Front Cover
 Table of Contents
 Editor's page
 The new Florida public archaeology...
 Introducing Shields Mound (SDU12)...
 Early Mississippian faunal remains...
 An investigation of St. Johns and...
 Bone, antler, tooth, and shell...
 Grant Mound: Past and present -...
 Archaeological overview of Mt....
 Toward an interpretation of Shields...
 FSA 2005 award recipients
 Book reviews
 Back Cover

Group Title: Florida anthropologist
Title: The Florida anthropologist
Full Citation
Permanent Link: http://ufdc.ufl.edu/UF00027829/00192
 Material Information
Title: The Florida anthropologist
Abbreviated Title: Fla. anthropol.
Physical Description: v. : ill. ; 24 cm.
Language: English
Creator: Florida Anthropological Society
Conference on Historic Site Archaeology
Publisher: Florida Anthropological Society.
Place of Publication: Gainesville
Frequency: quarterly[]
two no. a year[ former 1948-]
Subject: Indians of North America -- Antiquities -- Periodicals -- Florida   ( lcsh )
Antiquities -- Periodicals -- Florida   ( lcsh )
Genre: periodical   ( marcgt )
Summary: Contains papers of the Annual Conference on Historic Site Archeology.
Dates or Sequential Designation: v. 1- May 1948-
 Record Information
Bibliographic ID: UF00027829
Volume ID: VID00192
Source Institution: University of Florida
Rights Management: All rights reserved by the source institution and holding location.
Resource Identifier: oclc - 01569447
lccn - 56028409
issn - 0015-3893

Table of Contents
    Front Cover
        Front Cover 1
        Front Cover 2
    Table of Contents
        Table of Contents 1
        Table of Contents 2
    Editor's page
        Page 147
        Page 148
    The new Florida public archaeology network - Judith A. Bense
        Page 149
        Page 150
    Introducing Shields Mound (SDU12) and the Mill Cove Complex - Keith H. Ashley
        Page 151
        Page 152
        Page 153
        Page 154
        Page 155
        Page 156
        Page 157
        Page 158
        Page 159
        Page 160
        Page 161
        Page 162
        Page 163
        Page 164
        Page 165
        Page 166
        Page 167
        Page 168
        Page 169
        Page 170
        Page 171
        Page 172
        Page 173
        Page 174
    Early Mississippian faunal remains from Shields Mound (8DU12) - Rochelle A. Marrinan
        Page 175
        Page 176
        Page 177
        Page 178
        Page 179
        Page 180
        Page 181
        Page 182
        Page 183
        Page 184
        Page 185
        Page 186
        Page 187
        Page 188
        Page 189
        Page 190
        Page 191
        Page 192
        Page 193
        Page 194
        Page 195
        Page 196
        Page 197
        Page 198
        Page 199
        Page 200
        Page 201
        Page 202
        Page 203
        Page 204
        Page 205
        Page 206
        Page 207
        Page 208
        Page 209
        Page 210
    An investigation of St. Johns and Ocmulgee series pottery recovered from the Shields Site (8DU12) - Vicki Rolland
        Page 211
        Page 212
        Page 213
        Page 214
        Page 215
        Page 216
        Page 217
        Page 218
        Page 219
        Page 220
        Page 221
        Page 222
        Page 223
        Page 224
        Page 225
        Page 226
        Page 227
        Page 228
        Page 229
        Page 230
        Page 231
        Page 232
        Page 233
        Page 234
        Page 235
        Page 236
        Page 237
        Page 238
    Bone, antler, tooth, and shell artifacts from the Shields Mound Site (8DU12) - Thomas E. Penders
        Page 239
        Page 240
        Page 241
        Page 242
        Page 243
        Page 244
        Page 245
        Page 246
        Page 247
        Page 248
        Page 249
        Page 250
        Page 251
        Page 252
        Page 253
        Page 254
    Grant Mound: Past and present - Robert L. Thunen
        Page 255
        Page 256
        Page 257
        Page 258
        Page 259
        Page 260
        Page 261
        Page 262
        Page 263
        Page 264
    Archaeological overview of Mt. Royal - Keith H. Ashley
        Page 265
        Page 266
        Page 267
        Page 268
        Page 269
        Page 270
        Page 271
        Page 272
        Page 273
        Page 274
        Page 275
        Page 276
        Page 277
        Page 278
        Page 279
        Page 280
        Page 281
        Page 282
        Page 283
        Page 284
        Page 285
        Page 286
    Toward an interpretation of Shields Mound (8DU12) and the Mill Cove Complex - Keith H. Ashley
        Page 287
        Page 288
        Page 289
        Page 290
        Page 291
        Page 292
        Page 293
        Page 294
        Page 295
        Page 296
        Page 297
        Page 298
        Page 299
        Page 300
    FSA 2005 award recipients
        Page 301
        Page 302
    Book reviews
        Page 303
        Page 304
        Page 305
        Page 306
        Page 307
    Back Cover
        Back Cover
Full Text






~N. /

The Shields Site (8DU12):
New Perspectives on the
Early St. Johns II Culture
of Northeastern Florida


THE FLORIDA ANTHROPOLOGIST is published by the Florida Anthropological Society, Inc., P.O. Box 6356, Tallahassee, Florida 32314.
Subscription is by membership in the Society. Membership is NOT restricted to residents of the State of Florida nor to the United States
of America. Membership may be initiated at any time during the year, and covers the ensuing twelve month period. Dues shall be
payable on the anniversary of the initial dues payment. Members shall receive copies of all publications distributed by the Society during
the 12 months of their membership year. Annual dues are as follows: student $15, individual $30, family $35, institutional $30,
sustaining $40 or more, patron $100 or more, life $500, and benefactor $2500. Foreign subscriptions are an additional $5 U.S. to cover
added postage and handling costs for individual, family, or institutional membership categories. Copies of the journal will only be sent
to members with current paid dues. Please contact the Editor for information on recent back issues.

Requests for information on the Society, membership application forms, and notifications of changes of address should be sent to the
Membership Secretary. Donations should be sent to the Treasurer or may be routed through the Editor to facilitate acknowledgment in
subsequent issues of the journal (unless anonymity is requested). Submissions of manuscripts should be sent to the Editor. Publications
for review should be submitted to the Book Review Editor. Authors please follow The Florida Anthropologist style guide (49[1], 1996,
pp. 37-43) in preparing manuscripts for submission to the journal and contact the Editor with specific questions. Submit four (4) copies
for use in peer review. Only one set of original graphics need be submitted. The journal is formatted using Word Perfect 12.0. All
manuscripts must be submitted in final form on DOS-formatted 3.5 inch diskettes or CDs in Word Perfect, Word, or ASCII format.
Address changes should be made AT LEAST 30 DAYS prior to the mailing of the next issue. The post office will not forward bulk mail
nor retain such mail when "temporary hold" orders exist. Such mail is returned to the Society postage due. The journal is published
quarterly in March, June, September, and December of each year.


President: David Burns, 15128 Springview St., Tampa, FL 33624 (daveburns@prodigy.net)
First Vice President: Scott Mitchell, Silver River Museum, Ocala, FL (mitchells@marion.kl2.fl.us)
Second Vice President: Phyllis Kolianos, 880 Seminole Blvd., Tarpon Springs, FL 34689 (koliano@attglobal.net)
Corresponding Secretary: Joanne Talley, P.O.Box 788, Hobe Sound, FL 33475 (jo@whiticar.com)
Membership Secretary: Beth Sarney, 5331 Lansdowne Way, Palmetto, Fl 34221 (excav8@comcast.com)
Treasurer and RegisteredAgent: vacant
Directors at Large: Gloria Fike, 2815 SE St. Lucie Blvd., Stuart, FL 34997 (GloriaFike@prodigy.net)(Three Years); Barbara
Purdy, 1519 NW 25th Terr., Gainesville, Fl 32605 (bpurdy@ufl.edu)(One Year)
Immediate Past President: Sheila Stewart, 2130 Burlington Avenue North, St. Petersburg, FL 33713 (sjstew@gte.net)
Newsletter Editor: Sheila Stewart, 2130 Burlington Avenue North, St. Petersburg, FL 33713 (sjstew@gte.net)


Editor: Ryan J. Wheeler, P.O. Box 6356, Tallahassee, FL 32314-6356 (rwheeler@nettally.com)
Book Review Editor: Keith Ashley, P.O. Box 10074, Savannah, GA 31412 (kashley@scad.edu)
Technical Editor: vacant
EditorialAssistant: George M. Luer, 3222 Old Oak Drive, Sarasota, FL 34239-5019 (gluer@grove.ufl.edu)
Printer: Durra-Print, 717 South Woodward Ave., Tallahassee, FL 32304
BulkMail: Capital City Mailing, 4013 Woodville Hwy, Tallahassee, FL 32311
Back Issue Sales: Palm Beach Museum of Natural History (www.pbmnh.org)


Albert C. Goodyear, Institute of Archaeology and Anthropology, University of South Carolina, Columbia, SC 29208 (goodyear@sc.edu)
Jerald T. Milanich, Florida Museum of Natural History, University of Florida, Gainesville, FL 32611 (jtm@flmnh.ufl.edu)
Jeffrey M. Mitchem, Arkansas Archeological Survey, P.O. Box 241, Parkin, AR 72373 (jeffmitchem@juno.com)
Nancy Marie White, Department of Anthropology, University of South Florida, Tampa, FL 33620-8100 (nwhite@chumal.cas.usf.edu)
Robert J. Austin, P.O. Box 2818, Riverview, FL 33568-2818 (bob@searchinc.com)

NOTE: In addition to the above Editorial Review Board members, the review comments of others knowledgeable in a manuscript's
subject matter are solicited as part of our peer review process.

VISIT FAS ON THE WEB: www.fasweb.org



0, fV \
QW *

Volume 58 Numbers 3-4 'NCE 19 1
September-December 2005

Editor's Page 147

The New Florida Public Archaeology Network. Judith A. Bense 149

SPECIAL ISSUE: The Shields Site (8DU12): New Perspectives on the Early St. Johns II
Culture of Northeastern Florida

Introducing Shields Mound (8DU12) and the Mill Cove Complex. Keith H. Ashley 151

Early Mississippian Faunal Remains from Shields Mound (8DU12). Rochelle A. Marrinan 175

An Investigation of St. Johns and Ocmulgee Series Pottery Recovered
from the Shields Site (8DU12). Vicki Rolland 211

Bone, Antler, Tooth, and Shell Artifacts from the Shields Mound Site (8DU12). Thomas E. Penders 239
Grant Mound: Past and Present. Robert L. Thunen 255
Archaeological Overview of Mt. Royal. Keith H. Ashley 265

Toward an Interpretation of Shields Mound (8DU12)
and the Mill Cove Complex. Keith H. Ashley 287
Doran: Windover: Multidisciplinary Investigations of an Early Archaic Florida Cemetery. Robert J. Austin 303

Browman and Williams: New Perspectives on the Origins ofAmericanist Archaeology. Wm. Brian Yates 304

Contributors 307
Cover: Spatulate celt from the Shields Mound.
Published by the
ISSN 0015-3893


This issue begins with an introduction to the new Florida
Public Archaeology Network. The Network's founder, Judy
Bense, not only explains the origins and goals of this new
program in public archaeology, but introduces us to Bill Lees,
the Network Director. I have been involved in the develop-
ment of the Network, and it is clearly a great opportunity for
FAS and FAS Chapter members to get further involved in
public archaeology.
Presented here is a special issue dedicated to research on
the Shields site in Duval County and associated sites of the
Early St. Johns II Period, assembled by Keith Ashley. These
papers grew out of a symposium held at the 2001 Southeastern
Archaeological Conference annual meeting, also organized by
Ashley. The main focus of the issue is the Shields Mound,
which was initially studied by antiquarian Clarence B. Moore.
The first four articles summarize Ashley's recent excavations
at Shields, as well as studies of zooarchaeology by Rochelle
Marrinan, a study of ceramics by Vicki Rolland, and a study

of bone and shell artifacts by Tom Penders. Buzz Thunen's
paper on his excavations at the Grant Mound and Ashley's
synthesis on the Mt. Royal site broaden the perspective of the
issue and provide a welcome visit to these well-known St.
Johns River area sites. Ashley summarizes some of this
information and discusses the ways in which these sites are
connected in the issue's final article. This issue concludes
with two book reviews by Bob Austin and Brian Yates. I think
that many readers will find the Shields Mound and Early St.
Johns II Period focus to be especially interesting.
August saw the passing of two long-time and active FAS
members. Sally McKeige, President and Founder of the
Southeast Florida Archaeological Society passed away on
August 14, 2005. Sally was acknowledged with an FAS
President's Award at the May annual meeting. Chuck Wilde,
FAS Director and member of the Kissimmee Valley Archaeo-
logical and Historical Conservancy passed away on August 20,
2005. Many of us had an opportunity to work with Sally and
Chuck; they will both greatly be missed.

September, 2005

ULTU G5 ? > pr





VOL. 58(3-4)



Department ofAnthropology, University of West Florida, 11000 University Parkway, Pensacola, FL 32514
E-mail: jbense@uwf.edu

Public archaeology-and good public archaeology-is not
a new concept in Florida. What is new and on the horizon,
however, is a statewide program focused solely on the delivery
of archaeology to the general public from regional centers
located throughout Florida.
In 2004 the Florida Legislature established and in 2005
provided continuing funding for the Florida Public Archaeol-
ogy Network (FPAN) through the University of West Florida.
The Network will be administered by the University of West
Florida through a Memorandum of Agreement (MOA)
recently developed between UWF and the Division of Histori-
cal Resources, Florida Department of State. In the next
several years, regional public archaeology centers will be
established throughout Florida to help stem the rapid deterio-
ration of this state's buried past and to expand public interest
in archaeology. This will be achieved by a robust program of
public education, through the promotion and development of
archaeology-based heritage tourism, by active involvement of
the public through volunteerism, and through education and
outreach to local government officials, land managers, and
others whose actions affect the archaeological record.
Development and initial implementation of the MOA was
undertaken by a Steering Committee composed of representa-
tives from the Florida Archaeological Council, the Florida
Anthropolgical Society, the Division of Historical Resources,
and the University of West Florida. As provided by the MOA,
the Steering Committee has been replaced by an 11 person
Board of Directors, consisting of five core directors (three
appointed by and representing the University of West Florida,
one appointed by the Florida Anthropological Society, and one
appointed by the Florida Archaeological Council); two at-
large directors who are to be Florida residents; and two at-
large directors who are to be non-Florida residents. The
Network Director and the State Archaeologist serve as ex
officio core directors.
Three charter regional centers will be selected by the Board
of Directors from proposals received from institutions or
organizations willing to provide, at no cost, space for the
center. In exchange, the host institution will receive an annual
grant from the University of West Florida to provide for
professional staff and operation of the center. Other regional
centers will be selected by the Network Director and State
Archaeologist, based on recommendations from the board.
The University of West Florida has hired Dr. William B.
Lees, RPA, as Director of the Florida Public Archaeology
Network. His task will be to oversee the transformation the
FPAN from concept to reality. Dr. Lees comes to UWF from

William B. Lees, Florida Public Archaeology Network

Cultural Resource Analysts, Inc. in Lexington, Kentucky.
Starting in the mid-1980s, his career has been inextricably
linked with public archaeology. As an archaeologist with the
Kansas State Historical Society (1985-1993) he was involved
with and eventually directed the Kansas Archaeology Training
Program, a cooperative program with the Kansas Anthropo-
logical Association. Later, as Historic Sites Director with the
Oklahoma Historical Society (1993-2003), he administered a
state-wide public history/heritage tourism program and
developed within that an active public archaeology program
including involvement with the Oklahoma Anthropological
Society. Dr. Lees is an historical archaeologist who has
worked on both terrestrial and underwater projects in the Great
Plains and Southeast. He currently serves on the board of
directors and as Newsletter editor for the Society for Historical
Archaeology, and is a past president of the Register of Profes-
sional Archaeologists and the Plains Anthropological
Society-and he has recently submitted his membership
application to the Florida Anthropological Society!
For more information on the Florida Public Archaeology
Network, Dr. Lees may be contacted at wlees@uwf.edu.


