which has already begun with additi onal test unit excavations in 2011 and my own dissertation-focused analyses, will be gear ed toward addressing these questions in an even more complete and meaningful manner.

PAGE 326

CHAPTER 7 CONCLUSIONS AND RECOMMENDATIONS Kenneth E. Sassaman, Asa R. Randall, and Zackary I. Gilmore Four summers of fieldwork at Silver Glen Run has resulted in a fairly detailed account of archaeological sites spanning some 6000 years. Property of the Juniper Club fronting Silver Glen Run once contained some of the largest, most complex shell deposits in northeast Florida. Although mining operatio ns in 1923 compromised much of this archaeological record, intact deposits exist beneath the extant ground surface, as well as in escarpments left after mining shell. More over, the landforms fronting Silver Glen Run include several landforms that contain stra tified archaeological deposits unaffected by mining. One such location, known to us as 8L A1-West Locus B, contains a remarkable record of intensive activity dating from ca. 5700 to 3800 years ago. Another ridge nose, Locus C, houses the remains of a village site dating to the 14th century. Additional archaeological deposits are pervasive in this pa rt of the Juniper Club property; indeed, the land fronting Silver Glen Run is essentiall y one continuous archaeological site, with components distributed differentially but li nked together in a mo re-or-less continuous occupational history. The land today continue s to have great signi ficance for individuals who appreciate its d eep record of human occupation. The St. Johns Archaeological Field School has benefited immensely from the stew ardship of the land by the Juniper Club and the opportunity it has provided to bring th e ancient history of this land to light. In this concluding chapter we summarized br iefly the results of four field sessions at 8LA1 and follow with some general obs ervations and recommendations for further work at this site, as well as other portions of the Juniper Club prope rty. This report has been restricted to field school efforts da ting from 2007 to 2010. Fi eld school was also conducted in 2011 and is planned for the upcom ing summer of 2012. Thus, this report is but the first installment in a series of reports Likewise, graduate students who have been drawing on field school investigations in their doctoral research will issue additional information in their respective dissertations The first of these is expected to be completed in 2013 (Zackary Gilmore, on Locus B), followed by another in 2014 or slightly later (Elyse Anderson, on Locus C). SUMMARY OF INVESTIGATIONS Since the first field school along Silver Glen Run in 2007, a major focus has been on detecting the remnants of a massive U-shap ed shell deposit repor ted at the mouth of the run in 1875 by Jeffries Wyman of Harvards Peabody Museum. All above-ground shell from this once ~8-m-tall deposit was removed in 1923, leaving a relatively flat surface to the east of the club house. Field school investigations at what is known as 8LA1-East in 2007, 2008, and 2010 were only part ially successful in documenting the below-ground footprint of the U-shaped de posit. Coring and test-unit excavation across much of the area revealed a great deal of s ubsurface disturbance, particularly along the shoreline of Silver Glen Run, the presumed location of the north ridge of the U-shaped deposit. Small islands at the mouth of the run believed to be intact remnants of the shell 315

PAGE 327

316 St. Johns Archaeological Field School 2007-2010: Silver Glen Run deposit proved to be redeposited fill, apparent ly emplaced to create fish habitat. Other major disturbances at the mouth of the run relate to the cons truction of a slip and loading ramp for the mining operation. The south ridge of 8LA1-East was like wise severely reduced by mining, although augering and limited excavation in 2007 pr ovided good evidence for its position and orientation. Subsurface deposits observed in the profiles of several test units suggested that shell along the south ridge was emplaced directly over the existing (natural) ground surface. Subsurface shell in some locations of the south ridge appears to have been placed over inorganic sands, the natural subs trate of the landform. Throughout the area of the south ridge, shell-filled pits extended be low the old surface, into the sands below. Ground Penetrating Radar (GPR) deployed in 2010 offered evidence for a circular or arcuate arrangement of such f eatures. Circular villages of Orange age are not unexpected for the region, but we were fr ustrated that subsurface tes ting failed to reveal domestic features expected of a village occupation (e .g., heaths, house floors, post holes, etc.). It remains possible that the entirety of 8LA1-E ast was devoted to ritual activities that simply did not involve the sorts of domestic f eatures and refuse we expect from intensive dwelling. Irrespective of the act ual function of the south ridge, the combined efforts of subsurface testing confirms that it was added after the formation of the north ridge and that this activity resulted in a concentrati on of Orange Plain pottery in the former area and Orange Incised pottery in the latter area. We suspect that people of multiple cultural traditions were involved in the constructi on and use of the U-shaped deposit, some perhaps with roots on the coast, where circ ular or arcuate settlements were common during the Orange period. Investigations to the west of 8LA1 be gan in 2007 with reconnaissance survey. This area contains the remnan ts of a 200-m-long shell ridge that was mined, along with several sites with well-stratified midden depos its of varying age a nd composition, all in reasonably good shape. Systematic shovel te sting along property fronting Silver Glen Run shows that subsurface archaeological de posits are distributed widely across the 11.6ha survey tract. Some 80 percent of the 238 sh ovel test pits (STPs) excavated in the tract yielded pre-Columbian artifacts and/or anth ropogenic shell deposits the latter observed in 133 STPs. Shell density varied markedly, with dense subsurface shell coinciding with the footprint of the mined Mount Taylor shel l ridge designated Locus A, but occurring also across the terrace slopes of Loci B and C and in their respective shell domes to the south, forming the apex of adjacent ridge noses Several areas devoid of shell were also encountered. Most noteworthy is a small shell void at the ap ex of Locus C, to the far west of the survey tract. Ongoing work in this location is providing evidence for a St. Johns II-period village with a presumed centr al plaza, apparently kept clean of shell. Subsurface pre-Columbian artifacts, like sh ell, are distributed widely across the survey tract, revealing spatial patterning indi cative of distinct archaeological components. Pottery sherds are generally absent in Locu s A, the location of the Mount Taylor shell ridge. The oldest pottery, that of the Orange series, is co ncentrated in Loci B and C, largely in the shell nodes of each locus, but also in the downs lope portion of Locus C. St. Johns pottery is likewise distributed acro ss Loci B and C, with especially dense

PAGE 328

Conclusions and Recommendations 317 occurrences in the downslope aspects of both loci. Check-stamped St. Johns pottery is concentrated in Locus C and especially in th e shell-free ridge nose to the west of Locus C. In sum, reconnaissance survey of 8LA1-W est shows that the enti re expanse of land fronting the spring run contains intact subsur face deposits. Variation in the composition and density of subsurface shell and artifacts enables us to subdivide 8LA1-West into three loci (A, B, and C), each the subject of secondary testi ng, and Locus B the target of block excavation. Shell-mining escarpments of the ridge at Locus A encase up to 3 m of stratified deposits. Six 2 x 2-m test units excavated in 2007 and 2008 were distributed across three widely-spaced locations of the ridge to rev eal a consistent sequence of basal midden, accretional shell and sand, house mounds and associated midden accumulation, and possibly a final cap of sand. Seven AMS assays on charcoal from various layers of the ridge indicate that all deposition took place over a three-to-five-century period of the Mount Taylor phase, ca. 6300-5750 cal BP. Comm unities appear to have resided on this ridge as it accumulated, eventually constr ucting house mounds and imposing a formal spatial order to the placement of sand, she ll, and the outputs of daily living. A major finding at Locus A is the use of sand as a medium of mounding. Duri ng this time frame, sand was used in conjunction with shell for purposes that were cl early mortuary (Aten 1999). Evidently, the emplacement of sand at Locus A did not involve human interments (the two subadult humans uncovered and then re buried at the east end of the ridge are an exception, and are not typical of Mount Ta ylor mound burials involving the emplacement of sand [Aten 1999; Endonino 2010]). The use of sand at Locus A may have been simply practical, for instance, in the constructi on of small houses mounds, but if so, it was pervasive, because sand strata up to 40 cm th ick were observed in all profiles exposed to date. The volume of sand emplaced on the ridge is estimated to be a minimum of 2000 m3. One potential source for the sand is the depression 50 m to the south of the ridge, a presumed sinkhole. At ca. 2500 m3, the displaced volume of the depression is comparable to the minimal amount of sand deposited on the ridge. That the depression is indeed a sinkhole is speculative and awaits investigation. The most intensive excavations of the field school have been conducted at Locus B, a relatively small, but complex shell-bear ing site to the west of Locus A. Topographic mapping, auger testing, and a total of 45 m2 of test unit excavat ions were conducted between 2007 and 2010, the last three years under the direc tion of Zackary Gilmore. After the first year of testing, we assumed th e site to be an Ora nge-period habitation, but as the project continued in successive years, we came to understand it as a place whose function and meaning changed repeatedly over nearly a millennium spanning the late precermaic (Mount Taylor) and early ceramic (Orange) periods (ca. 5700-3800 cal BP). The earliest occupation resulted in stratified mi dden deposits similar to those at Locus A. With the introduction of pottery at ca. 4600 cal BP came a flurry of food processing activity involving massive pits, lots of them. While the pits in some cases were evidently shellfish steaming facilities, others have fill and other attributes that evoke a sense of ritual practice. After about five to six cen turies, pit digging and infilling ceased and the site was capped with a thick mantle of shell. Contained in the shell cap are Orangeperiod sherds of the Tick Island variety, a unique regional varian t appearing at about

