Show Summary Details

Page of

PRINTED FROM OXFORD HANDBOOKS ONLINE ( © Oxford University Press, 2018. All Rights Reserved. Under the terms of the licence agreement, an individual user may print out a PDF of a single chapter of a title in Oxford Handbooks Online for personal use (for details see Privacy Policy and Legal Notice).

date: 22 November 2019

Archaic Shell Mounds in the American Southeast

Abstract and Keywords

Freshwater and estuarine shellfish began to be exploited in the southeastern United States between 9000 and 7000 b.p. Shortly thereafter, shell mounds appeared in the mid-South Shell Mound Archaic, along the St. Johns River in peninsular Florida, and, somewhat later, in the Stallings Island area along the middle Savannah River. On the lower Atlantic and Gulf Coasts, shell rings arose. Until recently, all these mounds were considered middens—the accumulations of the remains of simple meals of mobile peoples who visited the same areas for hundreds or thousands of years. More recent scholarship indicates that these mounds were deliberate constructions—some of the first sculpted landscapes created by Archaic peoples to memorialize the past, celebrate the present, and provide for the future. In this chapter, recent research on shell sites in these four areas is discussed. The emphasis is on changing perspectives about the peoples who built them.

Keywords: Archaic Period; Shell Mound Archaic; Shell Rings; Stallings Island; shell midden; shell mound.

Research over the past 20 years or so has produced a sea change in the interpretation of Archaic mounded shell sites, which were deposited between ca. 9000 and 3200 cal b.p. in the southeastern United States (Figure 1). Into the 1990s, it was generally accepted that coastal and inland shell beds appeared only when sea level rose to approximate modern shorelines around 4,000–5,000 years ago (e.g., Milanich 1994). Most thought that, prior to this time, sea level rose so quickly that productive estuarine environments could not develop. Then, when sea level was high enough to exert back pressure on rivers, around 5000 b.p., river velocity slowed and channels wandered, creating, for the first time, the fecund floodplains and estuaries where (among other things) mystery snails (St. Johns River in Florida), mussels (interior southeast), and oysters (Atlantic and Gulf Coasts) grew in enormous numbers, ready for exploitation by hunters and gatherers already living in the area.

The massive shell piles resulting from this exploitation were testimony to the importance of shellfish, although why these relatively nutritionally poor resources were eaten in such large numbers was a puzzle. Under the ecological models of the day, reliance on low-quality resources was a response to resource stress caused by (for instance) population pressure. Some suggested that shellfish were marginal foods (e.g., Parmalee and Klippel 1974) and their bulky debris misrepresented their importance (e.g., Byrd 1977), while others argued that shellfish were important because they were sessile, and hence predictable, and could be gathered by women, children, and the aged (Yesner et al. 1980). Few questioned the conventional wisdom that the massive shell mounds of the interior Ohio River system and St. Johns River Valley, and the curiously shaped shell rings of the lower Atlantic and Gulf Coasts, were anything other than garbage accrued over centuries or millennia (cf., Cable 1997; Michie 1979; Waring 1968; Waring and Larson 1968). That these huge structures were accreted kitchen middens was “confirmed” by the presence of other (assumed) domestic discards: abundant food bone, pottery, and stone, bone, and shell tools.

Changes in this model over the past 20 years include the recognition that (1) shellfishing began much earlier than 5,000 years ago, and (2) some, if not most, of these shell structures were specifically created ritual landscapes rather than the daily discard of victuals. A subsidiary tenet of this focus is that shell in and of itself was (and is) “symbolically potent” (Claassen 2008; Sanger 2015, 489).

Archaic Shell Mounds in the American SoutheastClick to view larger

Figure 1. Location of areas and selected sites discussed in text. 1 = Fig Island site; 2 = St. Catherines Island shell rings. There are ca. 16 additional rings in the rectangle. Dotted line is Benton Interaction Zone, based on Kidder and Sassaman 2009.

Apparently, shellfish can exist in rapidly rising seas because oysters were exploited almost 9,000 years ago along the Texas Gulf Coast (and 13,000 years ago in California). The mussel (Unionoida) shell mounds of the mid-South began to accrue by at least 8900 cal b.p. (Bissett 2014), and, by 7300 cal b.p., Native Americans along the St. Johns River in Florida began to exploit gastropods (mystery snail, Viviparus georgianus, and apple snail, Pomacea paludosa) (Randall 2013). The earliest shell rings, created predominantly of oyster (Crassostrea virginica), appeared along barrier islands and the adjacent mainland of the lower Atlantic and Gulf Coasts by 5200 cal b.p. However, because the earliest rings are inundated, it would not be surprising to find earlier rings offshore (Sassaman 2010; Saunders and Russo 2011).

Shellfish also appear to have been associated with burial in the Southeast. Burial in shell middens is common in Archaic and early Woodland cultures. In general, this has been regarded by researchers as a practical solution to the problem of disposal of the dead. More recently, the association of burials and shell has been viewed as a referent to Native American core beliefs in “rebirth and rejuvenation” (Claassen 2010, 173). Claassen (1991, 1992, 2008, 2010, 2015) has presented the most extended argument for this viewpoint. Using ethnographic and archaeological examples from throughout the New World, she argued that the shell mounds of the Ohio Valley were, in fact, burial mounds. According to Sassaman (2006, 149), “the very first use of shell in the middle Savannah was for mortuary purposes.” Indeed, Sassaman suggested that Claassen’s hypotheses for the mortuary uses of shell in the Ohio Valley “may very well explain the origins of freshwater shellfishing in the middle Savannah, and perhaps the St. Johns basin too” (Sassaman et al., 2006, 561). Against this backdrop, it is curious that formal burials are not present in shell rings, although scattered human bone is found. Still, this is consistent with Claassen’s (2010, 36–37) observation that, except for the Stallings Island site (about which more later) shell-bearing sites founded relatively late in the Archaic on south-flowing rivers either lack burials or have very few. Although burials are not found in shell rings, Sanger (2015) agreed that shell carried a heavy symbolic load in those structures as well. As will be seen, this viewpoint is not without critics.

In all four of the aforementioned areas, some researchers now argue that shell mounds and rings were not simply deposits of domestic garbage but locations of population aggregation, interaction, ceremony, and feasting. In concert with the increasing acceptance of the presence of Middle and Late Archaic earthen mounds in Louisiana and Florida, more researchers suspect social and ceremonial functions for large, mounded, Archaic shell-bearing sites in the mid-South, lower Atlantic and eastern Gulf Coasts, and the St. Johns River. While details vary from region to region (and researcher to researcher), and dramatic differences exist in the types of materials deposited and depositional processes that created these structures, many agree that these sites are the first cultural landscapes intentionally created for remembrance of the past, celebration of the present, and preparation for the future.

In this chapter, I briefly describe mounded shell sites in each region and discuss models for their development and use. My own experience with shell-bearing sites is primarily with the shell rings of the lower Atlantic, and, to a lesser extent, the Gulf Coast. This bias is reflected in the amount of space given to rings. I have also given more space to the first area discussed, the Shell Mound Archaic of the Ohio River Valley because the interpretive issues reviewed for that region are similar to those in the other areas. Given space constraints, many important topics are not discussed. However, I hope to demonstrate that, after years of modeling Archaic societies as driven primarily by environmental opportunities or constraints, some very plausible social reconstructions are emerging that give depth and breadth to ancient populations.

Shell Mounds of the Mid-South: The Shell Mound Archaic

Although the term is sometimes more broadly applied, the “Shell Mound Archaic” (SMA) is used here to refer to a set of distinctive sites present only along the Ohio River and its tributaries (Claassen 2010, 36) in portions of northern Alabama, Tennessee, Kentucky, Indiana, and West Virginia (Figure 1). At least 60 shell sites are named, and many more exist but have not been described (Claassen 2010, figures 2.1, 2.2, 2.3). Recent dates from western Kentucky suggest initial sites were occupied as early as 8900 cal b.p. (Bissett 2014). According to Sassaman (2010), at this time, a new population entered the region (one of two ancestral populations of the Native Americans of the eastern United States). These people may have come from the Columbia Plateau, bringing a tradition of shellfishing with them. Site use intensified through time; most SMA sites were occupied in the Middle and into the Late Archaic. Most were abandoned by the end of the Late Archaic period, ca. 3500 b.p. (Claassen 2010, 47). However, a few Tennessee River SMA sites were used into the Woodland period (Peacock 2002).

Archaic Shell Mounds in the American SoutheastClick to view larger

Figure 2. Selected shell ring footprints, to scale.

Although a good deal of variation exists in site age, size, and contents across the region, site characteristics cohere at a general level and contrast strongly with sites outside the region (Sassaman 2010, 43). Much of our information comes from excavations of seminal sites done under the auspices of the Works Project Administration (WPA) and the Tennessee Valley Authority (TVA; Lyon 1996). However, there are a number of recent studies involving reanalysis of earlier materials and additional excavations (e.g., Bissett 2014; Claassen 2010; Marquardt and Watson 2005; Milner and Jeffries 1998; Moore 2011).

As noted, SMA shell mounds are enormously variable, but, in essence, they are large, subcircular to oblong mounds that accrued over centuries or even thousands of years, possibly with some hiatuses in use. Some shell mounds cover large areas, up to 24,281 square meters, and can be as tall as 12.2 meters above the floodplain or the low terraces where they are found (Bense 1994; Claassen 2010, 87–88). As such, they are imposing structures, rising starkly out of the floodplains or low terraces where they are located. Shell mound volumes calculated by Claassen (2010, Table 6.1) range from 1,500 to 14,000 cubic meters, the latter rivaling the volume of the smaller earthen mounds at Mississippi period sites.

Mussel shell is the principal component of most mounds, although gastropods predominate in a few areas (Claassen 2010, 160; Peres and Deter-Wolf 2016). Stratigraphy is often complex, with strata or lenses ranging from predominantly whole shell valves and, sometimes, quantities of unopened shell, through crushed shell with varying amounts of earth, to non-shell earth midden. In the SMA, as for other shell structures, some researchers consider clean, whole shell as evidence for rapid deposition and little post-depositional disturbance (e.g., Claassen 2010).

Features are dense within these deposits and include fire features and hearth dumps, fired clay floors, limestone or fire-cracked rock clusters, tool caches, and several different types of pits. However, postholes or other evidence for domestic structures are rare. Artifacts are abundant. For instance, at the Chiggerville site on the Green River, WPA excavators recovered 1,455 stone tools, including bifaces, scrapers, drills, adzes, cores, engraving tools, spokeshaves, pièce esquillées, and flake tools. Groundstone tools included some “200 pestles, axes, mano/hammerstones, [and] pitted stones.” Bone and antler tools were used for “perforation, weaving or matting, flaking stone tools. Bone and antler were also fashioned into fishhooks, atlatl hooks, projectile points, fleshing tools, and one bone flute” (Moore and Thompson 2012, 273).

Some mounds were used after the introduction of pottery in the area, and these may contain small amounts of Wheeler Fiber-Tempered pottery. However, pottery is conspicuously absent in some mounds that are late enough to include it (Claassen 2010, 47).