VOL. 58(3-4)





Department ofLiberal Arts, Savannah College ofArt and Design, Savannah, GA 31402-3146
E-mail: kashley@scad.edu

Looming large in a residential area of southern Jackson-
ville, the Shields Mound (8DU12) was originally excavated by
Clarence B. Moore in 1894 and 1895. This large mound and
causeway complex, along with Mt. Royal (8PU35) in Putnam
County and Thursby Mound (8V036) in Volusia County, are
touted today as "the three best known St. Johns IIb [A.D.
1050-1513] mounds" in Florida (Milanich 1994:269). A
fourth frequently referenced St. Johns II earthwork is the
Grant Mound (8DU14), located less than a kilometer from the
Shields Mound. The distinction of being "best known" is
somewhat misleading, however. While all four mounds were
extensively dug by Moore-with many of the recovered
artifacts featured in his and other publications-virtually
nothing is known about the villages associated with these
major St. Johns II Period mounds. In an attempt to rectify this
situation, systematic shovel testing and unit excavations were
undertaken in the vicinity of the Shields Mound between 1999
and 2002. These excavations yielded an impressive array of
artifacts including St. Johns II and Ocmulgee pottery types
(Rolland 2005, this issue), various bone and shell artifacts
(Penders 2005, this issue), stone points and debitage (Bland
2001), scraps of copper, and an abundance of vertebrate faunal
remains (Marrinan 2005, this issue).
This paper begins with a detailed review of C.B. Moore's
excavation of the Shields Mound. I next chronicle the history
of archaeological investigations in the vicinity of the mound,
with an emphasis on the findings of excavations conducted
between 1999 and 2002. The spatial focus of the paper is then
widened to include sites situated between Shields Mound on
the east and Grant Mound on the west (see Thunen 2005, this
issue, for a discussion on the Grant Mound). Archaeological
evidence along with a series of radiometric assays indicates
that the Shields and Grant mounds are contemporaneous and
spatially linked. Finally, the term Mill Cove Complex is
introduced to refer to the collectivity consisting of the two
large sand mound and intervening St. Johns II Period middens
that date to ca. A.D. 900-1250.

Ecological Setting

The Shields (8DU12) and Grant (8DU14) mounds are
spaced about 750 m apart on the western edge of Mill Cove
(Figure 1). Physiographically, the southern shoreline along
this stretch of the lower St. Johns River encompasses Pleisto-
cene beach ridges and paleo sand dune fields (Brooks 1981).
The area is described as undulating, which is unusual for this

part of Florida where the land surface typically has less than
one percent slope. The northern faces of the high sand ridges
that front the river are exposed as steep bluffs that in areas rise
some 16 m above the water (Figure 2). These bluffs have been
subject to severe scouring by river flow, and the rate of erosion
has been exacerbated over the past century by intentional
channel modification to facilitate ship traffic. The Shields and
Grant mounds were erected upon the two highest points along
the west side of Mill Cove, each perched with a magnificent
view of the river.
The upland ecology of the Mill Cove vicinity is character-
ized as a maritime hammock, consisting of large live oak,
laurel oak, red bay, and magnolia trees. Turkey oak and other
xeric vegetation dominate on the higher, excessively well-
drained sand ridges. The natural vegetation has been altered
to varying degrees in the past half century due to suburban
development. The properties situated between the two mounds
currently consist of two types of residential development. The
areas surrounding the Shields Mound and immediately east of
the Grant Mound consist of long linear, wooded lots, with
minimal site damage other than a riverfront house, a garage or
two, and a septic tank and drain field (Figure 3). These are
representative of the traditional house lots that formerly gave
the entire Mill Cove vicinity a bucolic air. More recent
residential communities, comprised of tightly clustered houses,
have claimed much of the Grant Mound itself and loci to the
south, as well as an approximately 200-m wide wedge situated
between the two mounds.
The most distinguishing ecological feature of the area is
Mill Cove, a natural indentation or meander within the
southern side of river channel, approximately 15 km west of its
mouth. The eastern half of the cove is bordered by expansive
salt marshes, whereas its western shoreline consists mostly of
mucky tidal flats with a thin marsh border. The integrity of
Mill Cove has been greatly altered over the past century due to
river maintenance, channel dredging, and spoil island forma-
tion, which, combined, have caused extensive silting. During
the early decades of the twentieth century, Mill Cove was the
scene of a thriving local fisheries industry. But what was once
"a fish-laden body of clear water whose depths ranged up to 18
feet" with white sandy beaches is now a mucky slough less
than a half-meter in depth at low tide (Anonymous 1986). The
presence of shell middens and mounds along the river banks
attests to the former productivity of Mill Cove and the broader
St. Johns River estuary, a rich and diverse aquatic environ-
ment that provided the local natives with resources beyond


VOL. 58(3-4)




Figure 1. Location of Shields Mound, Grant Mound, and other St. Johns H sites in northeastern Florida.

their immediate needs.
To date, more than 40 prehistoric midden and mound sites
of various time periods have been recorded along the upland
fringes of Mill Cove. Of the 18 St. Johns II Period sites
reported in the Florida Master Site File, 15 are tightly clus-
tered between the Shields and Grant mounds on the west side
of the cove. Based on their sheer size, number of mound

interments, and presence of exotic materials, the Shields and
Grant mounds dominated the cultural landscape of the Mill
Cove vicinity, as well as broader northeastern Florida' during
the early St. Johns II Period. It should be noted that in
contrast to the broader St. Johns River basin the St. Johns II
Period in northeastern Florida is restricted to circa A.D. 900-
1250 (Table 1).

2005 VOL. 58(3-4)



Figure 2. Bluff along northern edge of Shields site (view to the southeast).

C.B. Moore and the Shields Mound

In 1893, C.B. Moore (1894:204-205) briefly visited and
dug into an "Indian earthwork of great size" that he referred
to as "Mound near Mill Cove." But intensive digging of this
mound did not occur until the next year, when he and 31 men
spent 17 seven-hour days at the sand tumulus now called the
"ShieldsMound" (Moore 1895:452-468). According to Moore
(1895:455), the Shields Mound was 5.5-m tall and structurally
quite complex. It was "slightly oblong with rounded covers,"
and had a basal diameter of 65.2 m and a summit measure-
ment of 40.5 by 35.1 m. He characterized it as "a great
platform mound entirely unlike in form any aboriginal
earthwork on the river." Although a house atop the "platform
mound" had recently burned, Moore (1894:205) was rather
adamant about it being a flat-topped mound and asserted "it is
not probable that the mound was in any way leveled for its
[house] reception, since examination of the steep and symmet-
rical sides showed no appearance of deposits from above."
The mound was situated upon "high rolling ground,"
approximately 137 m south of the river bluff (Moore
1894:204; 1895:454). Admittedly, rolling does not often come
to mind when one considers the topography of northeastern
Florida, but, as previously discussed, it appropriately describes
the relic dune terrain that sweeps across this section of the
lower St. Johns River. Figure 4 depicts Moore's rendering of
the Shields Mound (see Morgan 1999:210 for an idealized
rendition of the Shields Mound based on Moore's data).

A "carefully graded approach" or earthen ramp, 38.2 m-
long and 26.7-m wide, led to the mound summit from the
north, whereas a series of "curious ridges" extended off the
other side of the mound in a south-southwest direction. Of the
latter, the main southern ridge is unique and shaped like a
fishhook. It began at the base of the mound with an elevation
of about 30 cm and continued southwesterly for about 150 m
where it rose to a maximum height of 4.1 m, before it turned
sharply back to the north and gradually descended until it
reached surface grade at the hook tip. According to Moore
(1895:454), the interior slope of the hook was "so abrupt as to
be difficult of ascent."
Another ridge apparently continued southwesterly at the
point where the main ridge began its hook turn. As this ridge
continued to descend gradually for about 100 m it remained
rather amorphous in shape, but eventually formed a discernible
low ridge that was about 10-m wide and 15 to 25 cm high.
Paralleling this was another low ridge that's shape was clearly
visible back closer to the hook. Although the two low ridges
were soon "lost in the surrounding territory," Moore
(1895:454) speculated that together they "served as a covered
way" about 20-m wide that led to a "small lake," some 546 m
from the platform mound.
Although the "platform mound" exists today (Figure 5), it
has been impacted by the construction of a house along the
summit's southern edge. Connected to the house, at ground
level, is a basement-like garage set into the southern slope of
the mound. A dirt road has been graded into its eastern slope,




Figure 3. Shields site environment, with Kinzey's Knoll in foreground and Reeves Rise in the background (view to the north).

with a cement wall erected along the western side of the road
to curtail erosion from above. According to one landowner,
the summit was modified and scraped to a lower grade in the
1960s to accommodate construction of the house (Kinzey
Reeves, personal communication, 1999). Portions of the
"ridges" nearest the mound also are still visible as part of an
open yard. The hook segment and its interior ravine are
discernible, but currently heavily overgrown with forest cover.
Although partly obscured by vegetation, the layout of this area
does not appear as well defined as Moore's sketch leads us to
believe. The southernmost extent, including the two low
ridges, have been destroyed or altered beyond recognition by
road construction and suburban development.
Moore perceived the mound complex to be solely the result
of aboriginal construction, but this may not be entirely true.
It appears that the mound and main ridge are part of a natural
relic dune formation that parallels the present Atlantic
coastline. While a considerable amount of fill may have been
added to create the platform mound, much of the main ridge
may have already existed in some form as a sand ridge.
However, the landform was sculpted to produce the desired
shape. Thus, as was the situation with the Grant Mound, the
builders of the Shields Mound may have taken advantage of a
prominent natural feature in the creation of "monumental
architecture." If this was indeed the case, this may explain
why the Grant Mound was positioned directly on the bluff and
the Shields Mound was placed over 100 m away; the builders

intentionally selected the highest points for mound construc-
Each mound clearly provided a commanding view of the
river, and each was bordered on the east by a very abrupt and
rather dramatic drop in natural elevation. However, this does
not mean that native peoples did not add to, modify, or sculpt
existing landforms; evidence clearly indicates that they did put
substantial effort into building the mounds. By working the
earthworks into the "rolling" natural topography, a grander
edifice was erected with less work. Depending on one's
vantage point, the Shields Mound exhibits varying heights and
appearances, but is most impressive when viewed from the
river to the north and the east.
Moore (1895:455) dug the bulk of the eastern slope of the
Shields Mound, including the easternmost 3 m of the summit
plateau, to a depth of approximately 5.2 m below the summit.
From that point, the entire mound top was excavated to depths
between 1.8 and 2.4 m below the summit surface. Although
"no uniform stratification" was apparent, Moore, in his own
convoluted and at times contradictory way, describes three
general strata. The deepest stratum (herein referred to as Zone
III) was a light yellow sand described as both "free from
admixture of any foreign substance" and "with occasional
particles of charcoal." Although Moore's measurements
varied, depending on location, it appears that Zone II was
encountered between 2.4 and 4.3 m below the summit, and
represents sterile dune soil.


2005 VOL. 58(3-4)

Table 1. Radiocarbon assays for St. Johns II contexts in northeastern Florida.

Material Measured C14 C13/C12 Conventional Calibrated Calibrated
Site Beta # Age (BP) Ratio (o/oo) C14 Age (BP) 1 Sigma (AD) 2 Sigma (AD) Reference
8DU12 137818 oyster 1080 + 80 -2.5 1450 + 90 865 -1035 765 -1155 Ashley 2002
8DU97 167858 AMS soot 1090 +40 -23.0 1120+ 40 890 980 810 -1000 Ashley 2002
8DU14 131314 oyster 1060 + 60 -3.7 1430 + 60 905 -1025 830 -1065 Thunen 2001
8DU12 141594 oyster 1000 + 60 -2.4 1370+ 60 990 -1070 910 -1170 Ashley 2002
8DU68 54644 oyster 990+ 60 -2.4 1360+ 60 990-1080 905-1185 Ashley 1995
8SJ14 105269 oyster 950 + 60 1360 + 60 1000 1085 930-1180 Dickinson & Wayne 1997
8DU66 167857 AMS soot 940 + 40 -25.3 940 +40 1030-1160 1010-1190 Ashley 2002
8SJ14 105270 oyster 850 + 50 1260 + 60 1065 1220 1025 1275 Dickinson & Wayne 1997
8DU12 165353 oyster 870 + 60 -2.0 1250 + 60 1070 1230 1030 1280 Ashley 2002
8SJ4772 178427 tagelus 850 + 60 -0.4 1250 + 60 1070 1230 1030 1280 Handley et. al 2002
8NA59 126311 oyster 820 + 60 1230 + 60 1105- 1250 1045 -1290 Dickinson & Wayne 1999
8DU626 180802 oyster 820 + 60 -1.8 1210 + 70 1125 -1275 1040 1310 Ashley 2003
8DU14 118371 oyster 870 + 60 -4.0 1210 + 65 1160 -1275 1065 1305 NEFAS 2001
8DU12 165354 oyster 850 + 60 -3.4 1210 + 60 1160 1270 1060 -1300 Ashley 2002
8DU1 120267 clam 790 + 60 -0.6 1190 + 60 1170 -1285 1065 1320 Ashley & Thunen 1999
8DU67 131463 oyster 770 + 60 +0.9 1190+ 60 1170 1285 1070 1310 Ashley 2002
8DU12 165352 oyster 750 + 60 -2.8 1120 +60 1240 1310 1180 -1390 Ashley 2002
* C13/C12 ratio estimated by Beta Analytic, Inc.
** Uncalibrated date previously interpreted as Swift Creek (Ashley 1995:26, 1998:200)


Figure 4. Plan drawing of Shields Mound (Moore 1895).

Surmounting Zone III, beneath the center of the summit,
was a "stratum of dark yellow sand," 0.9 to 1.5-m thick, with
considerable charcoal flecking (herein referred to as Zone II).
The basal section of this zone contained a thin charcoal-rich
layer "taken as indicating the base of the mound" (Moore
1895:45). Within the eastern slope area, however, he de-
scribes Zone II as containing "irregular and local strata"
consisting of yellow sand, charcoal, shell, and gray sand
darkened by fire and extending from 0.8 to 4.3 m below the
eastern summit floor (Moore (1895:456). It was his belief that
"these strata probably extend through the mound." In his
concluding remarks, he claims that Zone II lacked burials and
displayed strata of shell midden and fireplaces, "all the marks
of prolonged occupation" (Moore 1895:467).
In support of this argument, Moore cites his discovery of"a
large bed of oyster shells" near the center of the summit, about
2.1 m below its surface, in which he discerned a posthole, 20-
25 cm in diameter and 1.4 m deep. He attributed this to
"domiciliary" activities that preceded the summit's use as a

mortuary facility (1895:467). Contrary to Moore's belief, the
shell midden may represent either a premound surface related
to the initiation of mound use or a deposit related to a specific
episode of mound activity. Because of the ambiguity of some
of Moore's statements, it is unclear whether Zone II represents
mound fill free of interments or premound contexts, but it
appears that Moore favored the latter interpretation.
Above Zone II was "mortuary" mound fill (herein referred
to as Zone I) that consisted of an eclectic assortment of colored
sands, including: yellow, dark yellow, chocolate, white, and
red, all of which "varied at every stage of the digging." He
encountered both pockets and thin lenses of colored sands, as
well as an irregular layer of "light chocolate to brick red"
sand that covered much of the central and eastern part of the
summit. The latter stratum, colored various hues of red
through the addition of hematite, approached a thickness of
1.5 m in certain areas of the summit.
With regard to burials, Moore ran across a "few inter-
ments" along the outer margins of the mound, "possibly two




/.'! ;,
/ i .p"'/


/ 25m

2005 VoL. 58(3-4)



Figure 5. Shields Mound, 2004 (view to the southeast).

dozen" burials within the eastern slope-summit area, and an
undisclosed number of human remains at over "150 points"
within the main part of the mound. With regard to the latter
two loci, burials were restricted to Zone I. In fact, all human
remains were encountered within 0.9 m of the summit surface,
except for four interments revealed at a depth of 1.8 m below
surface. Unlike Grant Mound, secondary or disarticulated
burials appear to have outnumbered primary interments in the
Shields Mound. Unfortunately, because ofMoore's vagueness,
the actual number and distribution of burials are uncertain. A
few pathological bone specimens were observed as were a
number of "platycnemic tibiae," but no intact crania were
recovered (Moore 1895:456-457). Unfortunately, we are
unable to glean any sex or age information from Moore's
descriptions of the burial population at the Shields Mound.
As was the case at Grant Mound, Moore's excavation at
Shields Mound produced an array of local and nonlocal
materials that included various items and fragments of copper;
smaller amounts of galena and mica; 119 stone projectile
points; polished stone celts, and other ground stone imple-
ments; marine shell beads; whelk cups; ceramic and stone
pipes; and nine ceramic pots, a tureen shaped clay pendant,
and a bird shaped clay vessel.
Moore (1895:457) mentioned coming across occasional
sherds, especially in association with "midden refuse." Plain
sherds predominated, but check stamped and complicated
stamped also were found. According to Moore, these latter
sherds exhibited patterns unlike those found at other nearby
mounds (i.e., Late Swift Creek). The specimen illustrated in
Moore's volume is a classic example of charcoal-tempered
Early Swift Creek, which indeed typically displays design
patterns of poor quality compared to those of Late Swift Creek.