PAGE 329

318 St. Johns Archaeological Field School 2007-2010: Silver Glen Run 4000 years ago. Coincident with these developments is the construction and use of the large U-shaped shell deposit at 8LA1-East. In the concluding section of the chapter on Locus B (Chapter 6), Gilmore puts the results of Locus B into regional perspective. Much of the archaeological record of Locus B has no parallel in the documen ted literature, but some of its elements gain significance in the context of happenings elsewhere, notab ly on the coast. The argument need not be repeated here, but we note that Gilmores ongoing study of pottery from Locus A and 8LA1-East involves sourcing analyses that ha ve good potential to identify the provenance of clays used to make pots, and with th at, the geographic orig ins of communities participating in large-scale gatherings, such as those imagined for 8LA1-East. Finally, limited testing at Locus C in 2008 documented the presence of a St. Johns II period (ca. 600 years ago) midden at the top of the ridge nose, a presume village site. It is bounded to the north, toward the spring boil, by a 2-m thick, organic-rich midden spanning the past 300 years. Because testing of Locus C did not begin in earnest until the summer of 2011, the results are not included in th is report, but will be included in a later installment. More intensive testing at Locus C is scheduled for 2012 and perhaps beyond, and will figure prominently in th e dissertation projec t underway by Elyse Anderson. DISCUSSION The major goals of field school investigations in the first two years were to locate all archaeological resources in the Juniper Cl ub property fronting Silv er Glen Run and to begin to assess the age and composition of each component through controlled stratigraphic testing. In 2009 a limited crew of experi enced excavators, including field school supervisors, focused efforts on more detailed testing of Locus B in support of dissertation research of Zackar y Gilmore. As this work continued in 2010 we ramped up efforts to document the subsurface remains of the south ridge at 8LA1-East, deploying for the first time Ground Penetrating Rada r (GPR). The combined efforts through 2010 amount to 238 shovel tests across 8LA1-West, 119 m2 of excavation in four areas of 8LA1, about 1200 m2 of GPR survey in a portion of th e south ridge at 8LA1-East, and scores of cores and augers in a variety of contexts. As field work progressed, research questions evolved to encompass more than simply finding, characterizing, and dating archaeological remains. However, the basic goal of characterizing subsurface deposits at 8LA1 remains an ongoing project. The site is a large, complex amalgam of many distinct components, some stratified in vertical sequences, others dist ributed differentially across varied landforms paralleling the spring r un. Above all, 8LA1 is an anthropogenic landscape, whose surface contours and subs urface layers are th e product of human activity, much of it involving the deliberat e emplacement of shell, sand, and other materials. To date, 28 age estimates have been obtai ned on charcoal or soot samples from a wide variety of contexts at 8LA1. Figure 7-1 shows all age estimates as two-sigma calibrated date ranges, ordered by time from ol dest (left) to youngest (right) and divided

PAGE 330

Conclusions and Recommendations 319 Figure 7-1. Two-sigma age estimate ranges on AMS assays from carbon samples from sites 8LA1 and 8MR123, subdivided into components and showing culture-historical divisions of regional chronology.

PAGE 331

320 St. Johns Archaeological Field School 2007-2010: Silver Glen Run vertically from west (bottom) to east (top ) along the run. Included at the bottom are age estimates obtained for 8MR123, the site su rrounding the spring pool of Silver Glen on property of the U.S. Forest Service. Inve stigations at 8MR123 led by Asa Randall in 2010 incorporated field school st udents in the reconnaissance stage of fieldwork. Figure 7-1 is taken from Randalls report to th e Forest Service (Randall et al. 2011). As seen in Figure 7-1, human activity in the greater Silver Glen Run locality was more-or-less continuous from about 7000 to 3700 years ago. We know from shovel tests results and surface collections that occupation of the locality stretches back well before this time, evidently as much as 10,000 years based on the o ccasional occurrence of Early Archaic bifaces and unifaces, as well as heavily patinated flakes. However, archaeological deposits predating 7000 years ago have not presented th emselves in any of our excavations to date. Apparently, shellf ish were not collected and deposited in any significant fashion until this time and without the addition of she ll to archaeological matrix, the residues of human activity are far less conspicuous and less well preserved. The intensive and widespread practice of collecting shellfish and depositing their shells in mounds, middens, and pits after 7000 BP resulted a dramatic and highly visible archaeological record. As we have seen, mu ch of this remains encased deep under the ground, particularly the output of infilling large pits at Locus B and Locus C. After about 3700 BP intensive shellfish coll ecting and shell depositing app ears to have waned. This is more-or-less the end of the Orange period and the beginning of the St. Johns period. Sherds of this later period are common at many locations along the spring run, as they are throughout northeast Florida. That we have ye t to encounter intact deposits of the St. Johns I period (ca. 3700-1100 cal B.P.) is perh aps not too surprising considering the tops of shell deposits have been truncated by mi ning, and any occupation of surfaces that have been plowed or otherwise altered in the modern era would have displaced St. Johns materials from near-surface contexts. The one deeply buried context for a St. Johns I component came from the downslope midden of Locus C. Two other AMS assays for deposits of this time period came from subaqeuous contexts at 8MR123, as well as an additional terrestrial context at that site, in this case from an upland landform away from the spring pool. As work progresses at 8LA1 we are likel y to encounter better contexts for St. Johns I depos its, closing the apparent ga p in the radiocarbon record between Orange and St. Johns II. Well preserved archaeologica l deposits dating to th e St. Johns II period are encased in the portion of 8LA1-East known as Locus C (technically, this portion of the site extends into Marion Count y and is recorded in the Fl orida Master Site Files as 8MR3601). As mentioned repeatedly in this report, concentrated work on this location began in only 2011, and we have plans continue that project in 2012 and perhaps beyond. What information we have to this point sugg ests that Locus C was a circular or semicircular village with a small central plaza and a massive downslope midden that capped earlier deposits. We suspect th at the sand mound to the southeast of Locus C dates to this village occupation (~13th century ). One other AMS assay of this timeframe came from the apparent burned post in the south ridge of 8LA1-East. Placed in what appears to be a basin-shaped deposit of white sand, the post do es not appear to be associated with a