SMA involvement in long-distance trade is viewed differently by different researchers. Certainly, the southern portion of the SMA was involved in the regional Benton lithic distribution network (Kidder and Sassaman 2009; for Benton chronology, see McNutt 2008), which provided both utilitarian bifaces and spectacular oversized blades (Figure 1). However, most discussion centers on the possibility of more extensive trade networks, particularly in copper from the Great Lakes and marine shell from the Gulf or South Atlantic Coast. The former is relatively rare, whereas the latter is represented by thousands of marine-shell beads as well as cups and gorgets.

A number of researchers have considered how this trade might have been organized. Winters (1968; see Marquardt and Watson 2004) suggested that a site like the spectacular Poverty Point site (1700–1100 cal b.c., ca. 3700–3100 cal b.p.) would be found for SMA redistribution of exotics. Goad (1980) believed western Kentucky peoples were middle-men in the exchange of copper and marine shell products between the Great Lakes and the southern coast. Marquardt (1985, 83) thought that exotic goods could have been obtained by “trader-diplomats.” However, others have downplayed the role of trade in interregional relations. Jeffries (2009, 657) stated that, for the Green River area, “The few pieces of copper … could have been brought to the Green River region from the Lake Superior area by one person making one trip.” Indeed, there appears to have been a strong seam along the Ohio River that discouraged interaction of the SMA with northern cultures (Sassaman 2010).

Trade with the south is considered more open. However, Claassen (2010, 210–211) was unimpressed by the amount of marine shell. Some 18,443 shell beads made of Busycon whelks have been recovered from SMA sites (11,891 of these are from the Green River sites), which might suggest a lively traffic. However, when Claassen translated that number into Busycon shells, she determined that only 367 shells would be required to produce those tens of thousands of beads. “When divided over the probable 1000 years of most intensive use, the harvest is miniscule and need not depend on trading relationships” (Claassen 2010, 211).

Burials are common in SMA mounds; many contain hundreds. Although far from being the largest mound, Indian Knoll had the largest number of documented burials (n = 880). However, differences in preservation and in the proportion of mound excavated make raw numbers like this difficult to compare (Claassen 2010, 106–107). Still, controlling for the volume, Claassen estimated some 18,000 individuals were interred in shell mounds (Claassen 2015, 1; 2010, Appendix 1). Many of the deceased died as a result of violence, a finding that has led researchers to suggest population pressure was creating resource stress and that a rise in territoriality led to violence.

Most of the copper and marine shell artifacts became mortuary items, as did the aforementioned Benton blades. Bannerstones were also common mortuary items. Notably, dogs were buried, either with human burials or by themselves. None of the other Archaic cultures discussed in this essay buried dogs in human mortuary contexts.

Traditionally, SMA sites have been considered the aggregated discard of villages (Claassen 2010, 1), seasonal encampments (Milner and Jeffries 1998, 130), or base camps of “small groups of late Middle to Late Archaic hunter-gatherers” (Milner and Jefferies 1998, 119). As noted, Claassen (1991, 1992, 2008, 2010) has challenged this interpretation, arguing instead that the mounds were special-purpose sites. Their function was world renewal, achieved, at least in part, through mortuary rituals in conjunction with “feasting with shell.” She argued that shell mounds with burials were short-term encampments, but they were part of a larger community of ritual sites that also included tributary shell mortuariesshell mounds without burials, shell-free mortuaries, caves and rock shelters, and bluff-top sites (to which shell was transported). All these ritual sites were located away from domestic sites, which were encampments, not villages—Claassen (2010, 223) was pretty adamant that there were no Archaic villages (cf. Claassen 2010, 134).

Claassen began marshalling her arguments for this series of hypotheses in the 1980s. The massive amount of data used to substantiate her nuanced argument cannot be adequately summarized here, and only a few topics are discussed. Looking at site distributions on a regional basis, Claassen argued that mounds appeared only along specific waterways, whereas the environmental conditions for shellfish exploitation were widespread. Most mounds were on north- or northwestward-flowing rivers. More locally, mounds appeared in clusters, although there do not seem to be significant environmental differences in locations with mounds and those without. Indeed, a shell stratum at one mound may be contemporaneous with an earth stratum in a nearby mound. On a site-specific basis, Claassen noted that the number of paired shells and whole, clean shell strata indicate some shell was used only for construction, not subsistence, and that the lack of disturbance at the shell mounds (exemplified by paired shells and the integrity of strata and features) is inconsistent with the post-depositional mixing that would result from either village life or seasonal encampments. She also described differences in burial treatment and burial goods and found statistically significant differences in burials at the different types of mortuary sites. These results suggested that those who died of violence were sacrifices, not victims of skirmishes. In addition, she noted that the beginning and ending of the SMA cannot be associated with environmental change. Thus, social factors were involved, presumably changes in religious beliefs and practices.

Claassen’s publications have been met with skepticism (e.g., Bissett 2014; Hensley 1994; Peacock 2002; Milner and Jeffries 1998; Watson 1992). Bissett’s (2014) recent detailed critique is based on his review of seven SMA sites in western Tennessee. Among other disagreements, he (Bissett 2014, 545) argued that the mounds were not primary mortuary facilities—that the number of burials was “miniscule” given the time depth involved. Instead, mounds reflected a pattern of “incidental and sporadic use for burial during short stays by a variety of groups during their long histories.” Arguing from experience in the Green River area, Watson (1992) and Hensley (1994) noted that many burials in mounds are not associated with shell (other arguments are summarized in Marquardt and Watson 2004). These researchers model the mounds as slowly aggrading sites used at varying intensities, from village use to simple seasonal visits of short duration. Crothers and Bernbeck (2004) suggested that these varying intensities might be due to population flux—people moved around the region as individuals or families so that the resident population of any particular site was constantly changing. Bissett and others1 (Crothers 1999; Thompson 2010) acknowledged that the location of the mounds in environmentally productive regions and their reuse through deep time render them “persistent places” (Bissett 2014, 560). Persistent places are physical locations or cultural landscapes that have qualities that encourage repeat occupations. Through time, persistent places gain importance and power. The persistent places perspective is a bridge between a strictly materialist perspective on site function and one that incorporates social explanations. Expanding on the persistent places theme, Moore and Thompson (2012) used the animism they found in a “dwelling” perspective (in which all things are sacred) to further reconcile the secular and sacred viewpoints. They argued that SMA mounds were both daily living spaces and hallowed mortuary areas, but remarked that it is unlikely that Archaic peoples made these kinds of distinctions anyway.

In each of the four areas under consideration, site function is being reappraised in much the same way as for the SMA. The arguments in each area center on the tempo of shell deposition and the intentionality involved in shell mounding. In more traditional interpretations, villages and campsites aggrade slowly, through daily practices. Burials, no matter how reverently interred, did not change the essentially utilitarian function of the sites. The persistent places approach has been used for sites of many different time periods and many different cultures around the world (e.g., Barton et al. 1995, Littleton and Allen 2007; Nord 2009; Schlanger 1992), and it has been used to interpret Archaic period shell structures in all of the regions discussed here (e.g., Thompson 2010). Persistent places began as opportunistically located villages or campsites and generally aggraded slowly. Their importance, too, was enhanced gradually. At the opposite end of the interpretive spectrum, mounds as mortuary facilities (and shell rings; see later discussion) were intentional constructions from site inception, and there were at least pulses of rapid deposition of shell indicative of feasting involving relatively large populations.

Shell “Mounds” of the Mid-Savannah River Region: Stallings Island

In contrast to all the other early interpretations of Archaic shell structures, the Stallings Island site has always been considered the cultural center of the region and a “necropolis” (Jones 1873), where Late Archaic peoples buried their dead amidst dense piles of freshwater mussel shell. The site is sometimes considered a shell mound, created by processes similar to those traditionally modeled for SMA sites: that is, by gradual accretion of domestic debris (e.g., Bense 1994). However, the island itself is a dome-shaped, eroded clay, remnant formation, “approximately 120 m long, 80 m wide, and as much as 5 m above the alluvial apron that comprises the base of the island” (Sassaman et al. 2006, 541). The natural curvature in the topography of the site and the presence of shell midden of varying depths (ca. 0.4–3 m) over the surface give the illusion of a mound. If any mounding exists, it is only on the northern end of the site, and it was done by the last inhabitants of the site (Sassaman et al. 2006).

Sassaman’s recent work on the island and in the region has resulted in a reworking of the site chronology to include a preceramic period.2 He has also created a lively landscape of allies and factions in “historicizing” the Archaic. Thus, using Sassaman findings, it now appears that the earliest occupation at the Stallings Island site (ca. 5300 cal b.p.) may have been by descendants of the Middle Archaic Benton culture SMA peoples, who migrated to the coastal plain after about 5,700 years ago (Sassaman 2006, 42). In total, the site contains three components, two of which occurred before pottery was adopted in the region: Paris Island (ca. 5300–4700 cal b.p.) and Mill Branch (ca. 4700–4450 cal b.p.), after which there was a hiatus in site use (Sassaman 2006; Sassaman et al. 2006). During these earlier phases, the site was not occupied year round. When not at Stallings Island, Paris Island and Mill Branch peoples “occupied riverine and [upland] inter-riverine locations, presumably on a seasonal basis” (Sassaman et al. 2006, 551). Shellfish were not routinely exploited at these other sites. With a nod to Claassen, Sassaman (2010, 155) noted that this indicates shellfishing was a “cultural preference, not an ecological imperative.” Like Claassen, he saw “a close association between the consumption of shellfish and the interment of the dead” (Sassaman 2010, 156). The third component was that of the Classic Stallings culture, which produced an elaborately decorated fiber-tempered ware. Classic Stallings peoples occupied the island on a more permanent basis between 4200–3800 cal b.p., after which the island was abandoned.

During the pre-pottery occupations, rather than mounding shell, most of the discarded mussel shell, along with significant amounts of other artifacts, were draped over the edge of the island. Sassaman (2006, 98–99) mused that this discard pattern might have produced a ring-like midden around the perimeter of the island and that the site at this time may have been the interior analog of coastal shell rings. If so, the ring would have been 60 meters in diameter and some 3 meters high (Sassaman et al. 2006, 561). However, Sassaman ultimately dismissed the idea of a direct connection between the Classic Stallings of the mid-Savannah region and the Stallings culture on the coast.

In addition to lithic, bone and antler tools (notably bone fishhooks and beveled antler projectile points), stone netsinkers, soapstone cooking slabs, shell beads, and bannerstones, these peoples left behind “an untold number of human interments” (Sassaman et al. 2006, 552). Sassaman (2010, 156) noted that for the pre-ceramic piedmont peoples, Stallings Island was both a mortuary site and a seasonal habitation site. He also acknowledged the significance of shell in mortuary contexts, noting that the earliest burial discovered in the mid-Savannah region, dating to at “least 6500 b.p.” (Sassaman 2006, 149), was on a habitation site in a pit filled with shell. Other mundane features at the site (Mims Point) did not contain shell.