Although ambiguous, it appears that these pottery fragments
were coming form both Zones I and II, and represent either
unintentional inclusions in mound fill, submound refuse, or
perhaps instances of both.
Among the most impressive Shields Mound finds were two
long stemmed spatulate celts or what Moore (1895:461)
termed "spade-shaped implements." The Shields Mound
specimens, "probably of Saussurite," each demonstrated a
broad round bit with four tick marks along its sides, pro-
nounced barbs, and a slender, elongated poll or stem with a
rough or unfinished end. Moore (1895:461) provided a
drawing of the larger specimen (Figure 6), whereas Goggin
(1952:Plate 10) presents a photograph of each. Spatulate celts,
erroneously called spuds due to the early belief that they
represented an agricultural tool, have been found in mounds
throughout the Mississippian Southeast and often have been
interpreted as badges or emblems of office or high status
(Brown 1976:126; Brown et al.1990:264; Larson 1971;
Pauketat 1983; Pebbles 1971). Moore also recovered three
spatulate celts from Mt. Royal (see Ashley 2005a, this issue).
The presence of these and other exotic materials from the
Shields Mound indicates the existence of external exchange
relations that allowed the acquisition of a variety of non-local
metal, stone, and minerals from far-flung areas of eastern
North America during the period A.D. 900-1250.

Shields Mound after C.B. Moore

Following Moore, Shields Mound garnered the attention of
several archaeologists, but always from afar (e.g., Goggin
1952; Milanich 1994; Milianich and Fairbanks 1980; Wil-
liams and Goggin 1956). Either the mound's unique layout or




N i

2005 VOL. 58(3-4)

Figure 6. Spatulate celt from the Shields Mound [24 cm long] (Moore 1895:461).

i -.i


Figure 7. Archaeological sites near the Shields Mound.

its unusual artifacts was used in cultural syntheses on the St.
Johns II culture or on the Mississippian Period in general, with
total reliance on information generated by Moore's investiga-
tions. Additionally, in his synthesis of Hopewell traits in
Florida, Brose (1979) used Moore's excavation results to
misclassify the Shields Mound as Middle Woodland.
No formal testing or professionally run excavations were
undertaken until 1988, when limited shovel testing was
performed in the vicinity of the mound (Johnson 1988:93-
106). During this grant-sponsored project, residential yards
surrounding the Shields Mound were subjected to surface
reconnaissance and random shovel testing that resulted in the
documentation of 6 previously unrecorded sites, 8DU5603-
5608 (Figure 7). The small size and intermittent distribution
of these sites, more appropriately middens, are the byproduct
of limited shovel testing, since it appears that a only day or

two was spent in the Shields Mound vicinity.
The Shields Mound (8DU12) itself also was visited and
sampled, but no mention was made of the investigative
strategy involved in the collection of 32 artifacts, the "major-
ity" of which "came from the earthwork" (Johnson 1988:68).
The presence of a dirt road along the eastern edge of the
mound and other areas of ground surface exposure suggest that
the artifacts were surface collected. Although recorded as
seven separate sites, all are contained within an area that
measured approximately 300 by 300 m. As discussed below,
recent testing of this same location has shown that they are all
part of one large St. Johns II Period site.

Testing of the Shields site, 1999-2002

While worthwhile information can be gleaned on exchange




Figure 8. Shovel test and shell midden locations, Shields site.

networks and ceremonial life of the St. Johns II people by
examining C.B Moore's mound data, he provided no informa-
tion on villages or habitation contexts at Shields Mound. To
remedy this void, archaeological testing of non-mound loci at
the Shields site was initiated in July 1999 and continued
intermittently until February 2002. This investigation in-
volved, at one time or another, students from the University of
Florida, Florida State University, and the University of North
Florida (UNF), along with local residents and other volunteers.
Episodes of fieldwork were typically restricted to long week-
ends, although a three-week UNF field school was conducted
at the site in the summer of 2001.
In total, 69 50-cm square shovel tests (1-m deep) were
excavated within a riverfront section of the Shields site,
encompassing five contiguous house lots north and west of the
burial mound. In general, shovel tests were placed 25-m apart,

although some adjustments were made to compensate for areas
of disturbance (e.g., paved driveways, fences, septic fields,
garage, etc.) and to sample certain loci. The investigated
section of the site measured about 300 m (E-W) by 250 m (N-
S) and was bounded by the river to the north, a dramatic drop
in elevation to the east, a marked decline in artifact density to
the south, and recent residential development with tightly
spaced houses to the west. In addition to shovel testing, eight
1 x 2-m units were excavated within midden concentrations
located on one of the lots.
The location of the various tests excavated at the Shields
site is shown in Figure 8. The topography undulates, but the
general trend is for a gentle southward downslope from the
high bluff shoreline that, after about 125 m, begins a fairly
steep ascent. Much of the area is open and grass covered,
although large oak and hickory tress are scattered about (see

- 1 lON

- IO00N


800b 900E


2005 VOL. 58(3-4)



Figure 3). Situated closer to the houses are smaller, planted
trees and shrubs. The southernmost portion of the sampled
area is covered by secondary forest vegetation. All roadways
are dirt and have been created via continuous use, although a
limerock layer has been laid down to increase traction.
Standard field procedures were employed throughout
fieldwork. Shovel tests were dug in arbitrary 20-cm levels,
whereas larger units were excavated in natural levels that did
not exceed 10 cm in thickness. Fill from all units was screen-
ed through 6.35-mm hardware cloth, with all cultural materi-
als retained according to provenience (e.g., unit level, area,
feature). The total volume of shell from each shovel test or
unit level was measured in liters in order to develop shell
density and distribution maps. Soil samples of varying size
were taken from selected features and shell midden contexts
and subjected to either fine mesh (1.19 mm) water screening
or flotation.
Sixty-four shovel tests were dug acrossthe site on a 25-m
grid to sample for the presence or absence of cultural materials
and to gain a better understanding of the spatial organization
of the site. Five additional shovel tests were dug in specific
areas to test distinct rises, suspected to represent shell deposits
or other anomalies. Across the sampled section of the site,
shell density per shovel test ranged from a trace to over 100
liters. Using a volume of 20-liters per shovel test as a criterion
for high shell density, three high concentration loci were
delineated. From north to south, these shell middens are
designated Bluff Midden, Reeve's Rise, and Kinzey's Knoll
(see Figure 4). All three are partially discernible today as low
rises in a grass covered yard. Outside of these three concentra-
tions, only two shovel tests yielded a shell volume of 10 liters
or more, and both were determined to represent very localized
deposits of shell refuse. As expected, vertebrate faunal
remains density was positively correlated with shell density; a
consequence of the acidity reducing quality of calcium
leaching from the accumulated masses of shell refuse. There
was also a positive correlation between ceramic frequency and
shell midden density, although high and moderate concentra-
tions of pottery occurred in areas of little or no shell.
Of the 69 shovel tests, 66 yielded artifacts. Ceramic
distribution data indicate an unbroken spread of St. Johns II
Period wares over the entire area; types grossly contemporane-
ous with the Shields Mound, given Moore's results. To date,
no pre-St. Johns II Period and only a few post-St. Johns II
Period artifacts have been identified among the more than
5000 sherds retrieved to date. Based on his earlier testing of
this same area, Johnson (1988) reported finding small amounts
of Deptford (n=45), Swift Creek (n= 1), and San Marcos (n=3)
types among the more than 600 sherds from sites 8DU12 and
8DU5603-5608; St. Johns (n=448) and cordmarked (n=96)
wares were the most dominant. The number of Deptford
Check Stamped sherds is somewhat perplexing, since we have
sampled the same loci as Johnson and have yet to find a single
Deptford sherd. I suspect the Deptford Check Stamped sherds
recovered by Johnson are actually sandy St. Johns, a paste type
we have recovered and recognized microscopically (Rolland
2004). The three mission-period San Marcos sherds recovered

by Johnson were from the far northwestern part of the site; an
area not sufficiently sampled during our investigation. The
Swift Creek sherd was recovered well back from the river and
away from our primary loci of testing.
Thus, the area surrounding the Shields Mound represents
an almost pure St. Johns II Period site. The few St. Marys, San
Marcos, and Swift Creek series sherds suggest short-term site
use such as resource procurement encampments. More
substantial Swift Creek refuse deposits and sand burial mounds
were previously documented a short distance south of the main
concentration of St. Johns II Period refuse situated between the
Shields and Grant mounds. In an attempt to gain a better
understanding of the site's St. Johns II component, the 3 shell
middens (Kinzey's Knoll, Reeves Rise, and Bluff Midden)
were subjected to additional testing.

Kinzey's Knoll

Kinzey's Knoll is a distinct dome-shaped rise of densely
packed shell midden that measured about 15 m in diameter.
However, shovel testing and subsurface probing revealed that
the distribution of thickly deposited shell midden was actually
much larger and continued downslope to the west and north,
eventually covering an area that measured 30 m (N-S) by 25
m (E-W). This locus is consistent with the location of site
8DU5606, which Johnson (1988:100-102) described as a 1.2-
m thick midden measuring only 30 by 25 m.
In the summer of 1999, two closely-spaced 1 x 2-m units
(Units 1, 2) were dug into the northern section of Kinzey's
Knoll, approximately 50 m northwest of the Shields Mound.
The following summer, we expanded Units 1 and 2 with the
excavation of two more 1 x 2-m units (Units 3 and 4).
Together the four form two staggered, yet contiguous, 2-m
squares. Units 5 and 6, also 1 x 2 m in size, were excavated in
the summer of 2001 and were placed south and west of
Kinzey's Knoll (Units 1-4), respectively.
The excavation of Units 1-4 on Kinzey's Knoll revealed an
approximately 80-cm thick shell midden that contained an
abundance of pottery and large quantities of vertebrate fauna.
The top of the shell deposit occurred immediately below the
grass and thin layer of humus. The shell midden consisted
primarily of an upper stratum of dark brown sand followed by
a slightly lighter lower stratum of brown sand (Figure 9). For
the most part, the shell was densely packed, although pockets
and areas of more loosely accumulated shell were documented.
Within the four investigative units, a high shell to soil ratio
was documented, and a total shell volume of 3864 liters was
recorded. Thus, shell, not dirt, formed the matrix within which
the overwhelming majority of artifacts were recovered.
Oyster was by far the dominant shellfish constituent within
the midden, with much lesser amounts of quahog clam,
Carolina marsh clam, Atlantic ribbed mussel, and stout
tagelus. Small amounts of whelk debitage also were recov-
ered. Occasional lenses of these minority species were noted,
but no distinct concentrations were recorded. A small,
intrusive pit containing a mason jar, nails, and other
twentieth-century debris was encountered within the shell




Figure 9. Unit 4 and 2, south wall profile, Kinzey's Knoll.

midden. A few other historic items were recovered from the
upper 20 cm of the units.
Several features were identified at the shell midden-subsoil
interface, including a small, circular pit surrounded by a series
of 8 possible postmolds, 2 thin, amorphous stains of heat-
altered sandy clay, and an approximately 10-cm diameter sain
of finely ground hematite and sand (Figure 10). The pit
contained small amounts of shell and bone and a few St. Johns
series sherds, and the dispersal of postholes appeared random
with no discernible pattern. Also identified were a small
pocket of shell refuse, a dip or low spot in the overlying shell
midden, and a dark area of leachate soil. The function,
distribution, and precise relationship of these features to the
overlying shell midden are uncertain due to restricted areal
extent of unit excavations.
The shell midden, replete with preserved vertebrate faunal
remains, contained an impressive variety of fish, deer, bear,
canid, bobcat, turtle, small mammal, and bird bones (Marrinan
2005, this issue). With regard to plant remains, while no

formal ethnobotanical analysis has been performed, cursory
examination of both large and fine mesh screen samples failed
to identify any preserved maize cobs, kernels, or cupules.
Pottery from Kinzey's Knoll consisted mostly of St. Johns
Plain andSt. Johns Check Stamped, but chalky punctated, red
filmed, and Little Manatee varieties also were found. Most
conspicuous among the nonspiculate wares were sherds of grit-
tempered Ocmulgee Cordmarked, several of which possessed
interior red filming, an unusual trait combination also reported
by Moore (1895) on two partial vessels from the Shields
Mound. A recent technofunctional study by Rolland (2004,
2005, this issue) of more than 2100 sherds from two of the
four excavation units indicated that portions of over 350
separate vessels were represented. At least 30 were grit-
tempered cordmarked pots, including vessels that displayed
both folded and simple (unfolded) rims. One red filmed vessel
possessed a unique calcareous clay paste suggestive of a north-
central Florida production origin.
In addition to pottery, a striking array of other items made

20 cm

3 very dark grayish brown (10 YR 3/2); little shell

I- very dark grayish brown (10 YR 3/2); dense shell

IC- brown (10 YR 5/3); dense shell

S- dark grayish brown (10 YR 4/2) sand ; no shell

LU mottled dark grayish brown (10 YR 4/2) and
yellowish brown (10 YR 5/4); no shell

[n- grayish brown (10 YR 5/2) sand; no shell


2005 VOL. 58(3-4)




25 cm

I evidence of burning in the form of ash,
charcoal, or heat altered sand

Figure 10. Features at shell midden-subsoil interface, Kinzey's Knoll.