PAGE 332

Conclusions and Recommendations 321 structure and its function rema ins uncertain. Given the freq uency of St. Johns II checkstamped pottery in the water fronting 8LA1-E ast, as well as the spring pool of 8MR123, we are assured that occupation of the area dur ing this interval was substantial, even beyond the village site at Locus C. A few other observations of occupati onal chronology are noteworthy. Enough dates are currently available to suggest that the Mount Taylor occupation of Locus B was initiated at the same time Locus A was abandoned. As with the transformations in site use Gilmore outlines for Locus B, patterns of s ite use across the area seem to suggest that abandonments and relocations involved more than simply finding a bett er place to make a living. Indeed, the proximity of Loci A and B make sit hard to imagine that access to food and water resources was ever significantly different between the two, nor is there any reason to suspect that Locus A became uninhabitable due to flooding or any other catastrophic events. The changes appear instead to be cultural choi ces and may have had little to do with the material conditions or inha bitability of either location. We hasten to add, however, that the post-Moun t Taylor use of Locus A remains elusive. We suggested in Chapter 5 that the cap on top of the shell ridge at Locus A was emplaced as the site was abandoned, but we simply do not have th e data to substantia te this assertion. Another coincidence may be seen in th e transformation of the shell ridge at 8LA1-East into a U-shaped monument a nd the shell cap emplaced over the Orangeperiod pit assemblage at Locus B. This occu rred at about 4000 years ago, when the south ridge was apparently constructe d. This activity appears to ha ve been conducted by people using Orange Plain pottery like that documente d in pit features at Locus B. How the deposition of Tick Island Incised pottery in the shell cap at Locus B relates to construction of the south ridge is unknown, for we have yet to locate pottery of this type in the south ridge. Ongoing research on the provenance of these various wares by Gilmore has the potential to solve this puzzle. A lingering frustration in field school resear ch has been the lack of solid evidence for Orange-period habitation sites. Our assump tion that Locus B was a village of this age has not been realized. We remain optim istic that good evidence for Orange-period habitation will one day be located at 8LA1-East, the locus of an enormous assemblage of Orange Incised pottery, as well as the Orange Plain noted ea rlier at the south ridge. To date, however, nearly all in cidences of Orange Incise d pottery have come from underwater and the redeposited islands at th e mouth of Silver Glen Run. The upland landform of 8MR123 holds potential too for an Orange-period village, and the recent GPR survey of the bait field of Locus B at 8LA1-West offers suggestive evidence for a circular arrangement of subsur face features, another possible village. Solid evidence for places of dwelling during the Orange period are likely to be forthcoming, although we have seen enough of Ora nge-period at 8LA1 to know that what appear s to be habitation is often specialized activ ity that may have had little to do with day-to-day living.

PAGE 333

322 St. Johns Archaeological Field School 2007-2010: Silver Glen Run RECOMMENDATIONS The first four years of investigations at Juniper Club property along Silver Glen Run have been enormously productive, as we ha ve learned a great deal about an area that was not much more that a footnote in the accounts of 19th-century naturalists and antiquarians. And yet, as is the case in archaeo logy generally, our work has led to new questions as basic issues of chronology and site f unction are resolved. There is indeed much more that can be done. We are often asked by club members how long we plan on working on the property, and the answer has al ways been: As long as youll have us. Shortand long-range goals for future fiel dwork can be itemize d. First, we have yet to open a large enough area of Locus A to reveal spatial relationships among different types of deposits (e.g., house mounds, primar y deposition, secondary deposition, features, etc). Because most of the upper portion of the shell ridge at Locus A was removed by mining, the only option for block excavation is the floor of the mining pit. Upwards of one meter of stratified deposits are known to ex ist in certain locations of the mining pit, offering opportunities for examining more than si mply the basement of the ridge. In some locations, GPR may prove useful, alth ough the density of tree roots and nearsurface concreted midden will no doubt obscure any patterning of subsurface cultural features. Nonetheless, the aim in opening a large horizontal area, or multiple areas of Locus A is to collect direct evidence for the spatial arrangement of dwelling during the Mount Taylor phase, and with it paleoenvironm ental data from secure feature contexts. Second, large-scale testing at Locus C initiated in 2011 ough t to continue until we have sufficient data to reconstruct the size a nd configuration of the presumed St. Johns II village overlooking the spring pool, includi ng data on architecture, duration of occupation, and seasonal variations in activitie s. UF graduate student and field school teaching assistant Elyse Anderson is leading efforts to locate good contexts at Locus C for the differential use and deposition of animal remains. Going beyond the usual zooarchaeological questions about seasonal ity and ecology, Anderson is developing a strong case for religious beli efs (i.e., animism) that involved, among other things, ritualized treatment of animal bones. Locus C contains abundant pit fe atures with good bone preservation, as well as a 1.5-m thick dow nslope midden chock full of animal bone, plant remains, pottery, and othe r materials. Ultimately, data on the differential treatment of animal bones in pits and middens will he lp to inform broader aspects of St. Johns ritual, such as the use of sa nd burial mounds and effigy vessels. Third, it is time to break way from 8L A1 and begin to explore the greater archaeological record of the Juniper Club. At least two other sh ell mounds are known from the property, one at the mouth of Little Juniper Creek, and another to the west on a terrace overlooking wetlands paralleling Lake Ge orge. Neither of th ese mounds has been tested, and both appear to be fully intact. We are compelled for ethical reasons to leave these mounds alone, but basic information on chronology and function are needed to establish their significance in the broader c ontext of regional archaeology. Again, remote sensing is recommended, in conjunction wi th limited subsurface testing. To start,

PAGE 334

Conclusions and Recommendations 323 however, we need to conduct reconnaissance using either augers or shovel tests to establish the depth and extent of each deposit. The same terrace edge on which one of th ese mounds lies is in need of full-scale reconnaissance survey. We have traveled the dirt road pa ralleling the terrace edge with Resident Manager Gene Nelson, who pointed out several surf icial shell deposits. It would appear that most, if not all such occurrences of shell ar e recent emplacements, usually to enhance th e traction of steep grades along th e road. However, not all such occurrences may be recent, and irrespective of that, we expect to find evidence for subsurface remains lacking shell, including one dating to the early Holocene, before shellfish were collected and deposited in mo unds. We are compelled to survey this terrace edge whether or not it proves to contain intact subsurface deposits because sound knowledge of where sites are not to be found is just as important as its counterpart, and, equally important, field school students need experi ence in basic rec onnaissance survey, which is what most will spend the majority of their time doing if they find employment in cultural resource management. Long-range goals run the gamut from mo re reconnaissance survey to more intensive excavation. Regardi ng reconnaissance, there remains a need to expand survey of 8LA1-West beyond the southern boundary a nd into the adjoini ng uplands. The dense lithic assemblages found in the uplands of 8MR123 (Ra ndall et al. 2010) may have parallels south of 8LA1-West. There is also a need to test the wetlands in the interior of club property, as well as the southern boundary of the property along Juniper Run. As for additional excavation, we have unresolved que stions about the subsurface composition of the south ridge at 8LA1-East and the bait fiel d of Locus B. Likewise, the nature of saturated deposits along the south margin of the spring run, th e shoreline of Lake George, and below the water table of Shell Point rema ins unknown. On the north side of the run, saturated deposits date predominantly to the Thornhill Lake and Ora nge periods (Randall et al. 2011). A likely parallel may be found in Lake George proper, as well as the basal aspects of the north ridge at 8LA1-East. We suspect, but have ne ver investigated the likelihood that the massive north ridge recorded by Wyman in 1875 was constructed over an existing Mount Taylor shell ridge. Delving into these deeper deposits will likely require a draw down of the water table, something that is both economically and politically challengi ng, and thus not to be undertaken without good cause and without the full endorsement by those charged with st ewardship of the land and its water. The archaeological record bounded by prope rty of the Juniper Club is truly spectacular. Much of it was compromised by land-use practices that today seem irresponsible but in their time were necessary and commonplace. Field school investigations show that mu ch can be learned from sust ained efforts to locate and characterize the remnants of deposits long ago mi ned for shell. Field school efforts have also succeeded in locating s ubsurface deposits along the spri ng run that were unaffected by mining operations and prove to be unprecedente d in the extant literature of northeast Florida archaeology. The partnership between the Juniper Club and the University of Florida to bring this record to light in th e hopes of better understa nding the ancient past has been exceptionally fruitful and hope fully will continue for years to come.