Stallings Island was either abandoned or only lightly used between 4450 and 4200 cal b.p. (Sassaman et al. 2006, 552), during which time coastal Stallings peoples migrated into the area with their early, undecorated Stallings fiber-tempered pottery.3 They intermingled with Mill Branch peoples until about 4200 cal b.p., when at least a faction of Mill Branch people left the valley. Any Mill Branch people who remained presumably intermarried with the descendants of the coastal folk. This amalgamation of peoples reoccupied Stallings Island and created Classic Stallings culture. Its most distinctive attribute is an elaborately decorated pottery. Sassaman (2006) believed the punctated and incised vessels that appear at this time in the region were vehicles for establishing an identity distinct from other groups in the area. He noted stylistic similarities between Stallings Island decorated pottery and the pottery at the closest shell ring, the 200 kilometer distant Chesterfield ring, near Beaufort, South Carolina, but ultimately considered the two “distinct, albeit closely related, communities” (Sassaman et al. 2006, 554).

During Classic Stallings times, occupation on the island was more concentrated in the center of the landform. Stallings Island once again became a residential and a mortuary site. One mortuary area was within the center of a small, circular, sedentary village some 30–35 meters in diameter (Sassaman 2006, 102). Unlike their predecessors, Classic Stallings peoples deposited their shell and other refuse in deep pits (ca. 1 m deep—Sassaman et al. 2006, table 2) and also were responsible for what appears to be the only intentional mounding at the site, in the northern portion of the site, where another small village may have been located. Other sites of Classic Stallings age area may also contain circular settlements, but they do not contain central burials (Sassaman et al. 2006, 561).

During the late Paris Island and Mill Branch phases, steatite cooking stones and bannerstones originating in the upper and middle Savannah piedmont area circulated in the Coastal Plain, perhaps in exchange for marine shell beads. Steatite cooking stones are common in some sites, but bannerstones are rare (Sassaman et al. 2006, 552). Still, the distinctive, hypertrophic bannerstones were traded out in “wildly divergent directions” (Sassaman 2006, 66) and have been found in the St. Johns River area and in at least one shell ring (Sapelo). Sassaman (2006, 66) believed that the distribution of these items indicated that alliances with “outsiders” were not necessarily friendly and may have been a source of tension. Once the Mill Branch peoples left the valley, elaborate bannerstones and other aspects of the Piedmont cultures disappeared.

Sassaman (2006, 94) suggested that Classic Stallings was “a society turned inward, with figurative ramparts erected to preserve the autonomy and unique identity of discrete riverine communities. At the same time, the elaboration of Stallings pottery and emphasis on serving ware hints at a society designed to interact with ‘outsiders.’” Indeed, Stallings Island became a regional ceremonial center. Classic Stallings peoples of the island hosted feasts, probably associated with mortuary rituals. These rituals appear to have been conducted only at Stallings Island (Sassaman et al. 2006, 541) and included a vessel form, the carinated vessel, that is almost entirely restricted to Stallings Island. The carinated shape is appropriate for serving vessels, not cooking. Use of the aforementioned socketed, beveled antler points intensified. They may have been used in pursuit of large fish, like sturgeon, which, in addition to shellfish, might have been a feasting food (Sassaman et al. 2006, 560). The center did not hold, however, and the site was abandoned by 3800 cal b.p. Sassaman (2006, 153) hypothesized that environmental factors, the social cost of mortuary ritual, and/or internal strife were responsible.

After Stallings Island was abandoned, “post-Classic” Stallings occupied other areas in the vicinity but did not exploit shellfish intensively (Sassaman 2006, 168–169). Ultimately, the middle Savannah River region was abandoned; peoples dispersed into upland settlements, leaving their Classic Stallings culture behind (Sassaman 2006, 154). Direct causal evidence is unclear, but in several areas in the Southeast there is evidence for environmental perturbations after this time.

Shell Mounds of Peninsular Florida: St. Johns River Shell Mounds

A third set of shell mounds occurred along the Upper (southern) and Middle regions of the northward-flowing St. Johns River, in peninsular Florida. The bulk of these St. Johns River shell mounds were constructed during the preceramic Mount Taylor period (7300–4600 cal b.p.) A few sites contain fiber-tempered ceramic Orange period (4600–3600 cal b.p.) components (Randall et al. 2014, 18; further chronological breakdowns are presented in Beasley 2008; Endonino 2008; Randall 2013). Unlike in other areas, St. Johns shell mounds saw use into the Woodland and Mississippi periods, and some sites were used as late prehistoric mortuaries (Sassaman and Randall 2012).

The mounds began to accrue at least as early as 7300 cal b.p., when pre-existing regional populations began to aggregate along the river (Randall 2013, 196). This may be coeval with “a significant shift in atmospheric water availability” (Randall et al. 2014, 24) that may have affected the St. Johns River region by increasing surface water. Freshwater gastropods were the mollusk of choice, with the banded mystery snail (Viviparus georgianus) the most abundant shellfish used, followed by the Florida apple snail (Pomacea paludosa); mussels (Unionidae) were also exploited (Randall 2013). Blessing (2010; see also Sassaman 2010, 156) suggested that the snails were not always subsistence discards, but could be collected in large numbers along the shoreline during die-offs caused by environmental stresses or demographic cycles.

The St. Johns River shell mounds were part of a settlement system that evolved through time and ultimately included linear or crescent-shaped shell ridges; platform mounds created with intermittent lenses of clean shell, burned shell, and sand; shell and sand mortuary structures; and conical burial mounds of sand and shell.4 In addition, Wyman (1875) noted four colossal U-shaped structures spaced about 20–30 kilometers apart along the river bank (see Figure 2). To date, only a remnant of one of these has been identified, at the Mouth of Silver Glen Springs site. In addition to these monumental structures, there are also shallow, discontinuous shell fields, shell-free lithic sites, and (shell and non-shell) wet sites originating as deposits along shorelines (Randall 2013, 199). These types are not mutually exclusive. Larger, multimound sites may have many of these features in close proximity. Because so many of these mounds have been destroyed by mining, development, and looting, those that remain “are but a fraction of the scores” of the mounds once present (Randall et al. 2014, 18).

Today, ridges are the most numerous site type (15 are known); they range in size between 45 and 200 meters in length and between 2.0 and 8.3 meters high (Randall 2013). These were the basic domestic site and appear to have been occupied seasonally, at least initially. The multiple shell mound complexes are fewer in number, but much larger, ranging from 300–600 meters and more than 10 meters high. Not all of the volume can be attributed to Archaic occupants. According to Randall (2013, 197), all of the multi-mound complexes were used after the close of the Archaic.

Like the SMA, the St. Johns River shell mounds have been considered the remains of simple, redundant habitations that accrued over centuries or even thousands of years (e.g., Wyman 1875). Randall et al. (2014, 19) referred to this as “the standard model” that has persisted since the 19th century. As for the SMA, this explanation is being replaced by a coherent, if complex, settlement model involving considerable purposeful construction and intensive site remodeling. Staged sequences of construction indicate that many sites began as domestic shell ridges but were transformed into “commemorative platforms whose surfaces were renewed repeatedly, possibly during feasting events that included regional populations” (Randall et al. 2014, 25).

The Harris Creek site (once called Tick Island) is the earliest mortuary site known. Harris Creek contained more than 175 individuals interred between 7240 and 6480 cal b.p. (Randall 2011; Tucker 2009). About 15% of the individuals had grave goods, most of which were bone pins, beads, and other ornaments, or bifaces, although one individual may have worn antlers (Aten 1999). “None of the graves were lavish” (Sassaman 2010, 68). Stable isotope analysis (δ13C and δ18O) of bone indicated that most individuals came from throughout Florida, and two were possibly from Virginia or Tennessee (Quinn et al. 2008; Tucker 2009, 141). This result has implications for understanding mobility, marriage patterns, and regional alliances far beyond what was once considered probable for such an early period.

The isotopic data also indicated that freshwater shellfish were not a significant component of the diet of the peoples interred at Harris Creek. (One datum in the idea that some of the gastropods used in building were collected from massive die-offs.) Instead, Mt. Taylor peoples appear to have derived a significant portion of their diet from marine or estuarine sources. Settlement patterning for the period has been modeled either as sedentary or as involving seasonal mobility between the coast and the river, with base camps on the St. Johns during the warm months and dispersal to the coast in the cooler months. However, additional isotopic evidence instead suggests that coastal occupation occurred during the summer and fall (Tucker 2009).

Between 5700 and 4600 cal b.p., occupation in the valley contracted and was structured around a few key places. Conical burial mounds of earth and shell were constructed on pre-existing sites and might contain materials borrowed from earlier sites (Endonino 2008). The mounds have a comparatively small footprint, just 20–30 meters in diameter, but they can rise up to 4.9 meters above the surrounding landscape. The mounds are often found in pairs, although the temporal relationship between the two is poorly known.

Perhaps not surprisingly given the aforementioned evidence for long-distance movement and alliances, there is evidence for trade in elaborate exotic items into the St. Johns region during this time. Rare, hypertrophic bannerstones appear in graves in the conical mounds by ca. 5000 cal b.p. As noted earlier, these likely came from the Stallings Island area (Sassaman 2010, 69). Other items appearing in the mounds include stone beads, possibly from Mississippi, and marine shell beads, some of which may come from the southern tip of Florida.

The elaborate exotic items are rare. However, lithics for utilitarian artifacts were imported into the stone-poor St. Johns River Valley in some quantity. For instance, in the Silver Glen Springs area, stone used for larger bifaces and microliths came from quarries between 125 kilometers to the northwest and 150 kilometers to the southwest (Randall et al. 2014, 32). Lithic assemblages from other Mount Taylor period sites along the St. Johns indicate similar networks. Marine shell, especially busycon, was also obtained for tools (celts, axes) and for special-purpose vessels (Randall et al. 2014, 33).

The Orange period saw further settlement in the river valley. Sassaman and Randall (2012) believed these pottery-making people were immigrants from the coast, and that they brought both Orange fiber-tempered pottery and a penchant for U-shaped monuments. As in the Stallings Island area, the commingling of Mt. Taylor peoples with the newcomers produced a vigorous new society, but one that was contested (Sassaman and Randall 2012). As in the Stallings area, the tensions were partially manifested in the different ways that pottery was used on the St. Johns sites as compared to the coast. In both areas, Orange Incised pottery was used predominantly (or possibly exclusively on the St. Johns) at “rings.” However, on the coast, Orange Incised was an elaborately decorated serving ware. In the St. Johns River Valley, Orange Incised pottery was a thick, often heavily sooted, cooking ware; serving vessels were small and undecorated (Sassaman and Randall 2012; Saunders 2004b; Saunders and Wrenn 2014). As is the case in most other areas where Orange culture people resided (including most coastal shell rings), trade dropped off considerably compared to the Mount Taylor period (Gilmore 2015, 126). Lithic artifacts are rare; assemblages are dominated by fiber-tempered pottery.

During the Orange period, shell mounding may have taken place in only four areas (Gilmore 2015, 126)—those locations where Wyman observed the U-shaped monuments. At the Mouth of Silver Glen Run site, a massive U-shaped structure was built along the bank of the spring run, where it debouches into Lake George (Figure 2). This spring run watershed was inhabited throughout most of the previous Mount Taylor period —the area also contains two large Mount Taylor ridges, as well as shell fields, wet site deposition, a mortuary mound, and lithic scatters (Randall et al. 2014). The “ring,” really three conjoined shell ridges, was the site of feasting in conjunction with other ritual. Randall (2008, 16) proposed that “the abundance of decorated Orange vessels in near-shore contexts suggests that the deposition of vessels into the water was part of a larger commemorative event.” However, unlike in the SMA and at Stallings Island, feasting was not done in conjunction with mortuary rituals. Burial in shell (or sand) mounds ceased. In fact, only a very few Orange period burials have ever been found in the St. Johns River Valley (see later discussion).