of shell, bone, stone, and metal were retrieved. Items of
modified bone consisted ofundecorated and incised pins or pin
fragments, awls, shark centrum beads, dolphin tooth abraders,
and shark tooth gravers/scrapers and drills (Penders 2005, this
issue). There also were several informal tools crafted of
mammal bones as well as two antler tine fragments. Shell
artifacts were primarily beads, but a scraper, a pendant, and a
few columella punches or burnishers were part of the midden
assemblage (Penders 2005, this issue). What is most interest-
ing about the bone and shell artifacts is the high ratio of
decorative objects to tools, which led Penders to conclude that
these items were not routine domestic or living area garbage.
For example, a bone pendant (or dagger) exhibits an incised
design on both sides consisting of three eye motifs horizontally
stacked and each linked by four ladder motifs. James Brown
(personal communication, 2001) has commented that the

design is reminiscent of Muskogean cosmology that envisions
three worlds (Upper, This, Lower) connected by four cosmic
cords. This incised piece contained remnants of finely ground
hematite powder in the incisions.
Lithic material from the midden included sandstone
abraders and abrader fragments, a sandstone hone, two
greenstone celt fragments, several projectile points, a large
heat treated chert core, and a small amount ofdebitage (Bland
2001). The two pieces of greenstone appear to represent celt
fragments, one of which exhibited red pigment (iron oxide)
embedded in surface and edge pores. Of the 12 projectile
points recovered to date, 11 came from Kinzey's Knoll and
include seven small triangular Pinellas bifaces, a Citrus biface,
a Santa Fe biface, and an unusual side-notched biface (Bland
2001). The notched biface, the Santa Fe, and three of the
Pinellas points came from lower shell midden levels, whereas




the remainder of the stone points was recovered from contexts
immediately below the shell midden. However, all were
associated with levels yielding abundant St. Johns II Period
pottery. The Citrus and Santa Fe bifaces are earlier point types
probably scavenged from Archaic sites elsewhere. The same
may be true of the flakes-many of which were heat
treated-used to manufacture the Pinellas points (Bland 2001).
Moore (1895:460) recovered over 100 projectile points from
the Shields Mound and many of those illustrated are Archaic
types. The scavenging of earlier point types to use in ritual or
ceremonial contexts appears to have been a common practice
among St. Johns II peoples in northeastern Florida and at Mt.
Royal (see Ashley 2005a, this issue).
Three small scraps of copper and a small copper bead were
recovered and provide direct evidence of the acquisition of raw
materials derived from areas beyond northeastern Florida.
Copper was recovered from depths of 20 cm, 40-50 cm, and
60-70 cm below surface. The bead is a piece of rolled sheet
copper that may have been manufactured on site, and the
scraps may be pieces left over from working sheet copper (e.g.,
as covering for wood, stone, or bone artifacts). In addition,
several fragments of iron oxide or hematite (Fe203), a well-
known source of red pigment, were recovered. Hematite
powder or dust (ochre) also was observed in the cracks and
crevices of certain artifacts, such as sherds and a fragment of
Because St. Johns II pottery types dominated all excavation
levels from top to bottom, the Kinzey's Knoll shell midden can
be dated broadly to the period A.D. 750-1500. In an attempt
to narrow this span of time, oyster shell recovered from a
depth of 15 cm below surface and another from 67 cm below
surface were submitted for radiocarbon dating. The one-sigma
calibrated assays were A.D. 865-1035 and A.D. 990-1070,
respectively, and the two-sigma calibrated date ranges were
A.D. 765-1155 and A.D. 910-1170, respectively. These
radiocarbon dates, combined with a lack of sherd cross-
mending between excavation levels, suggest the midden was
laid down in distinct deposits over a relatively short period of
time between ca. A.D. 900-1150. The two dates bracket a
radiocarbon date from a St. Johns Period II feature beneath the
base of the Grant Mound (Thunen 2005, this issue), suggesting
contemporaneity between these areas of the Shields and Grant
sites during the period A.D. 900 to 1150.
Abutting the Kinzey's Knoll shell midden to the west and
south was a low-density spread of midden refuse. To sample
this peripheral area of shell midden, two 1 x 2-m units were
excavated, with one (Unit 5) placed to the west and the other
(Unit 6) to the south. Unit 5 intersected a locus of disturbance
associated with the construction and subsequent filling of a
mid-twentieth century well. Chert gravel and other historic
debris were abundant within the top 40 cm of soil. However,
undisturbed contexts below 40 cm indicated an organic
midden with scattered oyster shell, little vertebrate bone, and
St. Johns II Period pottery. A thin shell deposit was partially
exposed within the western section of the unit. Probing
indicated that this St. Johns Period II feature was oval and
measured 1.68 by 1.22 m. The shell deposit was 11 cm thick

along the unit's western wall, and thinned to a tapered point
that formed its eastern edge. This suggests that it represents
a small shell midden deposited on an old ground surface. The
only unusual artifact from Unit 5 was a portion of a greenstone
celt recovered from a mottled leachate zone beneath the
organic midden in the eastern part of the unit.
Unit 6 was positioned about 30 m south and slightly
downslope of the apex of Kinzey's Knoll at the base of a
southward rising slope. This locus corresponded to Johnson's
(1988:102-104) site 8DU5607, which he described as a
circular, shell midden rise that measured only 20 by 15 m.
Unit 6 revealed a thick, dark grayish-brown earth midden with
occasional whole and fragmented mollusk shells and large
amounts of St. Johns II Period pottery and vertebrate animal
bone. A large pit feature, more than 1 m in diameter, was
partially exposed at the midden-submidden interface. The
feature extended into the western and northern walls of Unit
6, such that only a small portion was within the unit and
excavated. The sloped walls and round bottom of the 1.1-m
deep pit were lined with a layer of well-packed shell midden
along with flecks and small chunks of charcoal.
An oyster shell from the base of the feature yielded a one-
sigma calibrated radiocarbon assay of A.D. 1240-1310 and a
two-sigma date range of A.D. 1180-1390. This represents the
latest (most recent) radiometric date associated with a St.
Johns II Period context in northeastern Florida, and it suggests
that feature post-dates the activities at Kinzey's Knoll.
Cultural materials recovered from the pit feature were similar
to those recovered from the overlying midden, which consisted
mostly of St. Johns II Period pottery and vertebrate bone.
Because the pit was only partially excavated, its precise
function is uncertain, but it may have been constructed for
food processing or cooking and later used as a receptacle for

Reeves Rise

This shell midden was located about 25 m north of and
upslope from the northern edge of the Kinzey's Knoll shell
midden. Its size (ca. 20 m in diameter) was delineated via
probing, but it stood out visually as a high spot directly in front
of the southern entrance to a modern house (Reeves family);
the northern quarter of the midden extended beneath the
elevated house. Testing here was at first restricted to a single
0.5-m square on the midden peak, since the landowner was
somewhat apprehensive about having a large unit opened in
this part of his yard. Following the success of the first shovel
test, however, he allowed us to excavate two more shovel tests
in close proximity to the first. Each was dug to a terminal
depth of 1 m. These shovel tests produced shell volumes of 72
liters, 70 liters, and 29 liters. The former tests were positioned
roughly along the same north-south axis but 3 m apart,
whereas the later test was situated 3 m south of the western
test, per the landowner's direction. Areas to the north and
west were apparently impacted by house construction and the
building of a cement pond filled in the 1960s.
The two northern shovel tests contained small amounts of



2005 VOL. 58(3-4)


glass, metal, construction rubble, scattered and broken oyster
shells, and a few potsherds in the top 15 cm of soil. Beneath
this disturbed layer was a consolidated oyster shell stratum that
extended from 15 to 48 cm below surface. Maximum shell
volume was documented at a depth between 30 and 40 cm
below surface. A radiocarbon assay on shell taken from near
the base of the midden (45 cm below surface) produced a one-
sigma calibrated date range of A.D. 1160-1270 and a two-
sigma calibrated date range of A.D. 1060-1300. This date
range suggests that the shell midden at Reeves Rise was
formed shortly after the one deposited at Kinzey's Knoll,
although the possibility exists that two are contemporaneous
to some extent. The southern shovel test revealed a loosely
packed shell midden that contained material down to a depth
of 40 cm below surface.
Cultural material from the three shovel tests consisted
exclusively of St. Johns ware fragments and small amounts of
grit-tempered cordmarked pottery. The most unique find was
an effigy adorno from a St. Johns Plain pot. Unfortunately,
only a small section of the vessel was recovered. The adorno
is a small animal head, with a thin snout, a slit for a mouth,
two small punctations for eyes, and distinctive pointed ears
resembling a'dog, fox, or some other canid (Rolland 2005, this
issue). Only a few pieces of modified bone and no lithic
artifacts were recovered. Formal bone tools and decorative
pins, so conspicuous in the Kinzey's Knoll shell midden, were
absent. Copious quantities of vertebrate faunal remains were
removed from the shell midden, including fish, bird, turtle,
deer, and small mammal (Marrinan 2005, this issue). Oyster
was the dominant shellfish constituent with trace amounts of
Carolina marsh clam, stout tagelus, Atlantic ribbed mussel,
and whelk shell fragments.


The Bluff Midden was distributed along a section of the
undulating bluff that overlooks the St. Johns River to the
north. It was bounded on the basis of five shovel tests, with
shell volumes that ranged from 10 to 32 liters (mean volume
of 17 liters). The BluffMidden, which measured 90 by 25 m,
was less dense and more variable in shell volume than the
other two shell middens. This locus was previously sampled
by Johnson (1988) and designated 8DU5605. He described it
as an "artificial rise and associated shell
midden...approximately 30 meters in diameter" (Johnson
1988:97-98). Recent testing revealed that the rise is not
artificial but part of the rolling bluffline and that the shell
midden extended downslope to the west. In fact, shovel testing
indicated that the shell midden was thicker and more consoli-
dated along the slope than on the apex of the bluff.
The midden is dispersed along a bluffline that rises from
west to east, with a 2.8 m differential between highest and
lowest surface elevations. A 2-m square, consisting of two
contiguous 1 by 2-m units (Units 7 and 8), was excavated in
the lower, western part of the shell midden. The unit was
placed on a slope in which the ground surface along its
western edge was approximately 20 cm lower than along its

eastern edge. The unit was dug to a depth of 1-m below its
southwestern or lowest surface corner.
The various strata revealed in the northern wall profile
followed the sloped contour of the ground surface (Figure 11).
Beneath the grass and humus zone was an approximately 27-
cm thick layer of very dark grayish brown sand with a loose
amalgam of molluscan shells, which overlay an approximately
36-cm thick layer of moderately packed shell within a very
dark grayish brown sand matrix. Although the latter demon-
strated a higher shell to dirt ratio than the surmounting
stratum, it was not as consolidated as the sampled portions of
the Reeves Rise and Kinzey's Knoll shell middens. Below the
lower midden was a leachate zone that was followed by yellow
brown sand, with rare shell. There was much less pottery,
bone, and shell within these lower zones.
Like Kinzey's Knoll and Reeves Rise, all contexts sampled
within the Bluff Midden yielded mostly, if not exclusively, St.
Johns II Period ceramics (Rolland 2005, this issue). In
addition to large quantities of chalky pottery and grit-tempered
cordmarked sherds, several lithic flakes and pieces of iron
oxide were recovered along with a few bone tool fragments
and other pieces of modified animal bone and whelk shell.
The most unique items were a minute piece of sheet copper
from the shell midden and a small, rolled sheet copper bead
from the leachate zone beneath the shell midden. Vertebrate
animal bone refuse was recovered in large quantities and
included mostly fish with occasional deer, turtle, and bird
(Marrinan 2005, this issue). Other than the two pieces of
copper, the remains from the BluffMidden appear to represent
typical domestic refuse. Once again, oyster was the primary
shellfish species with marsh clam, stout tagelus, Atlantic
ribbed mussel, and whelk scattered in small amounts through-
out the midden.
An oyster shell from near the base of the midden produced
a one-sigma calibrated radiocarbon date of A.D. 1070-1230
and a two-sigma date of A.D. 1030-1280. At the one-sigma
date range, this places the Bluff Midden between the earlier
Kinzey's Knoll and the later Reeve's Rise radiocarbon assays,
suggesting the three loci are sequential (although the Bluff
Midden and Reeve's Rise assays overlap during the years A.D.
1160 to 1230). However, there is temporal overlap among all
three loci at the two-sigma date interval. Presently it is not
known how much erosion has taken place along this section of
the bluff, so it is difficult to discern how far the midden may
have once extended to the north.


Because we have yet to uncover any direct structural
evidence, such as arrangements of postholes, pits, or hearths,
the precise nature and distribution of the St. Johns II Period
domestic structures and other buildings at the Shields site
cannot be ascertained. However, shovel testing and unit
results contribute important data on the dispersal of refuse
across the site. This spatial information in concert with the
kinds and range of materials recovered from specific site
contexts can be used to offer preliminary suggestions on site




20 cm
20 cm

E very dark gray (10 YR 4/3); little shell

B very dark grayish brown (10 YR 3/2); loose shell

[ very dark grayish brown (10 YR 2.5/2): moderately dense shell

[] light brownish gray (10 YR 6/2). w/light gray (10 YR 7/1) mottling, rare shell

] yellowish brown (10 YR 5/6). w/brownish yellow (10 YR 6/6) mottling, no shell

Figure 11. Unit 8, north wall profile, Bluff Midden.

activities and community layout.
Shovel test results indicate that there was a continuous
spread of St. Johns II Period ceramics across the sampled
section of the Shields site. The ubiquitous distribution of
pottery contrasted with the dispersal of shell refuse. Shell
midden concentrations were restricted to three discrete loci,
although two other small shell features were encountered
during shovel testing. Oyster was by far the dominant shell-
fish species within all sampled middens. As expected the
density of vertebrate faunal materials was positively correlated
with density of shell.
Of the 69 shovel tests dug on site, 57 (83%) yielded 8 liters
or less of shell, and 46 (67%) produced 1 liter or less of shell.
Though a trace scattering of shell is present throughout the
site, a distinct mantle of shell midden is not pervasive nor are
individual house middens peppered over the site, as is the case
I t

with later St. Marys II Period sites in the region. Compared to
the amount of shellfish refuse found on sites to the east, closer
to the coast or on the barrier islands, much more shell was
expected at the Shields site than was encountered, particularly
considering the fact that the site may have been occupied for
several centuries.
Factors contributing to the patterning of shell refuse at the
Shields site include the possibility that oysters were not
consumed as much by site inhabitants as they were by groups
inhabiting sites near the coast. Alternatively, perhaps a good
deal of oyster processing occurred off site, possibly near the
locus of exploitation. An interesting fact regarding the
location of both the Grant and Shields sites is that expansive
salt marshes do not front the river shoreline as is the case
farther to the east. Both the Mariano de la Rocque map of
1791 and the Charles Vignoles watercolor map of 1822


2005 VOL. 58(3-4)


provide detailed sketches of the lower St. Johns River, and
each shows that the distribution of marsh habitat terminated
within the center of Mill Cove, east of the Shields sites (see
maps in Ward 1985:87, 126-127). It is very likely that the
dispersal of salt marsh biomes coincided with that of viable
oyster beds, meaning that oyster populations may not have
been immediately available to the occupants of the Shields and
Grant sites. Along the lower St. Johns River west of the Grant
Mound, known archaeological sites are characterized by either
very low-density scatters of shell or no shell at all.
From the Shields Mound, site occupants may have had to
travel at least 2 km to the eastern fringes of Mill Cove before
they would have encountered expanses of marsh, according to
the earlier maps. However, reports from the early twentieth
century indicate that Mill Cove was a deep-water cove that was
less brackish than it is today (Anonymous 1986). According
to local fishermen whose families have been fishing these
waters for the past century, certain freshwater fishes were once
more common in Mill Cove than elsewhere along the lower St.
Johns River (Walter Wells, personal communication, 2002).
In fact, prior to major rechannelization efforts in the 1930s,
freshwater eelgrass grew along the banks of Mill Cove and
oyster beds were absent. The salinity level of the cove may
have been lower in pre-Columbian times due to a higher rate
of freshwater intrusion caused by north-flowing freshwater
within the original channel moving into Mill Cove. At the
least, salinity levels could have been variable due to hydaulic
or climatic conditions. In either case, it may have had a
pronounced effect on oyster populations in the cove, meaning
the occupants of the Shields site, at times, may have had to
travel even farther downstream to gather oysters.
Village cleaning is another factor we must keep in mind
when interpreting the spatial distribution of shell midden
refuse at the Shields site. If people were living there continu-
ously, instead of periodically, there may have been a need to
dispose of large amounts of non-biodegradable refuse (e.g.,
oyster shell) in a way that would not have impinged upon site
activities or living and work spaces. Thus, people may not
have been able to throw or sweep trash anywhere they wanted,
and there may have been an emphasis on keeping large
sections of the site clean or free of bulky refuse, meaning that
secondary communal dumps would have been localized and
situated away from houses. There may have been less concern
for where sherds and other smaller pieces of refuse were
discarded, particularly if these items could have been recycled.
This may have been a common practice at local St. Johns II
villages, but such a distribution becomes obscured on
multicomponent sites with a high degree of subsequent
reoccupation. A more detailed interpretation of the three
middens and the Shields site in general is provided in a later
paper (Ashley 2005b, this issue).

Mill Cove Complex

Grant Mound

The Grant Mound (8DU14) is located 750 m northwest of

the Shields Mound and situated on a high bluff overlooking
the river to the north (see Figure 1). Moore (1894:200)
described it as a "truncated cone" about 8.1-m high, with a
base diameter of approximately 65.8 m and a summit plateau
diameter of 7.3 m (Figure 12). Moore (1895:473) spent parts
of two summers digging the Grant Mound. He encountered
numerous human burials, although the exact number is
difficult to decipher from Moore's writings. Although some
mound artifacts were directly associated with human remains,
the overwhelming majority apparently were not (see Thunen
2005, this issue).
Artifacts of both local and nonlocal origin were recovered
and included polished celts, chipped stone tools, soapstone
elbow pipes, shell drinking cups, pearls, shell beads, bone pins
and tools, and pottery vessels. The recovery of 147 ground
stone celts from the mound by Moore is astonishing. Also
taken from Grant Mound were various mineral and metals
such as galena, mica, and copper, the latter of which included
beads, sheets with repouss6 bosses and beaded lines, two Long-
nosed god maskettes (actually earpieces), and a variety of
wood and stone items overlaid with sheet copper. The
maskettes and two copper covered "cones of wood" that
together form a biconical ear ornament suggest possible
connections to Cahokia and Spiro, or at least to the Midwest
region in general (Goggin 1952; Kelly and Cole 1931;
Williams and Goggin 1956). A direct connection between
Grant and Spiro, as well as between Grant and Gahagan
Mound in Louisiana (Webb and Dodd 1939), seems unlikely,
and it is more probable that the three share similar exotic
artifacts that originated from a common location, perhaps
Cahokia. These exotic items along with other artifactual
evidence place Grant Mound in the Early Mississippian Period
or local St. Johns II Period, ca. A.D. 900-1250 (Ashley
2003:139-144; Thunen 2005, this issue).