PAGE 335

324 St. Johns Archaeological Field School 2007-2010: Silver Glen Run

PAGE 336

REFERENCES CITED Blitz, John 1993 Big Pots for Big Shots: Feasting a nd Storage in a Missi ssippian Community. American Antiquity 58:80-96. Borremans, Nina T. 1990 The Paleoindian Period Florida Historical Contexts. Borremans, Nina T and Graig D. Shaak 1986 Preliminary Report on Investigations of Sponge Spicules in Florida "Chalky" Paste Pottery. In Papers in Ceramic Analysis, Ceramic Notes No. 3 edited by P. M. Rice, pp. 125-132. Occasional Pub lications of the Ceramic Technology Laboratory, Florida State Museum, Gainesville. Bradley, Richard 1998 The Significance of Monuments Routledge, London. Brain, Jeffrey P. and Drexel A. Peterson 1971 Palmetto Tempered Pottery. Southeastern Archaeological Conference Bulletin 13:70-76. Bullen, Ripley P. 1955 Stratigraphic Tests at Bl uffton, Volusia County Florida. The Florida Anthropologist 8(1):1-16. 1972 The Orange Period of Peninsular Florida In Fiber-Tempered Pottery in Southeastern United States and Northern Colombia: Its Origins, Context, and Significance edited by Ripley P. Bullen and James B. Stoltman. Florida Anthropological Society Pub lications 6, Gainesville. 1975 A Guide to the Identification of Florida Projectile Points Revised ed. Kendall Books, Gainesville. Clark, John E. 2004 Surrounding the Sacred: Geometry and Design of Early Mound Groups as Meaning and Function In Signs of Power: The Rise of Cultural Complexity in the Southeast, edited by J. L. Gibson and P. J. Carr, pp. 162-213. University of Alabama Press, Tuscaloosa. Clausen, C. J. 1964 The a-356 Site and the Florida Archaic. Masters Thesis, University of Florida, Gainesville. Clausen, C. J., A. D. Cohen, Cesare Emiliani, J. A. Holman and J. J. Stipp 1979 Little Salt Springs, Florid a: A Unique Underwater Site. Science 203:609-614. 325

PAGE 337

326 St. Johns Archaeological Field School 2007-2010: Silver Glen Run Clench, William J. and Ruth D. Turner 1956 Freshwater Mollusks of Al abama, Georgia and Florida. Bulletin of the Florida State Museum, Biological Sciences 1:108-111. Cooke, C. Wythe 1939 Scenery of Florida Interpreted by a Geologist Florida Geological Survey Bulletin no. 17, Tallahassee. Cordell, Ann S. 2004 Paste Variability and Possible Manuf acturing Origins of Late Archaic FiberTempered Pottery from Selected Sites in Peninsular Florida In Early Pottery: Technology, Function, Style and Inte raction in the Lower Southeast, edited by Rebecca Saunders and Christopher T. Hays, pp. 63-104. University of Alabama Press, Tuscaloosa. Cordell, Ann S. and Steven H. Koski 2003 Analysis of a Spiculate Clay from Lake Monroe, Volusia County, Florida. The Florida Anthropologist 56(2):113-124. Daniel, I. Randolph and Michael Wisenbaker 1987 Harney Flats: A Florida Paleo-Indian Site Baywood Publishing Company, Inc., Farmingdale, New York. Daniel, I. Randolph, Michael Wisenbaker and George Ballo 1986 The Organization of a Suwannee Tec hnology: The View from Harney Flats. The Florida Anthropologist 39:24-56. Darby, Philip C., Robert E. Bennetts, Steven J. Miller and H. Franklin Percival 2002 Movements of Florida Apple Snails in Relation to Water Levels and Drying Events. Wetlands 22:489-498. Doran, Glen H. 2002a Introduction to Wet Sites and Windover (8BR246) Investigations In Windover: Multidisciplinary Investigati ons of an Early Archaic Florida Cemetery, edited by Glen H. Doran, pp. 1-38. University Press of Florida, Gainesville. 2002b Windover: Multidisciplinary Investigati ons of an Early Archaic Florida Cemetery. University of Florida Press, Gainesville. Dunbar, James S., Michael K. Faught and David S. Webb 1988 Page/Ladson (8je591): An Underwater Paleo-Indian Site in Northwestern Florida. The Florida Anthropologist 41:442-452.

PAGE 338

References Cited 327 Dunbar, James S. and Ben Waller 1983 A Distribution Analysis of the Clovis/ Suwannee Paleo-Indian Sites of Florida: A Geographic Approach. The Florida Anthropologist 36:18-30. Dunbar, James S. and S. David Webb 1996 Bone and Ivory Tools from Submer ged Paleoindian Sites in Florida In The Paleoindian and Early Archaic Southeast edited by David G. Anderson and Kenneth E. Sassaman, pp. 331-. University of Alabama Press, Tuscaloosa. Endonino, Jon C. 2003 Pre-Ceramic Archaic Burial Mounds Along the St. Johns River, Florida Paper presented at the 59th Annual Southe astern Archaeological Conference, Charlotte, NC. 2007 A Reevaluation of the Gainesville, Ocala, and Lake Panasoffkee Quarry Clusters. The Florida Anthropologist 60:77-96. 2008 The Thornhill Lake Archaeol ogical Research Project: 2005-2008. The Florida Anthropologist 61:149-165. 2010 Thornhill Lake: Hunter-Gatherers, Monuments, and Memory. Ph.D. Dissertation, Department of Anthropology, University of Florida, Gainesville. Faught, Michael K. 2004 The Underwater Archaeology of Pale olandscapes, Apalachee Bay, Florida. American Antiquity 69:275-289. Ford, James A. 1969 A Comparison of Formative Cultures in the Americas Smithsonian Contributions to Anthropology 11. Smithsonian Institution, Washington, D.C. Gibson, Jon L. 2000 The Ancient Mounds of Poverty Point: Place of Rings University Press of Florida, Gainesville. Giddens, Anthony 1984 The Constitution of Society: Outlin e of the Theory of Structuration University of California Press, Berkeley. Gilmore, Zackary I. 2009 Orange Pottery Variability and Intras ite Spatial Organization at Silver Glen Run, Lake and Marion Counties, Flor ida. Manuscript on file at the Laboratory of Southeastern Archaeology, University of Florida, Gainesville. 2010 Shell-ving the Midden-Mound Dichotom y: A Diverse Hist ory of Archaic Period Shell Deposition Pract ices at Locus B, Silver Glen Run (8LA1),

PAGE 339

328 St. Johns Archaeological Field School 2007-2010: Silver Glen Run Florida. Paper presented at the 67th A nnual Meeting of the Southeastern Archaeological Conference, Lexington, KY. Goggin, John M. 1952 Space and Time Perspectives in Northern St. Johns Archaeology, Florida University Press of Florida, Gainesville. Gosselain, Olivier P. 1998 Social and Technical Identity in a Clay Crystal Ball In The Archaeology of Social Boundaries edited by M. T. Stark, pp. 79-106. Smithsonian Institution Press, Washington. Griffin, James B. 1945 The Significance of Fiber-Tempered Potte ry of the St. Johns Area in Florida. Journal of the Washington Academy of Sciences 35(7):218-223. Hayden, Brian 1995 The Emergence of Prestige Technologies and Pottery In The Emergence of Pottery: Technology and Innovat ion in Ancient Societies edited by W. K. Barnett and J. W. Hoopes, pp. 257-2 65. Smithsonian Institution Press, Washington. Hegmon, Michelle 1992 Archaeological Research on Style. Annual Review of Anthropology 21:517536.. Janus Research 1995 Archaeological Investigations at th e Summer Haven Site (8SJ46), an Orange Period and St. Johns Period Midden Site in Southeastern St. Johns County, Florida Janus Research, St. Petersburg, Florida. Jenks, Clifford Joseph 2006 Rethinking Culture History in Florida: An Analysis of Ceramics from the Harris Creek Site (8VO24) on Tick Is land in Volusia County, Florida, Unpublished M.A. thesis, Department of Anthropology, University of Florida, Gainesville. Johnson, Katherine Burger 1994 Juniper, Thats Me! The History of the Juniper Club: 1090-1993. Privately published, The Juniper Club, Louisville, KY. Johnson, Robert E. 2005 Phase III Archaeological Data Recovery at the Fort Florida Midden Site (8VO48), at Traderscove's Riverside at Debary Development, Volusia County, Florida Report Submitted to Traderscove Corp. by Florida Archeological Services, Inc. Jacksonville, Fl orida, Jacksonville, Florida.