Evidence of an Orange residential site is present at the Silver Glen Springs site complex. On the basis of pottery distributions, it appears there were multiple structures “arranged around small plazas, which were further organized around one large plaza” (Randall et al. 2014, 32). While Randall worried that this complex design may be “overly elaborate,” it has a clear topographic analog in the Rollins Shell Ring (see later discussion). With the exception of these possible structures, Orange settlements in the valley were “scattered and ephemeral” (Gilmore 2015, 138).

Shell Rings: Lower Atlantic Coast, Western and Northern Gulf Coast to the Pearl River

Shell rings are another set of shell structures created in the Archaic period (Figure 2). Like Stallings Island and most SMA sites, shell rings as described here existed only in the Archaic; most were in use from ca. 5200–3500 2 cal b.p. Subsequent Woodland or Mississippi period use was generally ephemeral and manifested by just a small number of sherds. Like the other shell structures described here, rings are highly variable and there are exceptions to almost any generalization one would make (Russo [2006] provides an indispensable overview). However, like the SMA mounds, rings do form a cohesive whole and contrast strongly with contemporaneous shell structures in the Southeast.

Instead of a massive structure of variable shell and earth strata containing abundant features and burials, rings are circular or open curved structures (U- or C-shaped) that encompass a shell-free interior area generally referred to as a plaza. Ring walls are generally comprised of large heaps of clean, whole shell, usually oyster, although two rings (McQueens, in Georgia, and Guana, in Florida) have substantial deposits of clam (Mercenaria mercenaria) (Russo et al. 2002; Sanger 2015; Saunders and Rolland 2006). There is generally little soil admixture in this shell, although in some instances there is soil build-up on slopes. At Rollins Shell Ring, very thin lenses of sand segregate large depositional episodes (Saunders 2004a). The clean shell suggests, and radiocarbon dates confirm, that most ring walls were completed quickly. For example, radiocarbon dates from the top and bottom of the shell deposits from the aforementioned Rollins site overlapped at one sigma. This is consistent across ring sites; in contrast to SMA and St. Johns River sites, individual ring construction and use was short-lived, perhaps only a few decades (Russo 2010, 157) or generations (Saunders 2004a). An abrupt sea level regression is cited by most as the reason for the demise of rings at the end of the Late Archaic (see papers in Thomas and Sanger 2010).

Although not absolutely sterile, plazas have relatively few artifacts and generally few features. Even where features are found, the plaza areas appear to have been intentionally kept clean (Sanger 2015; Saunders 2004a; Thompson 2006). In addition, rings are singularly self-contained (Figure 3). The area outside the ring wall is similarly clear of debris (Sanger 2015, 215). It is almost as if the rings themselves were little islands.

Archaic Shell Mounds in the American SoutheastClick to view larger

Figure 3. Shovel testing showing sterility of area outside the St. Catherines Island rings.

Image courtesy of Matthew Sanger (Sanger 2015, Figure 3.7).

Some rings are barely visible topographically, whereas the highest point of Fig Island 1 (the tallest measured to date) is a wall of shell 5.5 meters high. Interior diameters range from less than 40 to more than 200 meters. In South Carolina and Georgia, rings tend to be relatively small, closed circles, and multiple rings occur. Florida rings are U-shaped, occur singly, and are substantially larger than those in Georgia and South Carolina. Ringlets, possibly analogous in plan if not in function to the Orange residential compound at Silver Glen Springs, are well developed at Rollins and also occur at Fig Island. Four rings are associated with mounds. There is one at Fig Island 1, four at Horrs Island, one at Bonita Bay, and one at Claiborne (Russo 2006).

As for the shell structures in other regions, arguments about ring function have shifted over the years. McKinley’s (1872) suggestions ranged from dance grounds to torture chambers. More recently, the Sewee rings have been considered fish traps (Edwards 1965) or reservoirs in extensive freshwater capture systems (Middaugh 2011; see also Marquardt 2010). Although a few researchers have long argued that rings had a ceremonial function (e.g., Michie 1979), an influential article by Trinkley (1985) swayed many to the conclusion that rings were simply the result of the centuries-long aggregation of daily meals. The circular shapes were the product of the spatial patterning of egalitarian societies. Indeed, many ring reconstructions depict houses on top of the ring (see drawing in Russo 2014). However, features suggestive of structures have not been identified on rings; with a few site exceptions, features are rare in ring walls. Features can be abundant beneath walls, however (e.g., Sanger 2015; Saunders 2004a).

By the turn of the 21st century, information from additional excavations, extensive radiocarbon dating, and sophisticated seasonality studies have resulted in a different consensus: the large piles of shell and vertebrate food remains are viewed as the result of deliberate, careful deposition of the remains of feasting events that occurred in conjunction with population aggregations and ritual. As with the SMA sites, this viewpoint is not without its detractors. Marquardt (2010), for instance, argued that archaeologists have been too hasty in assigning ceremonial function to large piles of clean shell and that there are any number of reasons why large piles of “clean” shell would accrue as garbage. (Sanger [2015] has addressed many of Marquardt’s concerns.)

Details for how sites were used within the ceremonial spaces model vary considerably (e.g., Russo 2008, 2014; Sanger 2015; Saunders 2004a, 2014; Thompson 2007; Thompson and Andrus 2011). Rings in northeast Florida may have been ceremonial constructions from the beginning, and Rollins and Guana have possible residential sites nearby. I (Saunders 2004a; 2014) have stressed the integrative nature of these Florida structures, arguing that isolation from residential sites creates a neutral ground that promotes cooperation between groups. Other rings may have been sedentary villages and ceremonial spaces from their inception. Russo (2008, 2014) envisioned competitive feasting as the driving force in the erection of these rings. In a persistent places approach to shell rings, Thompson (2007, 2010) suggested that ring function changed through time. Rings began as villages and shell middens aggraded through time, first in pits beside structures and later in a ring shape. When a village was abandoned, its deep ties to the landscape and to the people rendered it a sacred space, and shell continued to be mounded onto the pre-existing village ring during ritual celebrations. Given the variability of rings through time and space, all of these arguments (and others) may be good approximations of ring functions.

At most rings, feasting involved quotidian foods—oyster and other denizens of the estuaries adjacent to the rings—in vast amounts. Small, net-able fishes provided the bulk of the protein. Occasional paired shells are found, but the vast majority of shell is presumed to be associated with subsistence.

Unlike most other shell structures in this discussion, feasting does not appear to be associated with mortuary rituals. Scattered human bone is found in rings, but these do not represent formal burials (see Russo 2006). The only confirmed burials are the cremations at the McQueens Ring on St. Catherines Island, Georgia (discussed in a bit more detail later). Indeed, Late Archaic burials are rare on the Atlantic and Gulf Coasts (Anderson 2010, 276). Possible Late Archaic burials have been found at the Daws Island site, in South Carolina, which is a sheet midden site very near two rings (Barrows and Patent Island). However, there has been no formal excavation at the Daws Island site, and there are no radiocarbon dates to confirm association with the rings. The only other significant Late Archaic burial population along the lower Atlantic coast is from a site that has not received the attention it should, the Summer Haven site, which was approximately 50 kilometers south of Guana Shell Ring (Bellomo 1995). Fourteen in situ burials were excavated from within and beneath an oyster and coquina midden, which, prior to heavy disturbance, was some 152 × 61 square meters of mounded shell 1.2 meters deep (Bellomo 1995; Bullen and Bullen 1961).

Although six rings have preceramic components,5 pottery is the most common artifact at rings, and three distinct potting traditions (considered cultures) were involved: Thoms Creek, Stallings/St. Simons, 6 and Orange. Michie (1979) and Saunders (2004b) argued that pottery is more elaborately decorated and more highly burnished at rings than at other contemporaneous site types. Trade items, including mundane lithic artifacts, are very rare. As for Classic Stallings Island in the interior, rings appear very insular. Coastal folk relied almost exclusively on their immediate environment for both secular and sacred objects.

Around 50 ring sites are known. The bulk of these are the small structures of the South Carolina and Georgia coast. Like all rings, these are on barrier islands or the mainland shore. The next set of rings, in northeast Florida, are much larger; this set consists of Oxeye (preceramic and almost completely inundated), Rollins, and Guana Shell Ring. Guana, near St. Augustine Florida, is the southernmost of these. There is a substantial break in the occurrence of rings south of Guana—Wyman’s interior U-shaped rings may fill this void. However, 300 kilometers south of Guana, another large ring appears, the Joseph Reed Shell ring. According to Russo (2006, 101), radiocarbon dates suggest that Joseph Reed may have been among the last rings built. The small pottery assemblage from the site consists of St. Johns sponge spicule-tempered pottery and Glades7 sandy pottery. This is the earliest manifestation of either type in south Florida; Joseph Reed is at least 130 kilometers south of the general production area of St. Johns pottery (Russo and Heide 2004).

Rounding peninsular Florida, in the Ten Thousand Island area just north of the Everglades, there are a set of as-yet poorly understood Archaic rings associated with later and larger Woodland–Mississippi period shell structures (Schwadron 2010). Because so little is yet known about these rings, they will not be dealt with here.

Moving further north along the Gulf Coast, the Horrs Island site contains evidence of some of the earliest shellfishing in Florida (ca. 7200 cal b.p.) along with a complex, pre-ceramic Late Archaic site containing a large, U-shaped shell ring, two linear ridges (Russo refers to one of these as a ramp), and four shell/sand ceremonial mounds (Russo 2006, 68). This array of different features is somewhat reminiscent of the complexity of the St. Johns River sites. Bonita Bay, north of Horrs Island, is another pre-pottery, large U-shaped ring associated with a single sand mound.

Hill Cottage is a poorly known, large, U-shaped ring near Sarasota, Florida. The site is overlain and disturbed by an early homestead and large botanical garden. Hill Cottage contains Orange fiber-tempered pottery but also has a substantial pre-pottery component (Bullen and Bullen 1976).

These four South Florida sites have some commonalities owing to geological and biological factors. Most have local limestone artifacts; the Gulf sites have a more diverse shellfish assemblage in the midden and a remarkable shell tool industry. The latter persists in this region throughout prehistory.

A large gap exists between Hill Cottage and the next possible rings, which are on the northern Gulf Coast: Buck Bayou on the Florida panhandle8 and Cedarland and Claiborne at the mouth of the Pearl River, in Mississippi. Although these sites are discussed when shell rings are discussed, they are a different phenomenon than the eastern rings. They clearly skew west and north in artifact assemblages and relationships, and their construction methods are quite different from the eastern rings.