Grant Mound Vicinity

In 1988 Johnson recorded 13 new sites between the Grant
and Shields mounds (Figure 13). The discontinuous distribu-
tion of these sites reflects both limited shovel testing and
incomplete survey coverage, since fieldworkers were at times
denied access to some properties. In addition to refuse deposit
observed along the western edge of Grant Mound (8DU14),
additional shell middens were noted to the west (8DU5597),
southwest (8DU5598), and southeast (8DU5599) (Johnson
1988:70-75; 78-85). The area immediately south and south-
east of the mound was not inspected due to cul-de-sac con-
struction and landowner refusal, respectively. However, other
landowners living south of the mound report little oyster shell
but abundant check stamped pottery. Although recorded
separately, the vast majority of artifacts from the three sites are
St. Johns II Period wares, along with trace amounts of
Deptford and Swift Creek sherds, which renders the middens
grossly contemporaneous with the Grant Mound. These
deposits indicate that a continuous spread of St. Johns II
material radiated out from the mound for a distance of
approximately 100 m to the west, south, and east. Though




Figure 12. Grant Mound (Moore 1894a).

shell midden density is variable across the rolling relict dune
landscape, shell and refuse accumulations were thickest (in
excess of Im in some loci), immediately west and southwest of
the mound.
In January of 1989, a large segment of the Grant Mound
was removed by earthmoving equipment (Thunen 2005, this
issue). Robert Thunen and students from the University of
North Florida (UNF) were permitted on the property during
bulldozing, but were not allowed to stop the removal of mound
fill. Following the mechanized removal of mound fill, UNF
crews were able to excavate four 2-m square units (Thunen
2005, this issue). Two were placed in areas where the mound
once stood in anticipation of uncovering either intact basal
remnants of the mound or submound features. The other two
units were placed outside mound boundaries to sample midden
deposits and search for domestic features. These excavations
revealed substantial cultural deposits both beneath the mound
and along the bluffline to the west. An oyster sample from
Feature 1, which contained several Papys Bayou sherds, was
radiocarbon dated (one-sigma calibrated) to A.D. 905-1025
(Thunen 2005, this issue). Since the radiocarbon-dated feature
was situated at the base of the midden, we can estimate when

the midden began to accumulate, but we presently do not know
how long it took to build up. Based on the artifacts recovered
from the mound by Moore (1894a; 1895) we can infer that the
mound and midden date grossly to the same era, and that it is
likely that some midden and mound related activities were
precisely coeval.
The most recent archaeological work conducted in the
Grant Mound vicinity has occurred on two contiguous
riverfront tracts, within an area of the site designated
8DU5599 by Johnson (1988). The first, undertaken in the
summer of 1999 by members of the Northeastern Florida
Anthropological Society (NEFAS 2001), involved the testing
of private property, located about 60 m southeast of the
mound. This tract is situated immediately southeast of and
slightly above the "abrupt dip of the land" mentioned by
Moore (1895:493), but still at an elevation 3 to 5 m lower than
the bluff on which the burial mound is perched.
Thirty-one 1-m square units were dug on a tight 10-meter
grid across a residential yard along the river. In addition, a
single 2-m square was dug to expose a feature encountered
during grid testing. While shell was scattered over the entire
area, only a few distinct shell deposits were encountered. A 1

2005 VOL. 58(3-4)




a t Mound

St. Johns River


, 8DU5600


Figure 13. Location of archaeological sites in the vicinity of the Grant and Shields Mounds.

by 2-m unit was excavated into one of these middens, which
displayed a densely-packed shell zone, 15 to 30-cm thick. The
unit yielded over 200 sherds, most of which are characteristic
of a St. Johns II Period pottery assemblage. An oyster shell
from the midden was radiocarbon dated to A.D. 1170-1285
(NEFAS 2001). Preliminary results, combined with my own

inspection of the recovered artifacts, indicate a strong St.
Johns II Period presence marked by St. Johns II and grit-
tempered cordmarked wares and Pinellas points; small
amounts of earlier Woodland Period and later St. Marys II
Period pottery types were also recovered.
Finally, a cultural resource assessment survey was made of





an approximately 1.6-ha tract, 120 m southeast of the Grant
Mound and contiguous with tract surveyed by NEFAS
(Hendryx and Smith 2002). Approximately 75 percent of the
630 sherds from the site unequivocally date to the local St.
Johns II Period; inclusion of some of the nondescript
plainwares could conceivably bring this figure up to about
eighty percent. Little in the way of Woodland Period pottery
was recovered, but later St. Marys II Period (A.D. 1250-1500)
and San Pedro (post-A.D. 1500) ceramics were more common.
Although occupational debris was dispersed over the entire
area, discrete shell deposits or middens were few, limited in
horizontal extent, and restricted to the area near the bluff.
This tract appears to have sustained a fair amount of distur-
bance, although intact St. Johns II Period deposits may exist.

Defining the Mill Cove Complex

With the data on hand, how might the spatial relationship
between the Grant and Shields mounds be interpreted?
Traditionally, the two have been discussed as separate St.
Johns II Period constructions (Goggin 1952:54-55), although,
in recent years, we have alluded to the possibility that the two
may somehow be related (Ashley 2002; Thunen and Ashley
1995:6). With a battery of radiocarbon dates and the results of
recent archaeological testing now available, it is proposed that
the two large mounds are indeed contemporaneous and
spatially connected. Positioned about 750 m apart, the two are
linked by an almost continuous spread of St. Johns II Period
refuse; however, midden materials are generally not present
along steeply sloped and low landforms not conducive to
human settlement (Hendryx and Smith 2002; Johnson 1988;
NEFAS 2001). Current data indicate that shell midden
deposits and discarded refuse are most heavily accumulated in
close proximity to each mound, with variable density midden
interspersed between the two. More data are needed from loci
midway between the two mounds, although these areas have
been heavily impacted by recent residential development.
The two sites appear coeval or at least so closely separated
in time as to be statistically indistinguishable during the period
A.D. 900-1250. For instance, one of the radiocarbon assays
from a shell midden near the Grant Mound is statistically
equivalent to an assay from a refuse deposit near the Shields
Mound. In another case, one of the Grant Mound dates is
tightly bracketed by two dates from Kinzey's Knoll. With
respect to the mounds themselves, both the Long-nosed god
maskettes (earpieces) from Grant Mound and the long-
stemmed spatulate celts from Shields Mound are considered
Early Mississippian and date to circa A.D. 1000-1200.
Admittedly, we still do not know the precise physical layout or
extent of the village community at any one time during the
three and a half century span of the local St. Johns II Period.
However, the site was clearly not a vacant mound center.
The term Mill Cove Complex is proposed to refer collec-
tively to the Grant and Shields mounds and the widespread
dispersal of St. Johns II Period middens scattered between
them (Figure 14). The names Shields Mound and Grant
Mound will continue to be employed to denote each respective

earthwork. Additionally, the word "mound" is dropped and
replaced with "site" when referring specifically to nonmound
areas adjacent to each mound (e.g., Shields site, Grant site).
Although a detailed discussion of the Mill Cove Complex is
presented in Ashley (2005b, this issue), a few general state-
ments on the mounds and middens are offered here.
First, a reminder for those unfamiliar with the Mill Cove
Complex vicinity, its topography is atypical of northeastern
Florida. The landscape is very undulating, with marked
changes in relief. Topographic features consist of rolling hills,
ridges, and ravine-like declivities, all representative of ancient
marine terraces, ridges, and dune fields. The builders of both
the Grant and Shields mounds apparently utilized the natural
terrain to their advantage and grafted their earthen monuments
onto the most prominent natural formations in the immediate
Second, an interesting distributional aspect of the area is
that St. Johns II Period artifacts are overwhelmingly concen-
trated along the bluff, whereas Woodland-period Swift Creek
materials are much more prevalent in middens or artifact
scatters away from the river on the sand ridges. Refuse
dispersal data suggest that St. Johns II Period peoples were
distributed in a linear fashion along the river, staying within
a 200 m or so of the bluff. That the shoreline has undergone
considerable erosion in the past century or two indicates
portions of the site's St. Johns II component also has eroded
into the river.
Third, the Shields site is virtually a pure St. Johns II Period
site, with occasional temporally unrelated ceramics recovered
within 20 cm of ground surface. In contrast, the Grant site has
demonstrated more mixing in the form of Woodland-Period
Deptford and Swift Creek sherds; however, only small
amounts have been recovered to date. The recovery of St.
Marys, San Pedro, and San Marcos series pottery also provides
evidence for post-St. Johns II Period occupations at the Grant
site. Again these wares pale in comparison to the number of
pottery types associated with St. Johns II Period occupation of
the Grant Mound vicinity.


As recently as a decade ago, the consensus among Florida
archaeologists was that the St. Johns II Period covered the
same general timeframe (A.D. 750-1500+) in northeastern
Florida as it did to the south in the St. Johns heartland. Recent
research, however, has dispelled this notion and demonstrated
a more compressed St. Johns II Period in northeastern
Florida-ca. A.D. 900-1250 (Ashley 2002:163-164, 2003:96-
104). Contrary to earlier beliefs, the Grant and Shields
mounds were not sequentially constructed and used earth-
works, but contemporary pieces of mortuary architecture dated
to circa A.D. 900-1250. Abundant amounts of St. Johns II
Period refuse deposited between the two large mounds indicate
that people also were living there during the same period, and
the site was not a vacant center or necropolis. Some middens
contain mere domestic trash, whereas others closer to the
mound suggest the byproduct of ritual and feasting. But more

2005 VoL. 58(3-4)



Figure 14. Location of the Mill Cove Complex.

archaeology remains to be conducted within the proposed
boundaries of the Mill Cove Complex. To date, no structures
have been identified nor has a specific complex layout been
determined for specific periods within the broad A.D. 900 to
1250 timeframe. Future excavations are planned with the
hope of filling these gaps within the archaeological record.


In this paper the term "northeastern Florida" is used in a specific
manner to refer to present-day Nassau, Duval, and northern St. Johns





The research associated with the Shields Mound project has
benefited from many people over the past five years. I owe a great
deal of thanks to several landowners who allowed me to shovel test
on their property and to examine their artifact collections; they
include the Reeves, Wells, Simmons, Pearsall, Sares, Masucci, and
Banks families. I especially want to thank Kinzey Reeves for making
his yard available for archaeological excavation. I appreciate the help
of volunteers from the local community, University of North Florida
students, and members of the Northeast Florida Anthropological
Society, who served diligently as dependable and efficient field
workers. I would particularly like to thank Mike Tarlton, Dave
Bishop, Walter Wells, Jim Freel, Brian Floyd, and Bob Richter. I
profited greatly from a symposium on the Shields site, and I thank all
the participants: Rochelle Marrinan, Vicki Rolland, Tom Penders,
Myles Bland, Duke Beasley, Buzz Thunen, and Jim Brown. My
thanks are also extended to Ken Sassaman, Buzz Thunen, Jerry
Milanich, Ryan Wheeler, Vicki Rolland, and an anonymous reviewer
for commenting on earlier versions of this paper. Funding for two
radiocarbon dates was provided by two Florida Archaeological
Council John W. Griffin Student Grants. Finally, thanks to Angela
Ashley and Jo Ann and David Mynatt for their support and help with
many facets of the project, including all graphics.

References Cited

1986 Rep. Bennett is a winner in long Mill Cove battle. Jack-
sonville Journal, October 24, 1986.

Ashley, Keith H.
2002 On the Periphery of the Early Mississippian World:
Looking Within and Beyond Northeastern Florida. South-
eastern Archaeology 21:162-177.

2003 Interaction, Population Movement, and Political Economy:
The Changing Social Landscape of Northeastern Florida
(A.D. 900-1500). Ph.D. dissertation, Department of
Anthropology, University of Florida, Gainesville.

2005a An Archaeological Overview of Mt. Royal. The Florida
Anthropologist (This issue).

2005b Toward an Interpretation of the Shields Mound (8DU12)
and Mill Cove Complex. TheFlorida Anthropologist (This

Bland, Myles C. P.
2001 Moore to the Point. Paper presented at the 581Annual
Southeastern Archaeological Conference, Chattanooga,

Brooks, H. Kelly
1981 Physiographic Divisions: State of Florida. Institute of
Food and Agricultural Sciences, University of Florida,

Brose, David S.
1979 A Speculative Model of the Role of Exchange in the
Eastern Woodlands. In Chillicothe Conference, edited by
D. E. Brose and N. Greber, pp. 3-8. The Kent University
Press, Kent, Ohio.

Brown, James A.
1976 The Southern Cult Reconsidered. Midcontinental Journal
ofArchaeology 1:115-135.

Brown, James A., Richard A. Kerber, and Howard D. Winters
1990 Trade and the Evolution of Exchange Relations at the
Beginning of the Mississippian Period. In The Mississip-
pian Emergence, edited by Bruce Smith, pp. 251-274.
Smithsonian Institution Press, Washington, D.C.

Goggin, John M.
1952 Space and Time Perspective in Northern St. Johns Archeol-
ogy, Florida. Yale University Publications in Anthropol-
ogy 47, New Haven.

Handley, Brent M., Greg S. Hendryx, Michael Arbuthnot, and Greg
C. Smith
2002 An Intensive Cultural Resource Assessment Survey of Las
Calinas East Tract, and Test Excavations at 8SJ3471,
8SJ3476, and 8SJ4773, St. Johns County, Florida. Report
on file, Division of Historical Resources, Tallahassee.

Hendryx, Gregory S., and Greg C. Smith
2002 An Intensive Cultural Resource Assessment Survey of
Windswept Point and Site Testing at 8DU5599, Duval
County, Florida. Report on file, Division of Historical
Resources, Tallahassee.

Kelly, Arthur, and Fay-Cooper Cole
1931 Rediscovering Illinois. In Blue Book ofthe State ofIllinois
1931-1932, pp. 318-341. Springfield.

Larson, Lewis H., Jr.
1971 Archaeological Implications of Social Stratification at the
Etowah Site, Georgia. InApproaches to the SocialDimen-
sions ofMortuary Practices, edited by James A. Brown,
pp. 58-67. Memoirs of the Society for American Anthro-
pology Number 25.

Marrinan, Rochelle
2005 Early Mississippian Faunal Remains from the Shields
Mound (8DU12). The Florida Anthropologist (This

Milanich, Jerald T.
1994 Archaeology of Precolumbian Florida. University of
Florida Press, Gainesville.

Milanich, Jerald T., and Charles H. Fairbanks
1980 Florida Archaeology. Academic Press, New York

Moore, Clarence B.
1894 Certain Sand Mounds of the St. Johns River, Florida, Parts
I and II. Journal of the Academy of Natural Sciences of
Philadelphia, Second Series 10:129-246.

1895 Certain Sand Mounds ofDuval County, Florida. Journal of
the Academy ofNatural Sciences ofPhiladelphia, Second
Series 10:449-502.

Morgan, William N.
1999 Precolumbian Architecture in Eastern North America.
Revised Edition. University Press of Florida, Gainesville.

2005 VOL. 58(3-4)


Northeast Florida Anthropological Society (NEFAS)
2001 Preliminary Results of Testing of the Wells Property
(8DU14), Duval County, Florida. Report on file, Archaeol-
ogy Laboratory, University of North Florida, Jacksonville.

Pauketat, Timothy R.
1983 A Long-Stemmed Spud From the American Bottom.
Midcontinental Journal ofArchaeology 8:1-15.