PAGE 340

References Cited 329 Le Baron, J. Francis 1884 Prehistoric Remains in Florida In Annual Report of the Smithsonian Institution pp. 771-790. Government Printing Office, Washington, D. C. Marquardt, William H. 2010 Shell Mounds in the Southeast: Middens, Monuments, Temple Mounds, Rings, or Works? American Antiquity 75(3):551-570. Marrinan, Rochelle A., H. Stephen Hale and William M. Stanton 1990 Test Excavations at Silver Glen Springs, Florida (8MR123) Miscellaneous Report Series Number 2. Department of Anthropology, Florida State University, Tallahassee. 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. Milanich, Jerald T. 1994 Archaeology of Precolumbian Florida University Press of Florida, Gainesville. Milanich, Jerald T. and Charles H. Fairbanks 1980 Florida Archaeology. Academic Press, New York. Miller, James A. 1997 Hydrogeology of Florida In The Geology of Florida edited by Anthony F. Randazzo and Douglas S. Jones, pp. 69-88. University of Florida Press, Gainesville. Miller, James J. 1998 An Environmental History of Northeast Florida University Press of Florida, Gainesville. Mills, Barbara J. 1989 Ceramics and the Social Contexts of Food Consumption in the Northern Southwest In Pottery and Technology: Ideas and Approaches edited by G. Bronitsky, pp. 99-114. Westview Press. 2007 Performing the Feast: Visual Disp lay and Suprahousehold Commensalism in the Puebloan Southwest. American Antiquity 72:210-239. Moore, Clarence B. 1894 Certain Sand Mounds of the St. J ohn's River, Florida. Part II. Journal of the Academy of Natural Sciences of Philadelphia 10:129-246.

PAGE 341

330 St. Johns Archaeological Field School 2007-2010: Silver Glen Run 1999 The East Florida Expeditions of Clarence Bloomfield Moore Classics in Southeastern Archaeology. University of Alabama Press, Tuscaloosa. Neill, Wilfred T. 1964 Trilisa Pond, an Early Site in Marion County, Florida. The Florida Anthropologist 17. Neill, Wilfred T., H. James Gut and Pierce Brodkorb 1956 Animal Remains from Four Preceramic Sites in Florida. American Antiquity 21:383-395. Newsom, Lee A. 1987 Analysis of Botanical Remains from Hontoon Island (8VO202), Florida: 1980-1985 Excavations. The Florida Anthropologist 40:47-84. 1994 Archaeobotanical Data from Groves' Orange Midden (8VO2601), Volusia County, Florida. The Florida Anthropologist 47:404-417. 2002 The Paleoethnobotany of the Archaic Mortuary Pond In Windover: Multidisciplinary Investi gations of an Early Archaic Florida Cemetery, edited by Glen H. Doran, pp. 191-210. University Press of Florida, Gainesville. Norman, Robert 2010 Images of America: Ocala National Forest Arcadia Publishing, Charleston. O'Donoughue, Jason M. 2011 Environmental and Archaeological Contexts In Archaeological Investigations at Salt Springs (8MR2322), Marion County, Florida edited by J. M. O'Donoughue, K. E. Sassaman, M. E. Ble ssing, J. B. Talcott and J. C. Byrd. Technical Report 11. Laboratory of Sout heastern Archaeology, Department of Anthropology, University of Florida, Gainesville. O'Donoughue, Jason M., Kenneth E. Sassaman, Meggan E. Blessing, Johanna B. Talcott and Julie Byrd 2011 Archaeological Investigations at Salt Springs (8MR2322), Marion County, Florida Technical Report 11. Laboratory of Southeastern Archaeology, Department of Anthropology, the Univer sity of Florida, Gainesville. Piatek, Bruce John 1994 The Tomoka Mound Complex in Northeast Florida. Southeastern Archaeology 13:109-118. Pikirayi, Innocent 2007 Ceramics and Group Identities: Towards a Social Archaeology in Southern African Iron Age Ceramic Studies. Journal of Social Archaeology 7(3):286301.

PAGE 342

References Cited 331 Potter, Alden 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 Stevenson B. Johnson, pp. 13-14. Civilian Conservation Corps, Company 1420, Ocala Camp, Florida F5. Purdy, Barbara A. 1975 The Senator Edwards Chipped Stone Workshop Site (8-Mr-122), Marion County, Florida: A Preliminary Report of Investigations. The Florida Anthropologist 28:178-189. Quitmyer, Irvy R. 2001 Zooarchaeological Analyses In Phase III Mitigative Excavations at Lake Monroe Outlet Midden (8VO53 ), Volusia County, Florida, pp. 1-25. Report Submitted to U.S. Department of Transportation Federal Highway Administration and Florida Department of Transportation District Five by Archaeological Consultants, Inc. and Janus Research. Randall, Asa R. 2007 St. Johns Archaeological Field Sc hool 2005: Hontoon Island State Park Technical Report 8. Laboratory of Sout heastern Archaeology, Department of Anthropology, University of Florida, Gainesville. 2010 Remapping Histories: Archaic Period Community Construction Along the St. Johns River, Florida. Ph.D. Dissertation, Department of Anthropology, University of Florida, Gainesville. Randall, Asa R., Meggan E. Blessing and Jon C. Endonino 2011 Cultural Resource Assessment Surey of Si lver Glen Springs Recreational Area in the Ocala National Forest, Florida Technical Report 13. Laboratory of Southeastern Archaeology, Department of Anthropology, University of Florida, Gainesville. Randall, Asa R. and Kenneth E. Sassaman 2005 St. Johns Archaeological Field Scho ol 2003-2004: Hontoon Island State Park. Technical Report 6. Laboratory of Sout heastern Archaeology, Department of Anthropology, University of Florida, Gainesville. Randazzo, Anthony F. and Dougl as S. Jones (editors) 1997 The Geology of Florida University Press of Florida, Gainesville. Rolland, Vicki L. and Paulette Bond 2003 The Search for Spiculate Clays near Aboriginal Sites in the Lower St. Johns River Region, Florida. The Florida Anthropologist 56(2):91-111.

PAGE 343

332 St. Johns Archaeological Field School 2007-2010: Silver Glen Run Russo, Michael 1990a The Archaic Period Florida Historical Contexts. 1990b East and Central Florida, 3200 B.P.-A.D. 1565 Florida Historical Contexts. 1991 Archaic Sedentism on the Florida Coast: A Case Study from Horr's Island. Unpublished Ph.D. dissertation, Departme nt of Anthropology, University of Florida, Gainesville. 1994 Why We Dont Believe in Archaic Ceremonial Mounds and Why We Should: The Case from Florida. Southeastern Archaeology 13:93-109. 1996 Southeastern Mid-Holocene Coastal Settlements In The Archaeology of the Mid-Holocene Southeast edited by Kenneth E. Sassaman and David G. Anderson, pp. 177-199. University Press of Florida, Gainesville. 2004 Measuring Shell 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. The University of Alabama Press, Tuscaloosa. Russo, Michael, Ann S. Cordell and Donna Ruhl 1993 The Timucuan Ecological and Historic Preserve, Phase III Final Report SEAC Acccession Number: 899. Florid a Museum of Natural History, Gainesville. Russo, Michael and Gregory Heide 2001 Shell Rings of the Southeast US. Antiquity 75:491-492. 2002 The Joseph Reed Shell Ring. The Florida Anthropologist 55:67-87. Russo, Michael, Gregory Heide and Vicki L. Rolland 2002 The Guana Shell Ring. Report Submitted to The Florida Department of State, Division of Historical Resources, Historic Preservation Grant No. F0126. Russo, Michael, Barbara Purdy, Lee A. Newsom and Ray M. McGee 1992 A Reinterpretation of Late Archaic Adaptations in Central-East Florida: Grove's Orange Midden (8VO2601). Southeastern Archaeology 11:95-108. Sassaman, Kenneth E. 2003a New AMS Dates on Orange Fiber-T empered Pottery from the Middle St. Johns Valley and Their Implications for Culture History in Northeast Florida. The Florida Anthropologist 56:5-14. 2003b St. Johns Archaeological Field Scho ol 2000-2001: Blue Spring and Hontoon Island State Parks Technical Report 4. Laboratory of Southeastern