Buck Bayou is more shell mound than ring (Figure 2). The area that might be defined as a plaza is not centered within the shell deposit and may be the result of disturbance (Russo 2006, 154). The site figures prominently in the current regional site settlement model as a population aggregation site of the Elliots Point culture (Campbell et al. 2004). A limited number of exotic artifacts, including copper beads, steatite bowl fragments, and microdrills from sites in the area have been used to suggest that the Elliotts Point culture was heavily involved in the Poverty Point culture trade network or was a “local expression” of Poverty Point culture (see discussion in Austin and Mikell n.d.). However, Russo (2006) noted that the bulk of the artifacts used to associate Elliotts Point culture with Poverty Point occur in many Late Archaic coastal sites and do not necessarily indicate especially strong ties with that enigmatic center. Indeed, the concept of Elliotts Point as a discrete culture is coming under increasing scrutinay and may not survive an onslaught of new data (Austin et al. 2014; Austin and Mikell n.d.). Notabley, this area is considered a possible source for marine shell in the SMA.

The Cedarland and Claiborne sites, at the mouth of the Pearl River (which divides Mississippi from Louisiana) are the westernmost ring sites known. Like Buck Bayou, these two sites were not rings in a strict sense. Shell midden was discontinuous both horizontally and vertically in the respective arcs. At Claiborne, two discrete linear midden areas were joined by an area of sterile sand. However, both sites were mounded shell and earth structures built by peoples who were actors in the larger Middle to Late Archaic scene.

Unfortunately, Cedarland was mined for shell before 1967, when it was first described (Gagliano and Webb 1970), and much of the information from these sites is based on private collections and salvage excavations carried out as the sites were being bulldozed for a port facility that never materialized. The two sites were C-shaped shell and earth rings separated by a 50 meter wide swale. Although similar in final configuration, their depositional histories are distinct. Cedarland was comprised of an oyster-shell midden core and an overlying stratum of black earth midden, whereas Claiborne was primarily composed of a black, sandy earth midden with a rangia (and some oyster) shell midden deposited primarily on its flanks. The two sites are very poorly dated, but the shift in mollusks has been considered to reflect delta formation processes, which produced a fresher-water regime through time (and ultimately formed the northeastern edge of Lake Pontchartrain). Thus, Cedarland, with its salt-tolerant oyster, was formed earlier than Claiborne, with its brackish water clam (Rangia cuneata). The radiocarbon dates9 are of little help. Still, Webb10 (1982) and Bruseth (1991, 14) have taken a 2040 ± 80 b.c. (uncorrected, uncalibrated) date to indicate that Claiborne was occupied before the major occupation of Poverty Point. However, the generalized dates for the pottery that appears at the site, Wheeler Fiber-Tempered and St. Johns early varieties, indicated that Claiborne should not be earlier than ca. 3500 b.p. (corrected, uncalibrated). However, Jenks (2006) recently reported a date of 4140 ± 40 b.p. (4820–4520 and 4470–4450 cal b.p.) from soot on a St. Johns sherd from Harris Creek. Although one cannot rewrite prehistory on the basis of a single sherd, the date does provide limited evidence that Claiborne could be significantly earlier than Poverty Point. (However, Wheeler pottery is currently dated to ca. 3700 cal b.p. (Kidder and Sassaman 2009.)

In any event, the idea that Cedarland was earlier than Claiborne was bolstered by a remarkable set of oppositions between the two sites that suggested Cedarland was in use in a generalized, but well-connected, Late Archaic, whereas Claiborne emerged later and interacted with the Poverty Point culture. Oppositions (presence/absence) include (Cedarland/Claiborne): oyster/rangia; clay-lined hearths/baked clay objects; no pottery/pottery; no steatite/steatite vessel caches; many red jasper beads, no effigy forms/few jasper beads, effigy forms; bipolar reduction/no bipolar reduction; no Jaketown perforators/Jaketown perforators; winged bannerstones/perforated “gorgets” (Bruseth 1991, table 1).

Claiborne is associated with a low sand mound about 170 meters east of the ring (Figure 2). The mound is aligned with a cache of 12 steatite vessels and other exotic items11 in an otherwise sterile area at the apex of the ring wall (Bruseth 1991, figure 6). The layout of these items suggested to the excavators that the artifacts were funerary items for burials that had eroded away. Notably, no one has suggested mortuary ritual involving feasting—with shellfish or any other foodstuff. However, the abundance of clay-lined hearths at Cedarland (which increase in higher strata [Gagliano and Webb 1970]) and the “enormous numbers” (Webb 1982, 35) of baked clay objects at Claiborne attest to cooking for large populations.

Bruseth (1991) identified both Cedarland and Claiborne as villages. Although Cedarland housed a local population, Bruseth argued that Claiborne was occupied by an intrusive group and that the ring walls were defensive. On the other hand, Webb (1982, 36) was convinced that the sites were sequential occupations of the same people. However, another possibility is that Cedarland and Claiborne were contemporaneous and either housed different societies or had different functions. That is, the differences between the two sites were social rather than temporal.

This lengthy recitation of site characteristics was provided because there is another set of ring sites that have oppositions that might suggest temporal differences similar to Cedarland and Claiborne, but are indisputably contemporaneous. The St. Catherines Island and McQueens shell rings, both on St. Catherines Island, Georgia, are well dated (ring deposition ca. 4400–3700 cal b.p.), similar in dimensions and shape, and less than 3 kilometers apart. These two sites also contained marked oppositions in material culture. For instance, both rings produced an unusually large lithic assemblage for the area, where lithics are “notoriously rare and homogeneous” (Sanger and Thomas 2010, 68). However, raw materials in the 5,000+ lithic assemblage from the St. Catherines ring were dominated by Coastal Plain chert, which can be obtained some 80 kilometers north of the island along the Savannah River (Sanger 2015). Lithic resources at McQueens were much more diverse and included “gray chert, metavolcanic materials, quartz, and quartzite” (Sanger and Thomas 2010, 68), resources that must come from the piedmont area further upstream. Baked clay objects were abundant at the St. Catherines ring and rare at McQueens. Stallings/St. Simons fiber-tempered pottery was present at both sites but approximately 99% of the pottery was plain at St. Catherines whereas 14% of the pottery at McQueens was decorated (Sanger and Thomas 2010). The pottery surface decorations and manufacturing techniques were also different.

Sanger (2015) hypothesized that the peoples who established the St. Catherines ring maintained earlier, Stallings culture traits, whereas the McQueens ring was established by peoples from outside the immediate St. Catherines Island area who brought new decorations, production techniques, and trading partners. Although there was clearly interaction between the peoples from the two rings, the population of each ring maintained a distinct community of practice and discrete social identities throughout their occupations. One might easily entertain the idea that Cedarland and Claiborne represent a similar phenomenon. Anderson (2010), however, suggested that the two Georgia rings had different functions, another good possibly for Cedarland and Claiborne.

The St. Catherines Island rings are quite possibly the best generalization-busters of all the rings; only a few of the details can be presented here. Profiles at both rings suggest more soil admixture in the shell than usual. Both St. Catherines Island rings began as mounded deposits, and the St. Catherines ring continued to be built by adding small mounds of diverse shell species interspersed with occasional small fire pits and areas of ash. However, sometime around 4200 cal b.p., the McQueens deposits were reworked and subsequent height was created by depositing broad, but thin lenses of shell and sand, possibly as many as 50, over a layer of clam shell (Sanger 2015, 192).

Although the evidence for cremations at the St. Catherines ring is tenuous, there is no doubt that the McQueens ring contained a central burial feature containing eight cremations: an upper, group interment of some seven individuals and a lower burial consisting of a young woman (Sanger 2015, 390, 407). The burial pit contained a copper object at the base of the upper burial. These are the only formal burials (as opposed to the occasional bone or tooth fragment) ever recovered from a ring site. Furthermore, cremation was not all that common in the South at this time, although it appears in SMA sites of Alabama (Claassen 2010, 127), in the Late Archaic Cowpen Slough site in Catahoula Parish, Louisiana (Mires 1991; Webb 1982, 32), possibly at Claiborne (Webb 1982, 36, 64), and some unidentified large mammal(s) were cremated on a platform at the base of the Middle Archaic Monte Sano Mounds site, in East Baton Rouge Parish, Louisiana (Saunders 1994).12

Another generalization-buster from the St. Catherines rings is the idea that there are few features in ring plazas. This idea is based partially on a lack of excavation. Many ring interiors are below the water table and cannot be excavated; others, like Rollins, are heavily wooded, and undisturbed access to the plaza surface is limited. Regardless, there is enough excavation in ring interiors to suggest that what was found at the St. Catherines rings is unusual, if not unique. High and dry, and only lightly plowed, the St. Catherines Island ring interiors share a pattern of features in three concentric circles. Dense pits occur along the ring interior next to the ring wall and a feature-free zone extends into the ring from there. The St. Catherines ring center contains an area of abundant, overlapping pits. At McQueens there is a single feature—the burial pit.

Sanger also recorded numerous features below the St. Catherines Island ring wall. These included several postholes, small hearths, and trash pits that did not occur elsewhere at the site. Like the Rollins sub-ring features, the St. Catherines features “anticipated” (Sanger 2015, 213, also pp. 215–216) the future configuration of the shell arc. From these data, Sanger suggested that a circular village may have preceded the construction of the St. Catherines ring from around 4800 cal b.p. to 4400 cal b.p.

Sanger (2015, 219) asserted that both of the St. Catherines Island rings, and perhaps most rings, were “ceremonial” villages, although this involved the redefinition of “village.” The lack of evidence for structures associated with the shell rings led him to conclude that the “villages” possibly did not have any long-term residents. Instead, populations eddied and flowed into and out of the ring, with the highest populations in the late fall (for mast harvest), winter, and early spring:

I suggest the possibility that although there was a stable communal fidelity to place at the rings, individuals and families likely moved into and out of the rings on a regular basis. As such, shell rings are perhaps best understood as a novel type of village defined less by continuous co-presence and habitation and more through short and long-term encounters in a locale imbued with a significant communal history.

(Sanger 2015, 233)

Indeed, this takes us back to the population flux settlement model proposed by Crothers and Bernbeck (2004) for SMA sites, which, if shorn of its insistence that shell sites are not purposeful constructions, might also serve to describe population movements through Claassen’s ritual and domestic landscape. It may also support Claassen’s (2010) contention that there were no villages in the Archaic. The Classic Stallings occupation on Stallings Island may be an exception, but, at least in some areas, Late Archaic peoples may have been less sedentary than they have been modeled of late (see, e.g., discussion in Tucker 2009).


The overarching goal of this chapter was to present the ways that researchers have replaced traditional, environmentally driven models of shell mounding in the Archaic with historical approaches that rely more on social explanations. As can be appreciated from the foregoing discussion, Archaic coastal folk are no longer seen as eking out a living in an unpredictable environment. Instead, they are modeled as masterful exploiters of a resource-rich ecosystem that allowed for population aggregation and an unprecedented rise in ceremonialism.