Peebles, Christopher
1971 Moundville and Surrounding Sites: Some Structural
Considerations of Mortuary Practices II. In Approaches to
the Social Dimensions ofMortuary Practices, edited by J.
A. Brown, pp. 68-91. Memoirs of the Society for American
Anthropology No. 25.

Penders, Thomas E.
2005 Bone, Antler, Dentary, and Shell Artifacts from the Shields
Site (8DU12). The Florida Anthropologist (This issue).

Rolland, Vicki
2004 Measuring Tradition and Variation: A St. Johns Johns II
Pottery Assemblage from the Shields Site (8DU12).
Unpublished Master's Thesis, Department of Anthropol-
ogy, Florida State University, Tallahassee.

2005 An Investigation of St. Johns and Ocmulgee Series Pottery
from the Shields site (8DU12). The Florida Anthropolo-
gist (This issue).

Thunen, Robert L.
2005 Grant Mound: Past and Present. The Florida Anthropolo-
gist (This issue).

Thunen, Robert L., and Keith H. Ashley
1995 Mortuary Behavior along the Lower St. Johns: An Over-
view. The Florida Anthropologist 48(1):3-12.

Ward, James R.
1985 OldHickory's Town. Old Hickory's Town, Inc., Jackson-

Webb, Clarence H., and Monroe Dodd, Jr.
1939 Further Excavations of the Gahagan Mound: Connections
with aFlorida Culture. Bulletin ofthe TexasArcheological
and Paleontological Society 11:90-128.

Williams, Stephen, and John M. Goggin
1956 The Long Nosed God Mask in Eastern United States. The
Missouri Archaeologist 18:1-72.




An Endowment to

Support production of
The Florida Anthropologist,
the scholarly journal published quarterly by
The Florida Anthropological Society since 1948.

Donations are now being accepted from
individuals, corporations, and foundations.

Inquiries and gifts may be directed to:

The Editor
The Florida Anthropologist
P.O. Box 6356
Tallahassee, FL 32314

The Florida Anthropological Society is a non-profit organization
under section 501(c)(3) of the Internal Revenue Code.
Contributions are tax-deductible as provided by
section 170 of the code.



Department ofAnthropology, Florida State University, Tallahassee, FL 32306-7772
Email: rmarrina@garet.acns.fsu.edu

Food and feasting have piqued the curiosity of anthropolo-
gists through the years. Food is central to sustained cultural
viability and functions through gifting, barter, redistribution,
and other means to meet an assortment of cultural needs that
transcend mere species survival. Ethnographies and theoreti-
cal studies have detailed the modes of acquisition or produc-
tion, recipes, customs, and uses of food (e.g., Boas 1966;
Powdermaker 1932; Douglas 1975; Goody 1982; Harris 1985;
Mintz 1997). In archaeological contexts, the study of plant
and animal food remains has grown since the 1960s. Research
designs that include subsistence evidence, and the develop-
ment of specialists who identify and interpret these data, have
generated a substantial database with which to address
behavioral issues of interest to anthropological archaeologists.
The ritual or ceremonial use of animals has received increased
attention in recent years, and in the past decade, a number of
anthropologists and archaeologists have addressed feasting -
its social implications, requirements, staging, and the kinds of
evidence that might be preserved (e.g., Dietler 1996; Dietler
and Hayden 2001; Hayden 2001). Several have addressed
Mississippian culture specifically (e.g., Jackson and Scott
1995, 2003; Kelly 2001; Knight 2001; Pauketat et al. 2002;
VanDerwarker 1999).
In this study, faunal remains recovered from midden
deposits adjacent to Shields Mound (8DU12), an Early
Mississippian Period site in Florida, are discussed (Figure 1).
Shields Mound is located on the St. Johns River east of
Jacksonville, approximately 14 miles from the river mouth.
Grant Mound (8DU14), better-known but mostly destroyed,
lies 750 m to the west along the river edge. Radiocarbon dates
(Ashley 2003:Table 4.1) suggest that the two mounds were
contemporaneous (ca. A.D. 900-1250). Whether shell midden
deposits in the intervening area represent continuous occupa-
tion of the site over a 350-year period or intermittent occupa-
tion for ritual events, cannot be determined without additional
excavation and dating. Ashley (2003) refers to this mound
group as the Mill Cove Complex. Chronology and mound
form indicate contact with Mississippian societies or ideology,
but some of the typical cultural markers of Mississippian
societies elsewhere are not in evidence (e.g., maize agriculture
and a clear demonstration of sedentism). Dating of this site,
evidence of long distance trade, and physical distance from the
heartland, suggest that the Shields Mound data will provide
insights from a less typical expression of Mississippian
Vertebrate and invertebrate faunal remains recovered from

eight excavation units and three shovel tests in midden areas
closely adjacent to Shields Mound have been analyzed in this
study. Ashley (2003) identified several discrete concentrations
of high density shell midden during a preliminary program of
shovel testing (Figure 2) and opened eight 1 x 2-m excavation
units in two of these areas for further assessment: Kinzey's
Knoll (six units) and the Bluff Midden (two units). Because
of residential structures, a third midden area, Reeves Rise,
could only be evaluated through shovel testing. The analysis
of three tests from that area is also included.
The most important question to be answered is whether
these assemblages possess attributes that indicate ritual
activities at this Early Mississippian Period site. Given the
proximity of these deposits to Shields Mound, is there evidence
of feasting or use of animals that can be shown to differ from
a domestic pattern? To answer these questions, I shall first set
Shields Mound within its archaeological context. Next, I shall
characterize the assemblages and evaluate their status within
the local environment relative to species availability using
diversity and equitability measures. To anchor a discussion of
ritual activity, I shall consider ethnohistoric evidence of
animal usage in Southeastern Indian societies and then briefly
examine coastal zone subsistence patterns through time using
data from a group of comparable sites. Lastly, criteria devel-
oped by Scott and Jackson (1995, 2003) will be used to
evaluate the Shields Mound assemblages for evidence of
feasting activities, elite consumption, or restricted access to
particular species.

Archaeology at Shields Mound

Excavations by C.B. Moore

Originally composed of a large ramped mound and
peripheral midden areas, Moore (1894:204) described Shields
Mound as an unusual type for the St. Johns River drainage. In
many of Moore's Florida investigations, he concentrated on
sand mounds where cultural materials were usually present in
quantity. Although disappointing from this perspective, his
excavations in Shields Mound recovered a variety of cultural
materials. With regard to faunal remains, Moore (1895:464)
reported shell midden and part of the lower jaw of a bear.
Bone pins were reported and Moore (1895:464) remarked that
there were "...often six or more with one interment." He
(Moore 1895:464-465) also noted the recovery of the
tarsometatarsus of a male turkey (lower leg just above the foot)


VOL. 58(3-4)




Figure 1. Location of Mill Cove Complex and other sites mentioned in this paper.

and drew parallels to an image, drawn by LeMoyne and
published by de Bry, of Outina's "leading counsellor" with a
bird foot ear ornament (Hulton 1977:Plate 106). Shell beads,
both columella segment and disk types, are reported along
with perforated Olivella sp. shells associated with a burial.
Canine teeth and "shell imitations" of canine teeth (Moore
1895:466), a drum tooth (a bony fish), and shark teeth were
associated with burials. Moore (1895:466) assumed that many
of the teeth he recovered were bear, but stated that one had
been identified as grey wolf.

Excavations by Ashley and Collaborators

The faunal remains discussed here were recovered from
excavation Units 1-4 (Kinzey's Knoll), Unit 5 (Kinzey's West,
northwest of the mound), Unit 6 (Kinzey's South, nearest the
mound), and Units 7 and 8 (Bluff Midden, farthest from the
mound). Also included are three shovel tests from Reeves
Rise, intermediate between the Bluff Midden and Kinzey's
Knoll. Recovery methods included dry screening through 1/4-

in hardware cloth and sampling with 1/8-in, 1/16-in, and 3/64-
in screens and columns. This study reports only the 1/4-in
sample which has been analyzed for all units and the three
shovel tests. The 1/16-in and 3/64-in samples await future
analysis. During excavation, invertebrate remains were
separated during screening and measured in liters. Based on
these data, the excavators concluded that oysters comprised 90
percent of the species present. For this study, the subsistence
contribution of oysters was estimated'.
Where possible, vertebrate data from the excavation units
have been compiled. Table 1 provides a species list for each
excavation area. Units 1, 2, 3, and 4 (Kinzey's Knoll) were
placed contiguously and are treated as one unit. Both Unit 5
and Unit 6 produced a small assemblage of vertebrate and
invertebrate fauna. Units 7 and 8 (Bluff Midden), farthest
from the mound, were also placed contiguously and their data
are joined. Although the Reeves Rise shovel tests were not
contiguous, their data are also compiled for the purposes of
Intrusive and commensal fauna occur within the assem-


2005 VOL. 58(3-4)


* *





St. Johns River N
St. Johns River N

* S

* a 0

. .I --1100N
TU5 T 4-O

Roe 0 No

SKinzey'sf I
Knoll jx |
*| JJ -1000N

TU5 TU1-4


MMound 900N

900E 1000E

Figure 2. Shields Mound (8DU12) locale illustrating location of mound, Kinzey's Knoll area, Reeves Rise area, Bluff Midden,
and relative positions of shovel tests (adapted from Ashley 2003:159).


blages. By intrusive is meant animals of European origin not
present in the Early Mississippian Period. In Units 3 and 4,
European rat (Rattus rattus) remains were recovered (Unit 3,
level 2: 10-20 cm; Unit 4, level 8: 70-80cm). Unit 8 produced
domestic cattle (Bos taurus, level 1:0-10 cm), and European
rats (levels 2:10-20 cm and 6:50-60 cm). The cattle remains
are four bones of the ankle. All rat remains are femurs. In the
Reeves Rise shovel tests, chicken (Gallus gallus) remains are
present (level 2:20-40 cm and level 5:80-100 cm). Intrusive
fauna listed here are not included in the species list (Table 1)
or compiled tables and do not figure in biomass or minimum
number of individual (MNI) calculations.
Commensal fauna include shrews (Soricidae), moles
(Scalopus aquaticus), rice rat (Oryzomys sp.), cotton rats
(Sigmodon hispidus), and frogs (Ranidae). The two rats are
indigenous species and appear frequently enough in prehistoric
faunal assemblages that they either are intruding burrowers

who commonly die in midden deposits or represent food. The
commensal animals have not been excluded, raising the MNI
slightly but scarcely affecting the estimated biomass.
The number of identified specimens (NISP) of worked
bone is 108. Most of the worked bone was identified as bone
pin fragments (NISP=56), but other less identifiable tools are
present in the assemblage (Penders, this issue). These frag-
ments have been included in the count, but generally excluded
from the weight category. The weight of butchered bone
(NISP=9) has been included and thus figures in the calculation
of estimated biomass.

The Shields Mound Faunal Assemblages

The Shields Mound collection is divided into assemblages
representing three discrete excavation areas of the site:




Table 1. Shields Mound species lists by assemblage, 1/4-in sample.

SKinzeys Knoll Kinzey's West Kinzey's Knoll Bluff Midden Reeves Rise
Scientific Name Common Name Units 1-4 Unit 5 Unit 6 Units 7 and 8 Shovel Tests
Mammalia, Large Probably deer or bear x x x x x
Mammalia, Medium Probably raccoon x x
Mammalia, Small Probably rabbit or squirrel x x x x x
Mammalia Unidentified mammal x x x x
Mammalia cf., Procyon Probably raccoon x
Mammalia cf., Ursus Probably bear x
Didelphis virginiana Eastern opossum x x
Soricidae Shrews x
Scalopus aquaticus Eastern mole x x
Sylvilagus sp. Rabbits x x x
Rodentia Rodents x x x x x
Sciurus sp. Squirrels x
Sciurus f., carolinensis Probably grey squirrel x
Sciurus carolinensis Grey squirrel x x x x
Sciurus niger Fox squirrel x x x x x
Oryzomys sp. Rice rat x
Sigmodon hispidus Hispid cotton rat x x x x
Delphinidae Dolphins x x
Tursiops truncatus Bottlenose dolphin x
Urocyon cinereoargenteus Grey fox x
Ursus sp. Bear x x x
Procyon lotor Raccoon x x
Lynx rufus Bobcat x
Artiodactyla Ungulate, probably deer x x x
Odocoileus virginianus Whitetail deer x x x x x
Aves All birds x x x x x
Podiceps auritus Homed grebe x x
Podilymbus podiceps Pied-billed grebe x
Phalacrocorax auritus Double-crested cormorant x
Phalacrocorax carbo Great cormorant x
Ardeidae Herons x
Ardea alba Great egret x
Anatidae Ducks, geese, and swans x
Anatidae Unidentified small ducks x
Anatidae Unidentified medium-sized ducks x x
Anatidae Unidentified large ducks x x
Anatidae Unidentified geese x
Anatidae cf., Mergus Duck, probably a merganser x
Mergus cucullatus Hooded merganser x
Branta canadensis Canada goose x
Meleagris gallopavo Wild turkey x x x
Larus sp. Gulls x x
Larus argentatus Herring gull x x
Larus delawarensis Ring-billed gull x x
Colinus virginiana Bobwhite quail x
Zenaida macroura Mourning dove x
Strix varia Barred owl x
Colaptes auratus Northern flicker x
Passeriformes Songbird x
Corvus ossifragus Fish crow x x
Alligator mississippiensis American alligator x x
Testudines All turtles x x x x x
Kinosternidae Mud and musk turtles x x x
Terrapene carolina Eastern box turtle x x x
Trachemys sp. Pond slider x x x
Malaclemys terrapin Diamondback terrapin x x x
Deirochelys reticularia Chicken turtle x x x x
Gopherus polyphemus Gopher tortoise x x x x x
Apalone ferox Softshell turtle x x x
Cheloniidae Sea turtle x

2005 VoL. 58(3-4)



Table 1. Shields Mound species lists by assemblage, 1/4-in sample, continued.

Kinzeys Knoll Kinzeys West Kinzeys Knoll BluffMidden Reeves Rise
Scientific Name Common Name Units 1-4 Unit 5 Unit 6 Units 7 and 8 Shovel Tests
Serpentes Snakes x x x
Farancia sp. Mud snake x
Colubridae Non-poisonous snakes x
Anura Frogs and toads x x
Anura cf., Ranidae Probably frog x
Caudata Amphibians, probably salamander x
Siren lacertina Greater siren x
Osteichthyes All bony fish x x x x x
Accipenser sp. Sturgeons x x x x
Lepisosteus sp. Gar x x x x
Amia calva Bowfin x x x x
Elops saurus Ladyfish x x x x
Anguilla rostrata American eel x x x x
Clupeidae Herrings x x x x x
Esox sp. Pickerel x
Siluriformes Catfishes x x x x x
Ictaluridae Freshwater catfishes x x x x
Ariidae Marine catfishes x x x x
Arius felis Hardhead catfish x x x x
Bagre marinus Gafltopsail catfish x x x x
Lepomis sp. Sunfish x x
Lepomis microlophus Redear sunfish x x
Micropterus sp. Bass x x
Pomatomus saltatrix Bluefish x x
Caranx sp. Jacks x x x x x
Lobotes surinamensis Tripletail x
Sparidae Porgies x x x
Sparidae/Sciaenidae Either porgy or drum x
Archosargus probatocephalus Sheepshead x x x x x
Sciaenidae Drums x x x x x
Bairdiella sp. Perch x
Cynoscion sp. Seatrouts x x x x x
Micropogonias undulatus Atlantic croaker x x x x
Pogonias cromis Black drum x x x x x
Sciaenops ocellatus Redfish x x x x
Stellifer lanceolatus Star drum x
Mugil sp. Mullets x x x x x
Sphyraena barracuda Great barracuda x
Bothidae Flounders, probably left-eye x x x x x
Gobionellus sp. Gobies x x x
Diodontidae Pufferfish x
Squaliformes Sharks x x x
Carcharhinidae Requiem sharks x x
Carcharhinus sp. Requiem sharks x x
Carcharhinus altimus Bignose shark x
Carcharhinus leucas Bull shark x
Carcharhinus limbatus Blacktip shark x
Galeocerdo cuvieri Tiger shark x
Negaprion brevirostris Lemon shark x
Rajiformes Rays x x x x x
Rhinobatidae Guitarfish x
Decapoda cf., Callinectes sp. Probably blue crabs x x x x
Crassostrea virginica Eastern oyster x x x x x

Kinzey's Knoll (Units 1-4, Unit 5, and Unit 6), Reeves Rise
(three shovel tests), and the BluffMidden (Units 7 and 8). A
series of Appendices (1-5) provides common
zooarchaeological measures for each assemblage: NISP,
weight of recovered specimens (in grams), estimated biomass
(in grams), the degree of thermal alteration (burned, charred,
or calcined bone), the presence of worked bone (tools, items of
personal adornment, or butchered bone), and MNI. Biomass
is an estimate of the meat-weight contribution of a specific

taxon using allometric scaling from the weight of identified
remains (Reitz et al. 1987b). Values used in this study are
provided in Table 2.
MNI was determined using the principle of paired elements
and a consideration of size, age, and gender, where possible.
Table 3 compiles information on excavation unit size, midden
depth, and a summary of recovered faunal remains for each
discrete assemblage.