PAGE 344

References Cited 333 Archaeology, Department of Anthropology, University of Florida, Gainesville. 2003c Crescent Lake Archaeological Survey 2002: Putnam and Flagler Counties, Florida Technical Report 5. Laboratory of Southeastern Archaeology, Department of Anthropology, the Univer sity of Florida, Gainesville. 2004 Common Origins and Divergent Historie s in the Early Pottery Traditions of the American Southeast In Early Pottery: Technology, Function, Style and Interaction in the Lower Southeast edited by Rebecca Saunders and Christopher T. Hays, pp. 23-39. The University of Alabama Press, Tuscaloosa. 2005 Hontoon Dead Creek Mound (8VO214) In St. Johns Archaeological Field School 2003-2004: Hontoon Island State Park edited by Asa R. Randall and Kenneth E. Sassaman, pp. 83-106. Technical Report 6. Laboratory of Southeastern Archaeology, Department of Anthropology, University of Florida, Gainesville. 2010 The Eastern Archaic, Historicized AltaMira Press, Lanham, MD. 2012 Drowning out the Past: How Humans Hi storicize Water as Water Historicizes Them In Big Histories, Human Lives: Ta ckling Problems of Scale in Archaeology edited by J. E. Robb and T. R. Pauketat. SAR Press, Santa Fe (in press). Sassaman, Kenneth E. and Michael J. Heckenberger 2004 Crossing the Symbolic Rubicon in the Southeast In Archaeology of the MidHolocene Southeast edited by K. E. Sassaman and D. G. Anderson, pp. 214233. University Press of Florida, Gainesville. Sassaman, Kenneth E., Jason M. O'Donoughue and Julie Byrd 2011 Material Culture. In 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 Byrd, pp. 49-64. Technical Report 11, Laboratory of Sout heastern Archaeology, Department of Anthropology, the University of Florida, Gainesville. Sassaman, Kenneth E. and Asa R. Randall 2012 Shell Mounds of the Middle St. Johns Basin, Northeast Florida In Early New World Monumentality edited by R. L. Burger and R. M. Rosenwig, pp. 53-77. University Press of Florida, Gainesville. Sassaman, Kenneth E., Asa R. Randall, Me ggan E. Blessing and Peter R. Hallman 2005 Hontoon Island North (8VO202) In St. Johns Archaeological Field School 2003-2004: Hontoon Island State Park edited by Asa R. Randall and Kenneth

PAGE 345

334 St. Johns Archaeological Field School 2007-2010: Silver Glen Run E. Sassaman, pp. 27-82. Technical Report 6. Laboratory of Southeastern Archaeology, Department of Anthropology, University of Florida, Gainesville. Sassaman, Kenneth E., J. Christian Russell and John Endonino 2000 St. Johns Archaeological Project Ph ase I: A GIS Approach to Regional Preservation Planning in Northeast Florida Technical Report 3, Laboratory of Southeastern Archeology, Departme nt of Anthropology, University of Florida, Gainesville. Saunders, Joe W., Rolfe D. Mandel, Roger T. Saucier, E. Thurman Allen, C. T. Hallmark, Jay K. Johnson, Edwin H. Jackson, Charles M. Allen, Gary L. Stringer, Douglas S. Frink, James K. Feathers, Stephen Williams, Kristen J. Gremillion, Malcolm F. Vidrine and Reca Jones 1997 A Mound Complex in Louisiana at 5400-5000 Years Before the Present. Science 277(5333):1796-1799. Saunders, Rebecca 2004a Spatial Variation in Orange Cu lture Pottery: Inter action and Function In Early Pottery: Technology, Function, Style and Interaction in the Lower Southeast edited by Rebecca Saunders and Christopher T. Hays, pp. 40-62. University of Alabama Press, Tuscaloosa. 2004b The Stratigraphic Sequence at Rol lins Shell Ring: Implications for Ring Function. The Florida Anthropologist 57(4):249-268. Saunders, Rebecca and Michael Russo 2002 The Fig Island Ring Complex (38C H42): Coastal Adaptation and the Question of Ring Function in the Late Archaic Grant 45-01-16441, South Carolina Department of Archives and History, Columbia. 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 of Florida Press, Gainesville. Schulderein, Joseph 1996 Geoarchaeology and the Mid-Holoce ne Landscape History of the Greater Southeast In Archaeology of the Mid-Holocene Southeast edited by Kenneth E. Sassaman and David G. Anderson, pp. 3-27. University of Florida Press, Gainesville. Scudder, Sylvia 2001 Archaeopedological Analyses Phase Iii Mitigative Excavations at Lake Monroe Outlet Midden (8VO53), Volusi a County, Florida. Report Submitted to U.S. Department of Transportati on Federal Highway Administration and

PAGE 346

References Cited 335 Florida Department of Transportati on District Five by Archaeological Consultants, Inc. and Janus Research. Sears, William H. 1960 The Bluffton Burial Mound. The Florida Anthropologist 13:55-60. 1973 The Sacred and the Secular in Prehistoric Ceramics In Variation in Anthropology: Essays in Honor of John C. McGregor edited by D. Lathrap and J. Douglas, pp. 31-42. Illinois Archaeological Survey, Urbana, IL. Shroder, Loyd E. 2002 The Anthropology of Florida Points and Blades American Systems of the Southeast, Inc., West Columbia, SC. Sigler-Eisenberg, Brenda, Ann S. Cordell, Richard Estabrook, Elizabeth Horvath, Lee A. Newsom and Michael Russo 1985 Archaeological Site Types, Distribut ion, and Preservation within the Upper St. Johns River Basin, Florida. Florida State Museum Miscellaneous Project and Report Series, Number 27. Department of Anthropology, Florida State Museum, University of Florida, Gainesville, Florida. Simpkins, Daniel L. and Dorothy J. Allard 1986 Isolation and Identification of Spanish Moss Fiber from a Sample of Stallings and Orange Series Ceramics. American Antiquity 51:102-117. Soil Survey Staff, Natural Resources Conserva tion Service, United States Department of Agriculture 2011 Official Soil Series Descri ptions. Electronic document, http://soils.usda.gov/t echnical/classification/osd/index.html, accessed 01/15/2011. Stanton, William M. 1995 Archaic Subsistence in the Mi ddle St. Johns River Valley: Silver Glen Springs and the Mt. Taylor Period. Unpublished Masters Thesis, The Florida State University, Tallahassee. Stark, Miriam T., Ronald L. Bishop and Elizabeth Miksa 2000 Ceramic Technology and Social Boundaries: Cultural Practices in Kalinga Clay Selection and Use. Journal of Archaeological Method and Theory 7:295331. Ste. Claire, Dana 1987 The Development of Thermal Alte ration Technologies in Florida: Implications for the Study of Prehistoric Adaptations. The Florida Anthropologist 40:203-208.