Sea level rise and the consequent initial formation of productive shellfishing locales are no longer considered the sole basis for the initiation of shellfishing and subsequent shell mounding in all areas. For instance, on the basis of site patterning, Claassen (2010, 83) maintained that shellfishing and shell mounding in the SMA was done for “socially determined reasons, not natural ones.” Shellfish were not marginal foods, nor was their prevalence a function of their ease of collection by otherwise lower-producing members of society. Instead, shellfish were symbols of regeneration and rebirth, and their inclusion in burials and other contexts was a purposeful act, replete with social meaning. Shell mounding was an act of reverence or even societal replication, not casual discard. Some shell mounds and rings may have begun as villages, but, ultimately, most became centers for population aggregation, ceremony, and feasting. That said, there is certainly enough time and enough space to admit considerable variation in individual sites and regions.

The degree to which the different regions were tied together in this symbolic system is difficult to ascertain. To a greater or lesser extent, long-distance trade was present in each region in the Middle Archaic. For instance, although the Harris Creek mortuary site in the St. Johns region contained few exotic items, it did contain two extralocal individuals. These were from Virginia or Tennessee, indicating ties with the interior and possibly with SMA peoples in Tennessee. Harris Creek also had a baked clay object sourced to Poverty Point (Hays et al. 2016). The St. Johns region and at least one ring are tied to Stallings Island through trade in hypertrophic bannerstones. But trade fell off markedly in the coastal plain cultures in the Late Archaic, except for four rings: McQueens and, to a lesser extent, St. Catherines, and Cedarland and Claiborne. It is tempting to look at these sites, located at the peripheries (northern13 and western) of the ring phenomenon, as some kind of boundary-maintaining or, conversely, gateway sites, where the material manifestations of diverse peoples or traditions were negotiated, displayed, and celebrated.

Elsewhere, the decline of trade in the Late Archaic may be a function of the rise and subsequent fall of the Benton Interaction Sphere and its derivatives (e.g., Stallings Island). Indeed, the demise of Middle Archaic mound building in the Lower Mississippi River Valley around 2800 cal b.c. and the subsequent 1,000-year hiatus in mound building (J. Saunders 2010) might suggest perturbations on a broader scale. Middle to Late Archaic transitions in the regions under discussion included the reorganization of St. Johns River Valley sites at the beginning of the Orange period, the rise of shell rings on the Atlantic and Gulf Coasts, and the development of Classic Stallings culture in the mid-Savannah River region. Claassen (2010, 47) noted a decline in shell deposition between 4499 and 4000 b.p., but, given the prevalence of hiatuses throughout the SMA and the poor resolution of the record after 4000 b.p., the significance of this decline is difficult to assess.

J. Saunders (2010) noted the correlation of the cessation of Lower Mississippi River Middle Archaic mound building with a strong El Nino Southern Oscillation (ENSO) event. The disappearance of shell mounding by the end of the Late Archaic has also been attributed to environmental factors: climate change and/or sea level change. Sea level regression is the most cited cause for the disappearance of shell rings (see papers in Thomas and Sanger 2010). However, Sanger (2010) found no correlation between sea level change and the ultimate abandonment of rings and emphasized social responses to environmental stress as decisive factors. Sassaman (2006) believed factionalism (“constant bickering”) was the major cause of the abandonment of Stallings Island. Claassen (2010) firmly rejected environmental explanations for the demise of the SMA. Among many other arguments, Claassen (2010, 82) noted that there was no evidence on the shellfish themselves that there was “any significant change in water conditions, river bottoms, or species mix during the Archaic or Woodland period.”

As part of her argument for social changes, Claassen (2010, 134) asked us to consider, if the SMA mounds were villages, why did village life come to such a “conclusive end”? (The same could be said if the mounds were base camps or indeed any manifestation of population aggregation; note that this does not apply to the St. Johns region.) But this “conclusive end” to shell mounding, when shell was clearly intensively exploited in the Woodland period, is its own topic. Whether the sites were abandoned because of climate change or vacillations in sea level, or because social systems were unable to suppress factionalism and endless bickering, the fact that these societies flourished for thousands of years and then completely disappeared, never to be even remotely replicated, is difficult to explain. Despite subsequent sea level rise, nothing like the SMA sites or the rings were ever built again.

Drawing on an analogy to the great Mississippian center of Cahokia, near St. Louis, Anderson (2010) suggested that shell mounding disappeared because environmental perturbations disrupted interaction networks, and, once disrupted, the networks could not easily be reconstructed or replaced. However, long distance trade declined by the beginning of the Late Archaic, with 2,000 years of shell mounding to go. Social and religious ties, however, may have taken more immaterial forms in the Late Archaic, and may have continued to bind peoples together. Although presently tenuous, I suspect that archaeologists will find more evidence of ties between the Archaic shell mounding regions. Use of more sophisticated tests for sourcing of artifacts and for people will demonstrate more movement of ideas, individuals, and items across the American Southeast.

To explain the lack of Woodland period shell mounding in the SMA region, Claassen (2010, 10) cited “a change in ritual practices that ended the ceremonial center concept and the large public feasts.” While acknowledging the differences (i.e., site location on southward-flowing rivers, log tombs, etc.), Claassen (2010, 216–217) highlighted the “shared geography” of the subsequent Adena (800–200 b.c.) burial mounds and noted that a number of characteristics of Adena mounds were likely derived from the SMA. These included isolation of mortuary sites from domestic sites, location of mortuaries by rivers, and long-term, accretional growth of mounds with shell and earth.

While Claassen stressed continuity, Russo (2010) noted that ancestor veneration as promulgated by the Adena and Hopewell (200 b.c.–a.d. 500) cultures constituted a significant break with the non-mortuary shell ring traditions. This veneration replaced “large-scale shellfish feasting associated with large public constructions” previously used for bonding peoples together (Russo 2010, 157). Given the lack of a widespread Archaic mortuary program in ring sites along the Atlantic Coast, it is not surprising that Adena-Hopewell mortuary programs gained little traction in the area.14


Emerson and McElrath (2014, 24) recently complained that models of Archaic societies were ensnared in “the tyrannical nature of the neo-evolutionary paradigm.” Archaic peoples adapted to environmental/economic conditions; agency was reserved for folks who left behind more than middens and simple tools. Although cultures who built mounded shell sites in each region had distinctive trajectories, archaeologists now argue that they shared an ethos that emphasized mounded monuments of shell as important points of contact between the past, present, and future. Archaeologists in the Southeast, as elsewhere, have begun to put culture in their models of Archaic societies, populating prehistory with peoples whose motivations are as likely to be social, political, or religious as environmental. Now, taking care not to conflate domestic garbage with ritual deposition (Marquardt 2010), some middens can be appreciated as monuments, purposefully constructed by societies marking their place in time and space.


Anderson, D. G. (2010). “The End of the Southeastern Archaic: Regional Interaction and Archaeological Interpretation.” In Trend, Tradition, and Turmoil: What Happened to the Southeastern Archaic?, edited by David Hurst Thomas and Matthew C. Sanger, pp. 273–302. New York: American Museum of Natural History Anthropological Papers no. 93.Find this resource:

Ashley, K. H. and N. J. Wallis. (2006). “Northeastern Florida Swift Creek: Overview and Future Research Directions.” The Florida Anthropologist 59(1): 5–20.Find this resource:

Aten, L. E. (1999). “Middle Archaic Ceremonialism at Tick Island, Florida: Ripley P. Bullen’s 1961 Excavations at the Harris Creek Site.” The Florida Anthropologist 52(3): 131–200.Find this resource:

Austin, R. J., and G. A. Mikell. (n.d.). “The Late Archaic of Northwest Florida and the Elliott’s Point Complex: A Reappraisal.” Unpublished paper in possession of author; used with permission.Find this resource:

Austin, R. J., C. Mickwee, and J. M. Torres. (2014). Cultural Resources Data Recovery at Bayou Park (8OK898), Eglin Air Force Base, Okaloosa County, Florida. On file, Florida Division of Historical Resources, Tallahassee.Find this resource:

Barton, R. N. E., P. J. Berridge, M. J. C. Walker, and R. E. Bevins. (1995). “Persistent Places in the Mesolithic Landscape: An Example from the Black Mountain Uplands of South Wales.” Proceedings of the Prehistoric Society 61: 81–116.Find this resource:

Beasley, V. R. (2008). “Monumentality during the Mid-Holocene in the Upper and Middle St. Johns River Basins, Florida.” PhD diss., Northwestern University.Find this resource:

Bellomo, R. V. (1995). Archaeological Investigations at the Summer Haven Site (8SJ46), an Orange Period and St. Johns Period Midden Site in Southeastern St. Johns County Florida. St. Petersburg, FL: Janus Research.Find this resource:

Bense, J. A. (1994). Archaeology of the Southeastern United States: Paleoindian to World War I. San Diego: Academic Press.Find this resource:

Bissett, T. G. (2014). “The Western Tennessee Shell Mound Archaic: Prehistoric Occupation in the Lower Tennessee River Valley between 9000 and 2500 Calendar Year BP.” PhD diss., University of Tennessee.Find this resource:

Blessing, M. E. (2010). “For Whom the Shell Tolls: The Use of Death Assemblages in the Deposition of Freshwater Shellfish.” Paper invited to the Symposium “Shell Hath No Theory like a Midden Formed: Alternative Approaches to Shell Site.” Presentation at the 67th Annual Southeastern Archaeological Conference, Lexington, KY, October 30, 2010.Find this resource:

Bruseth, J. E. (1991). “Poverty Point Development as Seen from the Cedarland and Claiborne Sites, Southern Mississippi.” In The Poverty Point Culture: Local Manifestations, Subsistence Practices, and Trade Networks, Geoscience and Man 29: pp. 7–26.Find this resource:

Bullen, A. K., and R. P. Bullen. (1961). “Summer Haven Site, St. Johns County, Florida.” Florida Anthropologist 14: 1–15.Find this resource:

Bullen, R. P., and A. K. Bullen. (1976). The Palmer Site. Gainesville: Florida Anthropological Society Publications 8.Find this resource:

Byrd, K. M. (1977). “The Brackish Water Clam (Rangia cuneata): A Prehistoric ‘Staff of Life’ or a Minor Food Resource.” Louisiana Archaeology 3: 23–31.Find this resource:

Cable, J. S. (1997). “The Ceremonial Mound Theory: New Evidence for the Possible Ceremonial Function of Shell Rings.” Poster for South Carolina Archaeology Week Shell Rings of the Late Archaic.Find this resource:

Campbell, L. J., P. M. Thomas, Jr., and J. H. Mathews. (2004). “Fiber-Tempered Pottery and Cultural Interaction on the Northwest Florida, Gulf Coast.” In Early Pottery Technology, Function, Style, and Interaction in the Lower Southeast, edited by Rebecca Saunders and Christopher T. Hays, pp. 129–149. Tuscaloosa: University of Alabama Press.Find this resource:

Claassen, C. (1991). “New Hypotheses for the Demise of the Shell Mound Archaic.” In The Archaic Period in the Mid-South: Proceedings of the 1989 Mid-South Archaeological Conference, Memphis, Tennessee, July 15, 1989, edited by C. H. McNutt, pp. 66–72. Jackson: Mississippi Department of Archives and History.Find this resource:

Claassen, C. (1992). “Shell Mounds as Burial Mounds: A Revision of the Shell Mound Archaic.” Current Archaeological Research in Kentucky 2: 1–12.Find this resource:

Claassen, C. (2008). “Shell Symbolism in Pre-Columbian North America.” In Early Human Impact on Megamolluscs, edited by A. Antczak and R. Cipriani, pp. 37–43. Oxford: Archaeopress.Find this resource:

Claassen, C. (2010). Feasting with Shellfish in the Southern Ohio Valley: Archaic Sacred Sites and Rituals. Knoxville: University of Tennessee Press.Find this resource:

Claassen, C. (2015). Beliefs and Rituals in Archaic Eastern North America: An Interpretive Guide. Tuscaloosa: University of Alabama Press.Find this resource:

Crothers, G. M. (1999). “Prehistoric Hunters and Gatherers, and the Archaic Period Green River Shell Middens of Western Kentucky.” PhD diss., Washington University.Find this resource:

Crothers, G. M. and R. Bernbeck. (2004). “The Foraging Mode of Production: The Case of the Green River Archaic Shell Middens.” In Hunters and Gatherers in Theory and Archaeology, edited by G. M. Crothers, pp. 401–422. Carbondale: Center for Archaeological Investigations, Southern Illinois University.Find this resource:

Edwards, W. E. (1965). “A Preliminary Report of the Sewee Mound Shell Midden, Charleston County, South Carolina.” Report submitted to the U.S. Forest Service. Ms. on file, South Carolina Institute of Archaeology and Anthropology, University of South Carolina, Columbia.Find this resource:

Emerson, T. E., and D. L. McElrath. (2014). “The Eastern Woodlands Archaic and the Tyranny of Theory.” In Archaic Societies: Diversity and Complexity Across the Midcontinent, edited by T. E. Emerson, D. L. McElrath, and A. C. Fortier, pp. 23–38. Albany, NY: SUNY Press.Find this resource:

Endonino, J. C. (2008). “The Thornhill Lake Archaeological Research Project: 2005–2008.” The Florida Anthropologist 61(3–4): 149–166.Find this resource:

Ford, J. A., and C. H. Webb. (1956). “Poverty Point, A Late Archaic Site in Louisiana.” Anthropological Papers of the American Museum of Natural History 46(Pt 1): 1–136.Find this resource:

Gagliano, S. M. and C. H. Webb. (1970). “Archaic-Poverty Point Transition at the Pearl River mouth.” In The Poverty Point Culture, edited by B. J. Broyles and C. H. Webb, pp. 47–72. Southeastern Archaeological Conference Bulletin 12.Find this resource:

Gilmore, Z. I. (2015). “Subterranean Histories: Pit Events and Place-Making in Late Archaic Florida.” In The Archaeology of Events: Cultural Change and Continuity in the Pre-Columbian Southeast, edited by Z. I. Gilmore and J. M. O’Donoughue, pp. 119–140. Tuscaloosa: University of Alabama Press.Find this resource:

Goad, S. I. (1980). “Patterns of Late Archaic Exchange.” Tennessee Anthropologist 5(1): 1–16.Find this resource:

Hays, C. T., R. A. Weinstein, and J. B. Stoltman. (2016). “Poverty Point Objects Reconsidered.” Southeastern Archaeology 35(3): 213–236.Find this resource:

Hensley, C. K. (1994). “The Archaic Settlement System of the Middle Green River Valley, Kentucky.” PhD diss., Washington University, St. Louis, MO.Find this resource:

Jeffries, R. W. (2009). “Archaic Cultures of Western Kentucky.” In Archaic Societies: Diversity and Complexity Across the Midcontinent, edited by T. E. Emerson, D. L. McElrath, and A. C. Fortier, pp. 635–666. Albany, NY: SUNY Press.Find this resource:

Jenks, C. J. (2006). “Rethinking Culture History in Florida: An Analysis of Ceramics from the Harris Creek Site (8VO24) on Tick Island in Volusia County, Florida.” PhD diss., University of Florida.Find this resource:

Jones, C. C. (1873). Antiquities of the Southern Indians: Particularly of the Georgia Tribes. New York: D. Appleton and Company.Find this resource:

Kidder, T. R., A. Ortmann, and T. Allen. (2004). “Testing Mounds B and E at Poverty Point.” Southeastern Archaeology 23(1): 98–113.Find this resource:

Kidder, T. R., and K. E. Sassaman. (2009). “The View from the Southeast.” In Archaic Societies: Diversity and Complexity Across the Midcontinent, edited by T. E. Emerson, D. L. McElrath, and A. C. Fortier, pp. 667–694. Albany, NY: SUNY Press.Find this resource:

Littleton, J., and H. Allen. (2007). “Hunter-Gatherer Burials and the Creation of Persistent Places in Southeastern Australia.” Journal of Anthropological Archaeology 26(2): 283–298.Find this resource:

Lyon, E. (1996). A New Deal for Southeastern Archaeology. Tuscaloosa: University of Alabama Press.Find this resource:

Marquardt, W. H. (1985). “Complexity and Scale in the Study of Fisher-Gatherer-Hunters: An Example from the Eastern United States.” In Prehistoric Hunter-Gatherers: The Emergence of Cultural Complexity, edited by T. D. Price and J. A. Brown, pp. 59–98. Orlando, FL: Academic Press.Find this resource:

Marquardt, W. H. (2010). “Shell Mounds in the Southeast: Middens, Monuments, Temple Mounds, Rings, or Works?” American Antiquity 75(3): 551–570.Find this resource:

Marquardt, W. H., and P. J. Watson. (2004). “The Green River Shell Mound Archaic: Interpretive Trajectories.” In Aboriginal Ritual and Economy in the Eastern Woodlands: Essays in Memory of Howard Dalton Winters, edited by A. -M. E. Cantwell, L. A. Conrad, and J. E. Reyman, pp. 113–122. Springfield: Illinois State Museum.Find this resource:

Marquardt, W. H., and P. J. Watson. (2005). “The Green River Shell Mound Archaic: Conclusions.” In Archaeology of the Middle Green River Region, Kentucky, edited by W. H. Marquardt and P. J. Watson, pp. 629–648. Institute of Archaeology and Paleoenvironmental Studies. Gainesville: University of Florida.Find this resource:

McKinley, W. (1872). “Mounds in Georgia.” Smithsonian Institution, Annual Report 27: 422–428.Find this resource:

McNutt, C. H. (2008). “The Benton Phenomenon and Middle Archaic Chronology in Adjacent Portions of Tennessee, Mississippi, and Alabama.” Southeastern Archaeology 27(1): 45–60.Find this resource:

Michie, J. L. (1979). The Bass Pond Site: Intensive Archaeological Testing at a Formative Period Base Camp on Kiawah Island, South Carolina. Research Manuscript Series 154. Columbia: Institute of Archaeology and Anthropology, University of South Carolina.Find this resource:

Middaugh, D. P. (2011). “Carolina Bays and the Sewee Shell Ring Site, South Carolina: Evidence for Natural and Planned Water Gathering.” Journal of North Carolina Academy of Science 127(3): 197–213.Find this resource:

Mikell, G. A., and R. Saunders. (2007). “Terminal Middle to Late Archaic Settlement in Coastal Northwest Florida: Early Estuarine Exploitation on the Northern Gulf Coast.” Southeastern Archaeology 26(2): 169–195.Find this resource:

Milanich, J. T. (1994). Archaeology of Precolumbian Florida. Gainesville: University Press of Florida.Find this resource:

Milner, G. R., and R. W. Jefferies. (1998). “The Read Archaic Shell Midden in Kentucky.” Southeastern Archaeology 17(2): 119–132.Find this resource:

Mires, A. M. (1991). “Sifting the Ashes: Reconstruction of a Complex Archaic Mortuary Program in Louisiana.” In What Mean These Bones? Studies in Southeastern Bioarchaeology, edited by M. L. Powell, B. D. Smith, G. R. Milner, C. S. Larsen, L. Eisenberg, and P. S. Bridges, pp. 114–130. Tuscaloosa: University of Alabama Press.Find this resource:

Moore, C. R. (2011). “Production, Exchange and Social Interaction in the Green River Region of Western Kentucky: A Multiscalar Approach to the Analysis of Two Shell Midden Sites.” PhD diss., University of Kentucky.Find this resource:

Moore, C. R., and V. D. Thompson. (2012). “Animism and Green River Persistent Places: A Dwelling Perspective of the Shell Mound Archaic.” Journal of Social Archaeology 12(2): 264–284.Find this resource:

Nord, J. (2009). Changing Landscapes and Persistent Places: An Exploration of the Bjäre Peninsula. Vol. 29. Lund: Lund University.Find this resource:

Parmalee, P. W., and W. E. Klippel. (1974). “Freshwater Mussels as a Prehistoric Food Resource.” American Antiquity 39(3): 421–34.Find this resource:

Peacock, E. (2002). “Shellfish Use During the Woodland Period in the Middle South.” In The Woodland Southeast, edited by D. G. Anderson and R. C. Mainfort, pp. 1–19. Tuscaloosa: University of Alabama Press.Find this resource:

Peres, T. M., and A. Deter-Wolf. (2016). “The Shell-Bearing Archaic in the Middle Cumberland River Valley.” Southeastern Archaeology 35(3): 237–250.Find this resource:

Quinn, R. L., B. D. Tucker, and J. Krigbaum. (2008). “Diet and Mobility in Middle Archaic Florida: Stable Isotopic and Faunal Evidence from the Harris Creek Archaeological Site (8Vo24), Tick Island.” Journal of Archaeological Science 35(8): 2346–2356.Find this resource:

Randall, A. R. (2008). “Archaic Shell Mounds of the St. Johns River, Florida.” SAA Archaeological Record 8(5): 13–17.Find this resource:

Randall, A. R. (2011). “Remapping Archaic Social Histories Along the St. Johns River in Florida.” In Hunter-Gatherer Archaeology as Historical Process, edited by K. E. Sassaman and D. H. Holly, Jr., 7:120. Tucson: University of Arizona Press.Find this resource:

Randall, A. R. (2013). “The Chronology and History of Mount Taylor Period (ca. 7400–4600 cal BP) Shell Sites on the Middle St. Johns River, Florida.” Southeastern Archaeology 32(2): 193–217.Find this resource:

Randall, A. R., K. E. Sassaman, Z. I. Gilmore, M. E. Blessing, and J. M. O’Donoughhue. (2014). “Archaic Histories Beyond the Shell ‘Heap’ on the St. Johns River.” In New Histories of Pre-Columbian Florida, edited by N. J. Wallis, and A. R. Randall, pp. 18–37. Gainesville: University Press of Florida.Find this resource:

Russo, M. (2006). Archaic Shell Rings of the Southeast U.S. National Historic Landmarks­National Register of Historic Places Theme Study. Tallahassee: Southeastern Archeological Center, National Park Service.Find this resource:

Russo, M. (2008). “Late Archaic Shell Rings and Society in the Southeast US.” SAA Archaeological Record 8(5): 18–22.Find this resource:

Russo, M. (2010). “Shell Rings and other Settlement Features as Indicators of Cultural Continuity between the Late Archaic and Woodland Periods of Coastal Florida.” In Trend, Tradition, and Turmoil: What Happened to the Southeastern Archaic?, edited by D. H. Thomas and M. C. Sanger, pp. 149–172. New York: American Museum of Natural History Anthropological Papers 93.Find this resource:

Russo, M. (2014). “Ringed Shell Features of the Southeast United States: Architecture and Midden.” In The Cultural Dynamics of Shell Middens and Shell Mounds: A Worldwide Perspective, edited by M. Roksandic, S. Mendonça de Souza, S. Eggers, M. Burchell, and D. Klokler, pp. 21–39. Albuquerque: University of New Mexico Press.Find this resource:

Russo, M., and G. Heide. (2004). “The Emergence of Pottery in South Florida.” In Early Pottery: Technology, Function, Style, and Interaction in the Lower Southeast, edited by R. Saunders and C. T. Hays, pp. 105–128. Tuscaloosa: University of Alabama Press.Find this resource:

Russo, M., G. Heide, and V. Rolland. (2002). “The Guana Shell Ring.” Historic Preservation Grant FO 126. Report submitted to the Florida Department of State, Division of Historical Resources, Tallahassee.Find this resource:

Sanger, M. C. (2010). “Leaving the Rings: Shell Ring Abandonment and the End of the Late Archaic.” In Trend, Tradition, and Turmoil: What Happened to the Southeastern Archaic?, edited by D. H. Thomas and M. C. Sanger, pp. 201–216. New York: American Museum of Natural History Anthropological Papers 93.Find this resource:

Sanger, M. C. (2015). “Life in the Round: Shell Rings of the Georgia Bight.” PhD diss., Columbia University.Find this resource:

Sanger, M. C., and D. H. Thomas. (2010). “The Two Rings of St. Catherines Island: Some Preliminary Results from the St. Catherines and McQueen Shell Rings.” In Trend, Tradition, and Turmoil: What Happened to the Southeastern Archaic?, edited by D. H. Thomas and M. C. Sanger, pp. 45–70. New York: American Museum of Natural History Anthropological Papers 93.Find this resource:

Sassaman, K. E. (2004). “Common Origins and Divergent Histories in the Early Pottery Traditions of the American Southeast.” In Early Pottery: Technology, Function, Style, and Interaction in the Lower Southeast, edited by R. Saunders and C. T. Hays, pp. 23–39. Tuscaloosa: University of Alabama Press.Find this resource:

Sassaman, K. E. (2006). People of the Shoals: Stallings Culture of the Savannah River Valley. Gainesville: University Press of Florida.Find this resource:

Sassaman, K. E. (2010). The Eastern Archaic, Historicized. Lanham, MD: Altamira Press.Find this resource:

Sassaman, K. E., M. E. Blessing, and A. R. Randall. (2006). “Stallings Island Revisited: New Evidence for Occupational History, Community Pattern, and Subsistence Technology.” American Antiquity 71(3): 539–565.Find this resource:

Sassaman, K. E., and A. 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. Gainesville: University Press of Florida.Find this resource:

Saunders, J. (2010). “Late Archaic? What the Hell Happened to the Middle Archaic?” In Trend, Tradition, and Turmoil: What Happened to the Southeastern Archaic?, edited by D. H. Thomas and M. C. Sanger, pp. 237–243. New York: American Museum of Natural History Anthropological Papers 93.Find this resource:

Saunders, R. (1994). “The Case for Archaic Period Mound Sites in Southeastern Louisiana.” Southeastern Archaeology 13(2): 118–134.Find this resource:

Saunders, R. (2004a). “Stratigraphy at the Rollins Shell Ring Site: Implications for Ring Function.” Florida Anthropologist 57(4): 249–270.Find this resource:

Saunders, R. (2004b). “Spatial Variation in Orange Culture Pottery: Interaction and Function.” In Early Pottery: Technology, Function, Style, and Interaction in the Lower Southeast, edited by R. Saunders and C. T. Hays, pp. 1–22. Tuscaloosa: University of Alabama Press.Find this resource:

Saunders, R. (2014). “Shell Rings of the Lower Atlantic Coast of the United States: Defining Function by Contrasting Details, with Reference to Ecuador and Japan.” In The Cultural Dynamics of Shell Middens and Shell Mounds: A Worldwide Perspective, edited by M. Roksandic, S. Mendonça de Souza, S. Eggers, M. Burchell, and D. Klokler, pp. 41–56. Albuquerque: University of New Mexico Press.Find this resource:

Saunders, R., and C. T. Hays, eds. (2004). Early Pottery: Technology, Function, Style, and Interaction in the Lower Southeast. Tuscaloosa: University of Alabama Press.Find this resource:

Saunders, R., and V. Rolland. (2006). “Exploring the Interior of the Guana River Shell Ring.” Report submitted to the Florida Department of State (Permit # 0405.11), Tallahassee: Division of Historical Resources.Find this resource:

Saunders, R., and M. Russo. (2011). “Coastal Shell Middens in Florida: A View from the Archaic Period.” Quaternary International 239(1): 38–50.Find this resource:

Saunders, R., and M. K. Wrenn. (2014). “Crafting Orange Pottery in Early Florida: Production and Distribution.” In New Histories of Pre-Columbian Florida, edited by N. J. Wallis, and A. R. Randall, pp. 183–202. Gainesville: University Press of Florida.Find this resource:

Schlanger, S. H. (1992). “Recognizing Persistent Places in Anasazi Settlement Systems.” In Space, Time, and Archaeological Landscapes, edited by J. Rossignol and L. A. Wandsnider, pp. 91–112. New York: Plenum Press.Find this resource:

Schwadron, M. (2010). “Prehistoric Landscapes of Complexity: Archaic and Woodland Period Shell Works, Shell Rings, and Tree Islands of the Everglades, South Florida.” In Trend, Tradition, and Turmoil: What Happened to the Southeastern Archaic?, edited by D. H. Thomas and M. C. Sanger, pp. 113–148. New York: American Museum of Natural History Anthropological Papers 93.Find this resource:

Thomas, D. H., and M. C. Sanger, eds. (2010). Trend, Tradition, and Turmoil: What Happened to the Southeastern Archaic? New York: American Museum of Natural History Anthropological Papers 93.Find this resource:

Thompson, V. D. (2006). “Questioning Complexity: The Prehistoric Hunter-Gatherers of Sapelo Island, Georgia.” PhD diss., University of Georgia.Find this resource:

Thompson, V. D. (2007). “Articulating Activity Areas and Formation Processes at the Sapelo Island Shell Ring Complex.” Southeastern Archaeology 26(1): 91–107.Find this resource:

Thompson, V. D. (2010). “The Rhythms of Space-time and the Making of Monuments and Places during the Archaic.” In Trend, Tradition, and Turmoil: What Happened to the Southeastern Archaic?, edited by D. H. Thomas and M. C. Sanger, pp. 217–228. New York: American Museum of Natural History Anthropological Papers 93.Find this resource:

Thompson, V. D., and F. T. Andrus. (2011). “Evaluating Mobility, Monumentality, and Feasting at the Sapelo Island Shell Ring Complex.” American Antiquity 76(2): 315–344.Find this resource:

Trinkley, M. B. (1985). “The Form and Function of South Carolina’s Early Woodland Shell Rings.” In Structure and Process in Southeastern Archaeology, edited by R. S. Dickens, Jr. and H. Trawick Ward, pp. 102–118. Tuscaloosa: University of Alabama Press.Find this resource:

Tucker, B. D. (2009). “Isotopic Investigations of Archaic Period Subsistence and Settlement in the St. Johns River Drainage, Florida.” PhD diss., University of Florida.Find this resource:

Waring, A. J., Jr. (1968). “The Archaic Hunting and Gathering Cultures: The Archaic and Some Shell Rings.” In The Waring Papers: The Collected Works of Antonio J. Waring, Jr., edited by S. Williams, pp. 243–246. Papers of the Peabody Museum of Archaeology and Ethnology No. 58. Cambridge: Peabody Museum, Harvard University.Find this resource:

Waring, A. J., Jr., and L. B. Larson. (1968). “The Shell Ring on Sapelo Island.” In The Waring Papers: The Collected Works of Antonio J. Waring, Jr., edited by S. Williams, pp. 263–278. Papers of the Peabody Museum of Archaeology and Ethnology No. 58. Cambridge: Peabody Museum, Harvard University.Find this resource:

Watson, P. J. (1992). “Archaic Period Mounds in the Southeast.” Discussant comments on the symposium. Southeastern Archaeological Conference, Little Rock, Arkansas.Find this resource:

Webb, C. H. (1982). “The Poverty Point Culture,” 2nd ed. Geoscience and Man 17.Find this resource:

Winters, H. D. (1968). “Value Systems and Trade Cycles of the Late Archaic in the Midwest. In New Perspectives in Archeology, edited by S. R. Binford and L. R. Binford, pp. 175–221. Chicago: Aldine.Find this resource:

Wyman, J. (1875). “Fresh-water Shell Mounds of the St. John’s River,” Florida Peabody Academy of Science Memoir 4.Find this resource:

Yesner, D. R., W. S. Ayres, D. L. Carlson, R. S. Davis, R. Dewar, M. R. González Morales, et al. (1980). “Maritime Hunter-Gatherers: Ecology and Prehistory [and Comments and Reply].” Current Anthropology 21(6): 727–750.Find this resource:


(1) Some, like Crothers, do not always use the term, but held similar views.

(2) The earliest dates for pottery on the coastal plain are ca. 5200 cal b.p., from the Rabbit Mount site located approximately 100 meters downriver from Stallings Island.

(3) Sassaman (2004) believes pottery was developed on the coast, at sites that are now inundated. A plain ware preceded decorated pottery.

(4) In the literature, the mortuary structures are sometimes referred to as “shell” and contrasted with the “sand” burial mounds.

(5) Russo (2006, 34) includes Claiborne as a seventh example, but fiber-tempered and St. Johns pottery were present at that site. The St. Johns is a trade ware, but the origin of the fiber-tempered is unknown.

(6) Explanation of the differences between these possibly distinct types requires more discussion than is appropriate here.

(7) A South Florida type generally considered the product of Woodland period peoples (i.e., after ca. 2500 b.p.).

(8) For Meig’s Pasture, see Russo (2006).

(9) Uncorrected, uncalibrated dates are 1240 +/− 130 b.c. for Cedarland and four dates ranging between 2040 and 1150 b.c. for Claiborne (Bruseth 1991, table 2).

(10) Webb’s conception of the relationship of Claiborne to Poverty Point is a bit unclear. He (Webb 1982, 68) also wrote “Claiborne was virtually a Poverty Point colony.”

(11) “Two sheet copper bracelets, a copper pendant, an oval shaped piece of galena, and a red jasper bead” (Bruseth 1991, 17). Gagliano and Webb reported 10 steatite vessels.

(12) Ford and Webb (1956) thought Mound B at Poverty Point was built over a crematory fire (Ford and Webb 1956). Kidder et al. (2004) reinterpreted the “ash” as an E horizon.

(13) Although the Sewee Ring is north of and at least partially contemporaneous with the St. Catherines Island rings.

(14) Swift Creek culture (a.d. 300–850; the culture begins and ends earlier in other areas) mounds on the northeast Florida coast contain many burials and some exotics, but these date considerably “after the heyday of Middle Woodland or Hopewellian interaction and ceremonialism” (Ashley and Wallis 2006, 12).