Table 2. Biomass values used in this study and their published sources.

Taxa Slope (b) Y-intercept (a) Published Source
Mammals 0.90 1.12 Reitz and Wing 1999:72
Birds 0.91 1.04 Reitz and Wing 1999:72
Alligator 0.89 0.91 Reitz, Marrinan, and Scott

Turtles 0.67 0.51 Reitz and Wing 1999:72
Snake/Siren 1.01 1.17 Reitz and Wing 1999:72
Bony Fish (Osteichthyes) 0.90 0.81 Reitz and Wing 1999:72
Catfishes (Siluriformes) 0.79 0.85 Reitz and Wing 1999:72
Bass and sunfish (Centrarchidae) 0.84 0.76 Reitz, Marrinan, and Scott
Porgies (Sparidae) 0.92 0.96 Reitz and Wing 1999:72
Drums (Sciaenidae) 0.74 0.81 Reitz and Wing 1999:72
Flounders (Pleuronectiformes) 0.89 1.09 Reitz and Wing 1999:72
Sharks and Rays (Chondrichthyes) 0.86 1.68 Reitz and Wing 1999:72
Crabs (Callinectes sp.) 0.99 0.92 On file, Environmental
Archaeology Laboratory,
Florida Museum of Natural
History, Gainesville, Florida
Oysters (Crassostrea virginica) 0.97 -0.77 Reitz and Wing 1999:72

Kinzey 's Knoll Group (Units 1-4, Unit 5, and Unit 6)

These units represent the closest midden deposits to the
mound that have been sampled, and they are quite varied.
Units 1-4 (Appendix 1) had the most dense and deepest shell
midden deposit and the greatest species diversity, estimated
biomass, and MNI values. Unit 5 (Appendix 2) and Unit 6
(Appendix 3) produced only a small quantity of vertebrate and
invertebrate remains. Table 4 presents faunal summaries by
class for each excavation area. Units 1-4 provide the most
complete view of vertebrate fauna (Table 4). Bony fishes
comprise the bulk of that assemblage with an estimated
biomass of 49 percent. Among the bony fishes, several
families are major contributors: catfish (both freshwater and
marine), sheepshead, drums, mullet, and flounders.
Mammalian fauna are second to bony fishes consistently,
except in the Reeves Rise shovel tests. The primary contribu-
tors are: rabbits, squirrels, bears, raccoons, and deer. The data
indicate that deer are the most important mammals in this
assemblage, but by contrast to bony fishes, their contribution
is less substantial in estimated biomass and MNI values.
Squirrels, particularly fox squirrels, are consistently present as
are rabbits. Raccoons (MNI=2), usually a significant contribu-

tor to faunal assemblages of this period, are not well-repre-
sented. Opossums are virtually absent (NISP=1). Bear
remains (NISP=29) in the Kinzey's Knoll units are concen-
trated in Units 1-4 (NISP=23). The skeletal elements are
predominately phalanges, particularly the first and third
(claws). Bones of the lower forelimbs, lower hind limbs, and
a single baculum were recovered. Bones of the cranium and
teeth were not identified.
Reptilian fauna, overwhelmingly turtles, is third in level of
contribution. A variety of turtles is present indicating concen-
tration on collecting terrestrial species (gopher tortoise and
box turtle) but freshwater, marine, and saltmarsh environ-
ments are also represented in low numbers. Alligator and
snake remains are present but contribute little. The mud snake
(Farancia sp.) is the only snake identified, and snake remains
are very few (NISP=9).
The avian assemblage from the Kinzey's Knoll area,
particularly Units 1-4, can be separated into five groupings:
grebes and cormorants, geese and ducks, turkeys, gulls, and
others. These assemblages contained the highest numbers of
birds, whether for food or byproducts (e.g., feathers, bone for
beads or tubes). The most consistently important are ducks,
geese, and turkeys. Species occurrence among birds provides

2005 VOL. 58(3-4)


Table 3. Estimated Excavated Volume and Standard Analysis Measures for Recovered Faunal Remains.

Estimated Estimated
Unit Midden Unit Projected NISP Weight in Vertebrate Invertebrate Heat Worked or MNP**
depth dimensions cubic volume grams* Biomass in Biomass in Altered Butchered
Kilograms Kilograms
1 85 cm 1x2 m 1.7 6,825 2,246.7 25.9 94.6 64 24 135
2 80 cm 1x2 m 1.6 7,156 2,146.4 23.8 64.5 91 23 118
3* 80 cm 1x2 m 1.6 9,155 2,876.4 33.4 77.8 89 24 213
4 80 cm 1x2 m 1.6 2,969 1,535.3 20.9 73.6 50 11 89
Totals 6.5 26,105 8,804.1 85.4** 297.9** 294 82 437

5 90 cm lx2 m 1.8 354 64.0 1.2 6.5 4 16

6 N 2 90 cm 1x2 m 2.25 4,700 910.6 12.4 21.2 44 13 89
S V 1.35 m

7 65 cm lx2 m 1.3 3,404 1,437.0 20.1 46.9 69 7 75
8 60 cm lx2 m 1.2 3,394 891.1 12.6 29.0 48 2 87
Totals 2.5 6,798 2,328.1 37.8** 94.2** 117 9 138

ST1032N,1000E 40 cm 50x50 cm 0.1 1.434 264.8 4.1 59.9 15 35
ST 1031N,1003E 40 cm 50x50 cm 0.1 212 127.8 2.0 2.6 1 3 20
ST 1029N,1000E 40 cm 50x50 cm 0.1 637 233.6 4.6 3.0 8 1 36
Totals 0.3 2,283 626.2 8.9** 68.2** 24 4 28

Unit 3 includes materials from Feature 1 which it shares with Unit 1. Feature 1 is a pit (Ashley 2003:Figure 5-9) encountered at 80 cm below surface.
** Biomass values are recalculated for compiled weights. These totals reflect values reported in Appendix 1-5. Vertebrate values do not include bone tools.
** MNI values have been recalculated for assemblage composites.

Table 4. Faunal remains by class. Percentages represent vertebrate fauna only. Intrusive domestic fauna have been removed from these data. Worked bone
(tools, adornment, and butchered) and thermally altered (burnt) bone are included.

Kinzey's Knoll Kinzey's Group Kinzey's Group Bluff Midden Reeves' Rise
Units 1-4 Unit 5 Unit 6 Units 7 and 8 3 Shovel Tests
Class NISP Biomass MNI NISP Biomass MNI NISP Biomass MNI NISP Biomass MNI NISP Biomass MNI
in Kg. in Kg. in Kg. in Kg. in Kg.
Mammals 1,092 31.3 37 42 0.6 3 135 2.5 10 315 1.3 17 54 1.3 6
Birds 429 4.5 28 16 .09 1 50 0.6 5 47 0.5 5 29 0.2 2
Reptiles 1,125 5.8 23 19 0.1 3 172 1.1 7 195 2.1 13 108 1.4 6
Amphibians 4 .007 2 2 .004 1 1 0.0 1 -
Bony Fish 22,509 41.9 338 181 0.3 8 4,023 6.5 64 5,944 13.7 100 1,993 6.0 41
Sharks 42 1.8 7 1 .02 1 21 1.1 2 12 0.5 2 1 .006 1
Unidentified 904 95 297 284 98 -
Totals 26,105 85.4* 435 354 1.1* 16 4,700 11.8* 89 6,798 29.7* 138 2,283 8.9* 56
Crabs 66 0.5 17 15 0.1 4 4 .044 2 2 .06 2
Oysters 297.9 6.5 21.2 74.4 68.2
Worked Bone 82 13 9 4 -
Burt Bone 294 44 100 24
* Indicates recalculated biomass values for assemblage composites.


the strongest seasonal evidence. The homed grebe (Podiceps
auritus), great cormorant (Phalacrocorax carbo), ring-billed
gull (Larus delawarensis), herring gull (Larus argentatus),
geese, and many ducks are winter residents of the Southeast.
Amphibians are poorly represented. A single fragment of
the large aquatic salamander (Siren lacertina) suggests very
low usage of this amphibian. The frog remains may be
commensal. Cartilaginous fishes (sharks and rays) appear to
be more important for tool use or as items of personal adorn-
ment. Sharks and rays add no perceptible increase to the
biomass estimates since most of the remains are teeth and
vertebrae having modifications that indicate use as tools or
items of personal adornment.
A total of 294 fragments of burnt bone was present in Units
1-4, 4 in Unit 5, and 44 in Unit 6. Most numerous were
fragments of bony fish (NISP=102; 34.7 percent) and mam-
mals (NISP=90; 30.6 percent). Table 5 details the distribution
of burnt bone and shows that its recovery is very low in
proportion to assemblage size (0.9 to 1.5 percent). In the
Units 1-4 assemblage, the greatest incidence of burnt bones
was among fragments that could be assigned only to class
(e.g., bony fish or mammal). Unit 5 had the lowest incidence
of burnt bone.
The majority of worked bone was recovered from Units 1-4
(NISP=82; 76 percent). Greatest vertebrate biomass was
generated from the same levels indicating an association of the
pin fragments with the densest faunal deposits (Table 6). Such
an association may indicate that the bone pins were used in the
same activities as the fauna or that this location was appropri-
ate for the discard of broken bone tools.

Bluff Midden (Units 7 and 8)

These excavation units intersected a concentrated shell
midden 60 to 65 cm in depth (Table 3). The faunal constitu-
ents are similar in proportion to Units 1-4 in the Kinzey's
Knoll group (Table 4). They also are similar in that bony
fishes are dominant, and the same taxa (families or genera)
that are important in the larger assemblage (Units 1-4) are also
prominent contributors to this assemblage (Appendix 4). The
Bluff Midden is the only excavation area to produce sea turtle
remains and, among the birds, the only area in which wood-
pecker (Colaptes auritus northern flicker) or owl (Strix varia
- barred owl) remains were recovered. These two birds are
mentioned because of their significance in Mississippian
iconography and traditional Southeastern cosmology, respec-
tively. Burnt bone (Table 5) is not common in this assemblage
(NISP=100), and only nine specimens of worked bone were
recovered (Table 6).

Reeves Rise

The compiled assemblage from the three shovel tests,
because of their relative size (Table 3) and shallow depth, is
small. Appendix 5 details the recovered fauna. The integrity
of these deposits is also a matter of concern since the recovery
of domestic fowl from level 5 suggests that the midden has

been deeply disturbed. In this assemblage, burnt bone is not
numerous (NISP=24) nor is worked bone (NISP=4). A single
bone pin fragment was recovered from this midden (level 1).

The Faunal Classes

This discussion provides an assessment of the composition
of the classes, an evaluation of the skeletal elements present,
and also considers animals not present in this sample.

Invertebrate remains

The biomass estimates for invertebrate fauna indicate a
contribution substantially higher than vertebrate fauna (Table
3). Although oysters are the dominant invertebrate, Ashley
(2003:164-173) also lists quahog clam, marsh clam, stout
tagelus, Atlantic ribbed mussel, and whelks as components of
the midden fill. Crab remains, identified as blue crab
(Callinectes sp.), were recovered in small quantities in all
midden deposits. Their occurrence was highest in Units 1-4.
Blue crabs are present throughout the St. Johns River drainage
(McLane 1955; Tagatz 1965).


White-tailed deer is usually a major constituent of mamma-
lian faunas in Southeastern faunal assemblages, and the
sample from Shields Mound does not deviate from this
expectation. The epiphyseal fusion rates for deer (Reitz and
Wing 1999:76) suggest that most of the Shields Mound
individuals ranged in age from approximately 11 months to
greater than 42 months. Although these data provide insight
about the composition of the deer death assemblage, fusion
rates do not assist in confidently pinpointing the time of year
these deer were killed. The distribution of elements suggests
that the complete carcass was processed at the site, but there
is slightly more representation of the hind limbs.
Small game, in the form of squirrels (grey and fox) and
rabbits, is an important part of the mammalian assemblages.
Also present are single individuals of bobcat (Lynx rufus), fox
(Urocyon cinereoargenteus), and dolphin (Delphinidae). The
dolphin remains consist primarily of teeth (NISP=4) all of
which, from the damage done to the enamel, seem to have
functioned as tools. Three of the teeth were recovered from
Units 1-4, the fourth from the Bluff Midden. A single dolphin
(Tursiops truncatus) inner ear ossicle also was recovered.
Some mammals generally expected to be significant contribu-
tors, particularly raccoons (MNI=2) and opossums (MNI=1),
are present in very low numbers.
Bear (NISP=31) is present in Units 1-4 (NISP=23,
MNI=3), Unit 6 (NISP=6, MNI=1), and Bluff Midden assem-
blages (NISP=2, MNI=I). The bear remains are postcranial
and the majority are elements of the distal lower limbs (radius,
ulna, fibula) and feet suggesting removal of the paws. A
single fetal humerus represents an upper forelimb. The
majority of elements (NISP=15) are phalanges. This distribu-
tion of elements, and the knowledge that bears were a major



Table 5. Inventory of thermally altered (burnt) bone by class. Percentages indicate value within class except for totals which are calculated for the entire
thermally altered bone collection.

Kinzey's Knoll Kinzey's West Kinzey's South Bluff Midden Reeves Rise
Categories Units 1-4 Unit 5 Unit 6 Units 7 and 8 Shovel Tests Totals

Number burnt 294 4 44 100 24 466

Total Vertebrate NISP 26,105 354 4,700 6,797 2,283 40, 239

Percentage burnt 1.1 1.1 0.9 1.5 1.1 1.2


Mammals 90 8.2 2 4.8 7 5.2 32 10.2 5 9.3 136 29.18

Birds 12 2.8 1 2.1 13 2.79

Reptiles (Turtles only) 40 3.6 9 5.4 7 3.6 2 58 12.45

Bony Fishes 102 0.5 2 1.1 15 0.4 45 0.8 14 0.7 178 38.20

Sharks and Rays -

Unidentified vertebrate 50 5.5 13 4.4 15 5.3 3 3.1 81 17.38

Totals 294 4 44 100 24 466 100.00

Table 6. Inventory of Worked Bone. Percentages indicate value within Class for individual collections except for Totals which were calculated for the entire
worked bone collection.

Categories Kinzey's Knoll Kinzey's Knoll Area Bluff Midden Reeves Rise
Units 1-4 Unit 6 Units 7 and 8 Shovel Tests Totals

NISP worked 82 13 9 4 108

Bone pin fragments 46 6 3 1 56

Butchered 5 -3 1 9

Other 31 7 3 2 43

Total vertebrate NISP 26,105 4,700 6,798 2.283 39,886

Percentage worked 0.3 0.3 0.1 0.1 0.3


Mammals 17 1.6 2 1.5 4 1.3 1 1.9 24 22.22

Birds 1 3.4 1 0.93

Reptiles (Turtles) 6 0.5 1 33.3 7 6.48

Bony Fishes 2 0.0 1 0.0 3 2.78

Sharks 12 28.6 1 4.8 13 12.03

Unidentified vertebrate 45 5.0 10 3.4 4 1.4 1 1.0 60 55.56

Totals 82 -13 9 -3 108 100.00


source of fat, suggests that the larger bones of the body were
processed elsewhere or are present in this midden deposit in
such a fragmentary condition as to be unrecognizable. It also
suggests that the head and teeth may have been reserved for
ceremonial usage and deposited elsewhere as Moore
(1895:464) reported.