PAGE 347

336 St. Johns Archaeological Field School 2007-2010: Silver Glen Run 1990 The Archaic in East Florida: Arch aeological Evidence from Early Coastal Adaptations. The Florida Anthropologist 43:189-197. Talcott, Johanna B. 2011 Paleoethnobotanical Assemblage In Archaeological Investigations at Salt Springs (8MR2322), Marion County, Florida, edited by Jason M. O'Donoughue, Kenneth E. Sassaman, Megga n E. Blessing, Johanna B. Talcott and Julie Byrd, pp. 87-104. Technical Re port 11, Laboratory of Southeastern Archaeology, Department of Anthr opology, the University of Florida, Gainesville. Taylor, Allen R. 1935 Shell Tumuli 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 Stevenson B. Johnson, pp. 14. Civilian Conservation Corps, Company 1420, Ocala Camp, Florida F5. Thulman, David K. 2009 Freshwater Availability as the Cons training Factor in the Middle Paleoindian Occupation of North-Central Florida. Geoarchaeology: An International Journal 24:243-276. United States Department of Agriculture 1975 Soil Survey Report Maps and Interpre tations: Lake County Area, Florida Deptartment of Agriculture Soil Conservation Service, Washington. 1979 Soil Survey of Marion County Area, Florida Deptartment of Agriculture Soil Conservation Service, Washington. 1980 Soil Survey of Volusia County, Florida Deptartment of Agriculture Soil Conservation Service, Washington. Watts, William A., Eric C. Grimm and T. C. Hussey 1996 Mid-Holocene Forest History of Flor ida and the Coastal Plain of Georgia and South Carolina In Archaeology of the Mi d-Holocene Southeast edited by Kenneth E. Sassaman and David G. Anderson, pp. 28-38. University of Florida Press, Gainesville. Wheeler, Ryan J. and Ray M. McGee 1994 Technology of Mount Taylor Peri od Occupation, Groves Orange Midden (8VO2601), Volusia County, Florida. The Florida Anthropologist 47:350-379. 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.

PAGE 348

References Cited 337 White, William Arthur 1970 The Geomorphology of the Florida Peninsula Bureau of Geology Division of Interior Resources Florida, no. 51, Tallahassee. Wiessner, Polly 1983 Style and Social Information in Kalahari San Projectile Points. American Antiquity 48(2):253-276. Wing, Elizabeth S. and L. McKean 1987 Preliminary Study of the Animal Remains Excavated from the Hontoon Island Site. The Florida Anthropologist 40:40-46. Wobst, H. Martin 1977 Stylistic Behavior and Information Exchange In For the Director: Research Essays in Honor of James B. Griffin edited by C. E. Cleland, pp. 317-342. Anthropological Papers, no. 61. Museum of Anthropology, University of Michigan, Ann Arbor. Wyman, Jeffries 1872 Journal and Field Notes. Manuscr ipt on file, Harvard Countway Medical Library. (HMS B54.1). Cambridge, MA. 1875 Fresh-Water Shell Mounds of the St. John's River, Florida. Memoirs of the Peabody Academy of Science 1 (4).

PAGE 349

APPENDIX A: CATALOG 339

PAGE 350

340 St. Johns Archaeological Field School 2007-2010: Silver Glen Run CATALOG CODES MATERIAL MATTYPE FORM DESCRIPTION BOTANICAL CHARCOAL C14 MATERIAL FOR C14 ANALYSIS BOTANICAL CHARCOAL FRAG UID CARBONIZED PLANT REMAINS CONCRETION UID UID UID CONCRETION HISTORIC METAL ** (**) FORM OF METAL FRAGMENT HISTORIC METAL FRAG METAL FRAGMENT HISTORIC MOD UID MISC UNIDENTIFED HISTORIC/MODERN LITHIC MOD BIFACE BIFACIALLY WORKED STONE LITHIC MOD FLAKE FLAKE WITH MODIFIED MARGINS LITHIC MOD HAFTEDBIFACE BIFACE WITH MODIFICATION FOR HAFTING LITHIC MOD HAMMERSTONE LITHIC MOD UNIFACE LITHIC UNMOD FLAKE LITHIC MOD SANDSTONE MODIFIED SANDSTONE MARINESHELL MOD FRAG MARINESHELL MOD WHOLE MARINESHELL UNMOD FRAG MISCROCK UNMOD LIMESTONE MISCROCK UNMOD PEBBLE POTTERY OFTI BODY ORANGE INCISED POTTERY OFTI CRUMB ORANGE INCISED POTTERY OFTI RIM ORANGE INCISED POTTERY OFTP BODY ORANGE PLAIN POTTERY OFTP CRUMB ORANGE PLAIN POTTERY OFTP RIM ORANGE PLAIN POTTERY SJCS BODY ST. JOHNS CHECK STAMPED POTTERY SJCS CRUMB ST. JOHNS CHECK STAMPED POTTERY SJCS RIM ST. JOHNS CHECK STAMPED POTTERY SJP BODY ST. JOHNS PLAIN POTTERY SJP CRUMB ST. JOHNS PLAIN POTTERY SJP RIM ST. JOHNS PLAIN POTTERY SJIN BODY ST. JOHNS INCISED POTTERY SJIN CRUMB ST. JOHNS INCISED POTTERY SJIN RIM ST. JOHNS INCISED POTTERY STP BODY SAND TEMPERED PLAIN POTTERY STP CRUMB SAND TEMPERED PLAIN POTTERY STP RIM SAND TEMPERED PLAIN POTTERY STCS BODY SAND TEMPERED CHECK STAMPED POTTERY STCS CRUMB SAND TEMPERED CHECK STAMPED POTTERY STCS RIM SAND TEMPERED CHECK STAMPED POTTERY GTP BODY GRIT TEMPERED PLAIN POTTERY DCS BODY DEPTFORD CHECK STAMPED BIVALVE BIVALVE WHOLE FRESHWATER BIVALVE (>50%) BIVALVE BIVALVE FRAG FRESHWATER BIVALVE (<50%) TERRESTRIALSNAIL EUGLANDINA UID VERTFAUNA ANTLER ANTLER VERTFAUNA MOD BONE VERTFAUNA UNMOD ANTLER VERTFAUNA UNMOD FRAG VERTFAUNA UNMOD SHARKTOOTH VERTFAUNA UNMOD CONCRETION FIRECLAY UID UID POTTERY SJPNT BODY ST. JOHNS "PAINTED", PROBABLY DUNN'S CREEK RED POTTERY SJPNT RIM POTTERY SJPNT CRUMB MISCROCK MOD STONEBEAD PALEOFECES UNMOD FRAG

PAGE 351

Appendix A: Catalog 341

PAGE 352

342 St. Johns Archaeological Field School 2007-2010: Silver Glen Run

PAGE 353

Appendix A: Catalog 343

PAGE 354

344 St. Johns Archaeological Field School 2007-2010: Silver Glen Run

PAGE 355

Appendix A: Catalog 345

PAGE 356

346 St. Johns Archaeological Field School 2007-2010: Silver Glen Run

PAGE 357

Appendix A: Catalog 347

PAGE 358

348 St. Johns Archaeological Field School 2007-2010: Silver Glen Run

PAGE 359

Appendix A: Catalog 349

PAGE 360

350 St. Johns Archaeological Field School 2007-2010: Silver Glen Run

PAGE 361

Appendix A: Catalog 351

PAGE 362

352 St. Johns Archaeological Field School 2007-2010: Silver Glen Run

PAGE 363

Appendix A: Catalog 353

PAGE 364

354 St. Johns Archaeological Field School 2007-2010: Silver Glen Run

PAGE 365

Appendix A: Catalog 355

PAGE 366

356 St. Johns Archaeological Field School 2007-2010: Silver Glen Run

PAGE 367

Appendix A: Catalog 357

PAGE 368

358 St. Johns Archaeological Field School 2007-2010: Silver Glen Run

PAGE 369

Appendix A: Catalog 359

PAGE 370

360 St. Johns Archaeological Field School 2007-2010: Silver Glen Run

PAGE 371

Appendix A: Catalog 361

PAGE 372

362 St. Johns Archaeological Field School 2007-2010: Silver Glen Run

PAGE 373

Appendix A: Catalog 363

PAGE 374

364 St. Johns Archaeological Field School 2007-2010: Silver Glen Run

PAGE 375

Appendix A: Catalog 365

PAGE 376

366 St. Johns Archaeological Field School 2007-2010: Silver Glen Run

PAGE 377

Appendix A: Catalog 367

PAGE 378

368 St. Johns Archaeological Field School 2007-2010: Silver Glen Run

PAGE 379

Appendix A: Catalog 369

PAGE 380

370 St. Johns Archaeological Field School 2007-2010: Silver Glen Run

PAGE 381

Appendix A: Catalog 371

PAGE 382

372 St. Johns Archaeological Field School 2007-2010: Silver Glen Run

PAGE 383

Appendix A: Catalog 373

PAGE 384

374 St. Johns Archaeological Field School 2007-2010: Silver Glen Run

PAGE 385

Appendix A: Catalog 375

PAGE 386

376 St. Johns Archaeological Field School 2007-2010: Silver Glen Run

PAGE 387

Appendix A: Catalog 377

PAGE 388

378 St. Johns Archaeological Field School 2007-2010: Silver Glen Run

PAGE 389

Appendix A: Catalog 379

PAGE 390

380 St. Johns Archaeological Field School 2007-2010: Silver Glen Run

PAGE 391

Appendix A: Catalog 381

PAGE 392

382 St. Johns Archaeological Field School 2007-2010: Silver Glen Run

PAGE 393

Appendix A: Catalog 383

PAGE 394

384 St. Johns Archaeological Field School 2007-2010: Silver Glen Run

PAGE 395

Appendix A: Catalog 385

PAGE 396

386 St. Johns Archaeological Field School 2007-2010: Silver Glen Run

PAGE 397

Appendix A: Catalog 387

PAGE 398

388 St. Johns Archaeological Field School 2007-2010: Silver Glen Run

PAGE 399

Appendix A: Catalog 389

PAGE 400

390 St. Johns Archaeological Field School 2007-2010: Silver Glen Run

PAGE 401

Appendix A: Catalog 391

PAGE 402

392 St. Johns Archaeological Field School 2007-2010: Silver Glen Run

PAGE 403

Appendix A: Catalog 393

PAGE 404

394 St. Johns Archaeological Field School 2007-2010: Silver Glen Run

PAGE 405

Appendix A: Catalog 395

PAGE 406

396 St. Johns Archaeological Field School 2007-2010: Silver Glen Run

PAGE 407

Appendix A: Catalog 397

PAGE 408

398 St. Johns Archaeological Field School 2007-2010: Silver Glen Run

PAGE 409

Appendix A: Catalog 399

PAGE 410

400 St. Johns Archaeological Field School 2007-2010: Silver Glen Run

PAGE 411

Appendix A: Catalog 401

PAGE 412

402 St. Johns Archaeological Field School 2007-2010: Silver Glen Run

PAGE 413

Appendix A: Catalog 403

PAGE 414

404 St. Johns Archaeological Field School 2007-2010: Silver Glen Run

PAGE 415

Appendix A: Catalog 405

PAGE 416

406 St. Johns Archaeological Field School 2007-2010: Silver Glen Run

PAGE 417

Appendix A: Catalog 407

PAGE 418

408 St. Johns Archaeological Field School 2007-2010: Silver Glen Run

PAGE 419

Appendix A: Catalog 409

PAGE 420

410 St. Johns Archaeological Field School 2007-2010: Silver Glen Run

PAGE 421

Appendix A: Catalog 411

PAGE 422

412 St. Johns Archaeological Field School 2007-2010: Silver Glen Run

PAGE 423

Appendix A: Catalog 413

PAGE 424

414 St. Johns Archaeological Field School 2007-2010: Silver Glen Run

PAGE 425

Appendix A: Catalog 415

PAGE 426

416 St. Johns Archaeological Field School 2007-2010: Silver Glen Run

PAGE 427

Appendix A: Catalog 417

PAGE 428

418 St. Johns Archaeological Field School 2007-2010: Silver Glen Run

PAGE 429

Appendix A: Catalog 419

PAGE 430

420 St. Johns Archaeological Field School 2007-2010: Silver Glen Run

PAGE 431

Appendix A: Catalog 421

PAGE 432

422 St. Johns Archaeological Field School 2007-2010: Silver Glen Run

PAGE 433

Appendix A: Catalog 423

PAGE 434

424 St. Johns Archaeological Field School 2007-2010: Silver Glen Run

PAGE 435

Appendix A: Catalog 425

PAGE 436

426 St. Johns Archaeological Field School 2007-2010: Silver Glen Run

PAGE 437

APPENDIX B RADIOCARBON DATA 427

PAGE 438

428 St. Johns Archaeological Field School 2007-2010: Silver Glen Run Beta Measured Conventional Lab 14C 13C/12C 14C 2-sigma 2-sigma Prov. Material Number Age BP Ratio Age BP Cal AD/BC Cal BP TU44-Tree wood 285044 120 40 -24.8 120 40 AD 1670-1780 280-160 AD 1790-1960 160-0 Feature 22 charcoal 298846 610 50 -24.8 610 50 AD 1280-1420 670-530 Feature 63 charcoal 289502 680 40 -25.6 670 40 AD 1270-1330 680-620 AD 1340-1400 610-560 Core 1-IV peat 236136 1330 40 -23.4 1360 40 AD 620-690 1330-1260 TU18-Core charcoal 248526 2440 40 -25.1 2440 40 BC 760-400 2710-2350 Feature 18 charcoal 298845 2790 60 -23.9 2810 60 BC 1120-820 3070-2770 Feature 36b charcoal 264443 3570 40 -23.9 3590 40 BC 2030-1880 3980-3830 Feature 38m charcoal 264446 3590 40 -25.0 3590 40 BC 2030-1880 3980-3830 Feature 1 charcoal 255903 3610 40 -25.8 3600 40 BC 2110-2100 4060-4050 BC 2040-1880 3990-3830 Feature 37 charcoal 264444 3630 40 -24.1 3640 40 BC 2130-1900 4080-3850 Feature 38b charcoal 264445 3670 40 -25.1 3670 40 BC 2190-2180 4140-4120 BC 2140-1940 4100-3890 Vessel 12 soot 166671 3690 60 -25.8 3680 60 BC 2210-1900 4160-3850 Feature 54 charcoal 285043 3690 40 -25.9 3680 40 BC 2190-2170 4140-4120 BC 2150-1950 4100-3900 Feature 33 charcoal 264442 3730 40 -24.1 3740 40 BC 2280-2240 4230-4190 BC 2240-2030 4190-3980 Feature 15 charcoal 255904 3820 40 -24.3 3830 40 BC 2460-2190 4410-4140 BC 2180-2140 4120-4100 Feature 26 charcoal 264441 3960 40 -24.5 3970 40 BC 2570-2440 4520-4390 BC 2420-2400 4370-4350 BC 2380-2350 4320-4300 Vessel 6 soot 166672 4020 60 -25.2 4020 60 BC 2850-2820 4800-4770 BC 2680-2430 4630-4380 Vessel 27 soot 166673 4060 40 -24.4 4070 40 BC 2860-2810 4810-4760 BC 2690-2480 4640-4430 Feature 50 charcoal 285042 4180 40 -24.8 4180 40 BC 2890-2630 4840-4580

PAGE 439

Appendix A: Radiocarbon Data 429 Beta Measured Conventional Lab 14C 13C/12C 14C 2-sigma 2-sigma Prov. Material Number Age BP Ratio Age BP Cal AD/BC Cal BP Feature 48 charcoal 285041 4240 40 -25.4 4230 40 BC 2910-2850 4860-4800 BC 2810-2750 4760-4700 BC 2720-2700 4670-4650 TU46-VIa charcoal 285045 4490 40 -24.9 4490 40 BC 3360-3020 5300-4970 TU46-XIa charcoal 285046 4940 40 -24.4 4950 40 BC 3880-3650 5740-5600 TU10A-C14-5 charcoal 298849 5130 40 -24.7 5130 40 BC 3990-3910 5940-5860 BC 3880-3800 5830-5750 TU10A-C14-4 charcoal 248528 5160 50 -25.4 5150 50 BC 4040-3910 5990-5860 BC 3880-3800 5830-5750 TU8-STR19 charcoal 298848 5280 40 -25.2 5280 40 BC 4240-3980 6190-5930 Feature 6 charcoal 299734 5290 40 -25.0 5290 40 BC 4240-3990 6190-5940 TU5E-XXII charcoal 236137 5290 40 -24.7 5290 40 BC 4240-3990 6190-5940 TU5-STRVI charcoal 298847 5320 30 -25.0 5320 30 BC 4250-4040 6200-6000 TU10A-C14-3 charcoal 248527 5390 40 -24.4 5400 40 BC 4340-4230 6290-6180 BC 4200-4170 6150-6120