Birds are unusually abundant in these assemblages. David
Steadman (personal communication, 2000), of the Florida
Museum of Natural History, characterized this assemblage as
a cold weather grouping, specifically November to March. In
addition to birds previously mentioned as seasonal indicators
and others commonly using water body habitats, turkeys,
mourning dove, and quail remains also indicate the use of
species inhabiting terrestrial habitats. Although the greatest
percentage of avian biomass was contributed by geese and
ducks, turkeys are a staple resource.


Turtles are the dominant reptile group and gopher tortoise
(Gopheruspolyphemus) is consistently the greatest contributor
to biomass, the highest values occurring in Units 1-4. This
tortoise is a slow-moving, burrowing land-dweller. Another
land-dweller is the box turtle, recovered in small quantities
across the site. Two drilled specimens (Units 1 and 3) indicate
that the box turtle served other needs. Water turtles include
pond sliders, chicken turtle, diamondback terrapin, mud or
musk turtle, and softshell turtles, most represented by a single
individual. Remains of a juvenile sea turtle were recovered
from the Bluff Midden. Given the quantities of estimated
biomass contributed and MNI values, gopher tortoise must be
considered a staple. Alligator (NISP=13) and snake remains
are present (NISP=12), but they are not significant contribu-
tors to biomass or MNI. At least one of the alligator speci-
mens, judging from the scute size and condition, was from a
very young individual.

Bony Fishes

As indicated previously, bony fishes make the most
substantial contribution to the food base in these assemblages.
Among identified species, the catfishes indicate use of both
freshwater and marine species. Other freshwater species are
present (pickerel, sunfish, bass, and gobies) as well as species
considered to be marine ladyfishh, herrings, tripletail, bluefish,
barracuda, and jacks). McLane's (1955) extensive survey of
the St. Johns River documented the distribution of many
marine "invaders" along the river course, a situation that
complicates efforts to determine prehistoric environmental
exploitation preferences. In truth, most of the fish species
identified in the Shields Mound assemblages could have been
taken from the river near the site.
A very small number of fragments identifiable as sturgeon
(Accipenser oxyrhincus, Atlantic sturgeon, or Accipenser

brevirostris, shortnose sturgeon) were recovered (NISP=25),
the majority (NISP=16) in the Bluff Midden. These fish are
anadromous, living most of their lives in marine waters, but
returning to freshwater rivers to spawn. Adult Atlantic
sturgeon have been reported from southern river systems in
mid-to-late February (Gilbert 1989:8). American eels are
catadromous fishes that live most of their lives in freshwater,
but migrate as adults during the winter to the ocean to spawn
and die. Eel remains (NISP=201 vertebrae) are more abun-
dant in the Kinzey's Knoll assemblages (NISP=177) than in
the BluffMidden (NISP= 1) or Reeves Rise (NISP=13). Both
sturgeons and adult eels suggest fall-winter activity at the site.
The gar (Lepisosteus sp.) remains are atypical. Faunal
assemblages from shell middens typically contain hundreds, if
not thousands, of gar scales. In the Shields Mound assem-
blages more vertebrae (NISP=158) and cranial fragments
(NISP=74) were recovered. Fewer scales (NISP=49) are
present in these assemblages suggesting that these fish were
"skinned" elsewhere. Because many of the fragments are from
the elongated maxilla and dentary of the fish (NISP=34),
selection of elements also may be a factor. A single butchered
dentary (Unit 2, level 10) suggests separation of the toothed
parts of the dentary from the jaw hinge. Garjaws are used in
Southeastern traditional ritual practices, particularly in
preparation for various dances (Swanton 2000:554).

Cartilaginous Fishes

Five different shark species were identified: bignose,
blacktip, bull, lemon, and tiger sharks. Sharks are represented
by both teeth (NISP=17) and vertebrae (NISP=22). It is
possible that all of the teeth functioned as tools, but clear
evidence of use-wear was present only on 13 specimens. Rays
are represented by mouth plate segments (NISP=26) and
vertebrae (NISP=12). No spine fragments were recovered.

Diversity and Equitability

Measures of diversity and equitability (Cruz-Urribe 1988,
Grayson 1984, Reitz and Wing 1999) are used to evaluate the
richness of the environment (diversity) and the pattern of
choices made by human groups exploiting these environments
(equitability). The approach that is most commonly used to
calculate diversity is the Shannon-Weaver Index (Shannon and
Weaver 1949). In these calculations, the values for diversity
(H') are read on a scale of 0-5, with 5 the highest value. For
equitability (V'), the Sheldon (1969) scale is used: values of 0-
1, with 1 equaling a choice of one of each taxon available in
the environment. Table 7 indicates diversity and equitability
values calculated for each unit using only taxa for which an
MNI was reported (Appendices 1-5).
Both MNI and estimated biomass have been used to
calculate diversity and equitability because they are derived
from different source values and provide different perspectives.
When MNI is considered, Units 1-4 have higher diversity and
lower equitability values suggesting a broad use of resources
(species) but concentration on a more restricted group. When

2005 Vol. 58(3-4)


Table 7. Diversity (H') and Equitability (V') values calculated for the Shields Mound assemblages (Hale 1989). Diversity
values range from 0 to 5; Equitability values range from 0 to 1.

Excavation Unit MNI Estimated Biomass

Diversity Equitability Diversity Equitability

Kinsey's Knoll Units 1-4 3.3 .76 2.9 .68

Kinsey's West Unit 5 2.7 1.0 2.2 .82

Kinsey's South Unit 6 3.3 .86 3.1 .82

Reeves Rise Shovel Tests (3) 2.9 .88 2.7 .81

Bluff Midden Units 7 and 8 3.2 .85 2.5 .65

biomass values are used, Units 1-4, Reeves Rise, and the Bluff
Midden have moderate values suggesting similarity in species
usage. The low diversity value in Unit 5 probably results from
the small sample; equitability is clearly affected by sample
size. The two equitability groupings Units 1-4 and the Bluff
Midden compared to Unit 5, Unit 6, and Reeve's Rise are a
strong reflection of the effects of small sample size.

Animal Use in Southeastern Societies
(Ethnohistorical Sources)

Although the Spanish entradas of the first half of the
sixteenth century are believed to have skirted the study area,
the French presence in the St. Johns River brought natives and
Europeans into close association beginning in 1562. Rene de
Laudonniere, the French leader at Ft. Caroline, was plagued by
food shortages and forced to rely on the indigenous population
(Bennett 2001). Although it is a recurrent theme in his
journal, specific information about native subsistence practices
is lacking. "King Molona," a local chieftain, welcomed him
with a feast of bread and fish (Bennett 2001:76) and he
recorded presents of deer, turkey, panthers, bear cubs, fish,
acorn and corn meal, and cassina from various chiefly person-
ages. Although Laudonniere also observed that indigenous
people dispersed during the months of January, February, and
March to live by hunting since little else was available, it is
not clear if this is a prehistoric pattern or a post-Columbian
accommodation. Ethnohistoric sources for the sixteenth
through eighteenth centuries have been compiled, summa-
rized, and evaluated by Swanton (1979, 2000) and Hudson
(1976) and the following characterization owes its substance
to these sources.
Indigenous Southeastern peoples, particularly those of
Muskhogean linguistic groupings2, lived in a universe that
stressed opposition, balance, and purity. Humans lived in This
World, between an Upper World associated with the sky,
sacred fire (sun), time past, and order and an Under World

associated with ghosts, monsters, future time, and disorder.
Animals had specific occurrences within this tripartite
existence determined largely by their behavior. Birds, because
of their ability to fly, were creatures of the Upper World.
Chief among them was the bald eagle, but the falcon was also
significant. Animals of the Under World-fish, reptiles,
amphibians, and possibly insects-were classed together and
considered vermin, the dominant representative being the
rattlesnake. The four-footed animals existed in This World
and most significant among them was the deer.
Maintaining balance among these worlds and their
associated forces required a knowledge of the qualities of
animals derived from their natural behaviors and mythological
associations, performance of the proper hunting rituals, proper
preparation of the human and the animal for consumption, and
the state in which the animal could be consumed safely. Many
animals were not eviscerated, but cooked after only the hide or
skin was removed. Several observers stress that Southeastern
people did not consume raw meat, but cooked it thoroughly
(Swanton 1979:368). They also avoided animals that were
found dead, when the manner of their dying could not be
explained. Animals whose behavior was atypical required
special concern (e.g., birds that fly at night, bats). Scholars
such as Brown (1989), Emerson (1989), Knight (1986), and
Strong (1989) have observed that Mississippian elites reserved
the symbols of these relationships for their own use and
legitimization. Association with the animals of the Upper
World, for example, testified to the historic rights of the elites
to their position and authority in society as well as transferring
to them, properties associated with sacred animals.

Prehistoric Faunal Usage in the Coastal Zone

Are the Shields Mound assemblages typical or atypical of
faunal usage in the coastal zone? I have chosen a group of
shell midden sites from the coastal zone that represent a
temporal span from Late Archaic to Mississippian periods




(Table 8). All of the sites are midden accumulations com-
prised of large quantities of invertebrate remains within which
cultural materials and vertebrate fauna have been recovered
(Figure 1). The Late Archaic Period site is the Cannon's Point
shell ring (9GN57) located on St. Simons Island, Georgia
(Reitz et al. 1987a). Radiocarbon dates bracket the lower and
upper levels at 2240-1815 B.C., respectively (Marrinan
1975:35). The Second Refuge Site (38JA61) is located within
the Savannah National Wildlife Refuge and dated, in its basal
levels to 1070 B.C. (Lepionka et al. 1983:38). The site serves
as an example of an Early Woodland Period refuse deposit
(Reitz et al. 1987a). It is a shell midden composed of oyster
and a freshwater mussel (Elliptio icterina). Two sites are
included that represent the Middle Woodland Period: the
Artesian Well site on the Kings Bay Naval Base in Camden
County, Georgia (Quitmyer 1985:110-111) and the nearby
Greenfield #7 site (8DU5543) analyzed by deFrance (1993).
Both sites are identified as Swift Creek and both are fine-
screened column samples. For the Mississippian Period,
midden data from the nearby McCormack/Goodman midden
(8DU66) are used. Located on the east side of Mill Cove,
several miles distant from the Mill Cove Complex, the site was
identified and tested by Sears (1957:20-22) and excavated by
Jordan (1963). Vertebrate fauna from this site were analyzed
by Elizabeth S. Wing (1963). These data indicate that by Late
Archaic times, a foodbase incorporating a wide variety of
vertebrate and invertebrate species, but concentrated depend-
ence on a narrower set of species, was established in the
coastal zone (Marrinan 1975; Reitz 1982; Reitz et al. 1987a;
Wing 1977). This foodbase persisted until European contact.
The collections vary in species composition by environmental
zone, but they illustrate reliance on deer and on a restricted
group of animals (e.g., in the Shields Mound assemblages:
deer, turkeys, gopher tortoises, catfish, sheepshead, sea trout,
black drum, mullet, and flounders). Within this comparison,
the Shields Mound assemblages appear consistent with long-
established subsistence practices.
A variety of technologies is indicated by the Shields Mound
species list (Table 1). Foraging and collecting by women (and
children) seeking plant foods may account for species such as
the gopher tortoise and box turtle. The gopher tortoise may
have been targeted at its burrow and taken there by digging or
waiting for its emergence. Small game, such as rabbits,
raccoons, or opossums can be taken by snares laid along game
trails. Larger species such as fox, bobcat, or even deer could
be taken in this manner. Bear fat was a commodity much in
demand for food preparation and ritual needs (Swanton
1979:321-324). Bears contain the most fat during the winter.
The fetal specimen indicates a winter death. The presence of
so many bear phalanges is puzzling given the pervasiveness of
bear ceremonialism (Hallowell 1926) and the use of claws as
items of adornment. None of the distal phalanges (claws) was
drilled or modified. Although the Shields Mound specimens
do not seem to have been retained for ornamental purposes,
theirdeposition in the Kinzey's Knoll midden in high numbers
mirrors Bogan's (1983) findings at the Dallas Phase Toqua
site in Tennessee. There, bear remains were distributed across

the site, but the highest density of remains was in a midden
believed to represent high status disposal (Bogan 1983:315).
Baculae were recovered in Bogan's (1983:315) samples from
Toqua, all from either mound or high status midden contexts.
Squirrels and terrestrial birds can be lured to bait and taken
by deadfalls. Birds also can be taken with nets or bolas.
Dugout canoes were present in Florida, at least by Middle to
Early Archaic times (Wheeler et al. 2003), and waterborne
access to upstream freshwater locations as well as downstream
collection sites in saltier waters must be assumed. Evidence
for hook and line fishing is sparse in prehistoric sites. It is
possible that fish weirs were erected in tributary feeder streams
or in tidal creeks nearby. Fish poisons are another possibility
for tidal pools or quiet eddies of the river. Nets also could be
used and others (e.g., Lee et al. 1984) have cited the small size
of fishes in archaeological samples to argue for the presence of
fine mesh nets. Larger bony fishes and sharks, perhaps even
some turtles, might be taken with gorges. Crabs can be taken
in traps or nets. Oysters were gathered at low tide from bars
and hard clams from mud or sand flats at salt marsh margins
or along beaches. Also gathered were bivalves, such as marsh
clams and mussels, and gastropods such as whelks.
The LeMoyne/deBry engraving entitled, "Preparations for
a Feast," (Hulton 1977:148) illustrates cooking on a wooden
grill elevated over a fire. Mammals, fish, an alligator, and a
snake are represented. Although the mammals appear to have
been skinned, there is no artistic suggestion that they have
been eviscerated and early observers insist that they were not
(Swanton 1979:368). This practice may account for the low
number of bones that contain cut marks from butchering or
It is, I believe, important to consider what is not present.
Canids (dogs and wolves) are absent from the assemblages.
No species of aquatic, fur-bearing mammals were identified:
river otter, marsh mink, muskrat, or beavers. Mustelids,
musk-producing mammals (skunks, weasels, otters, and
minks), are not included. Squirrels appear to be a staple, but
other rodents are generally absent, particularly those with non-
terrestrial habits (beaver, muskrat, and water rat). Squirrels
may be perceived differently than rats and mice given their
bushy tails. The absence of opossum remains in any quantity
also may be significant and suggest that they were classed with
other bare-tailed mammals and avoided.

Evidence of Ritual Activities

The staging of ritual activities requires a suite of material
objects and immaterial behaviors prescribed by tradition to
successfully negotiate ritual need. The Kinzey's Knoll Units
1-4 data suggest ritual activity. The large number of ceramic
vessels (Rolland 2004, this issue), the presence of exotic
cultural materials (Ashley 2003:323), and the large number of
bone pins from Shields Mound (Moore 1895:464) and pin
fragments from the Kinzey's Knoll Units 1-4 assemblage
(Penders, this issue), suggest that these deposits are not typical
domestic discard. From their work in Arkansas Mississippian
sites, Jackson and Scott (1995:107) proposed "archaeological


2005 VOL. 58(3-4)


Table 8. Comparative faunal collections from selected sites.

9GN57 38JA61 Artesian 8DU5543 8DU12 8DU66 -
Well Midden
Late Archaic Early Swift Creek Swift Creek Mississippian Mississippian
Taxon Common Name Marrinan Reitz et al. Quitmyer deFrance Units 1-4 Wing 1963
1975 1987 1985 1993
Mammalia (large) Probably bear or deer x x
Mammalia (Medium) Probably opossum or x x
Mammalia (Small) Probably rabbit or squirrel x x x
Mammalia Unidentified mammals x x x
Didelphis virginiana Easter opossum x x x x x
Soricidae Shrews x x x
Cryptotis parva Least shrew x
Blanna brevicauda Short-tailed shrew x x x
Scalopus aquaticus Eastern mole x x
Sylvilagus sp. Rabbits x x x x x
Rodentia Rodents x x x x
Castor canadensis Beaver x
Glaucomys volans Flying squirrel x
Sciurus carolinensis Grey squirrel x x x x
Scirurus niger Fox squirrel x
Oryzomys sp. Ricerat x x x x
Peromyscus sp. Glossy rat x
Sigmodontidae New World rodents x
Sigmodon hispidus Hispid cotton rat x x x x
Cetacea Whales x
Delphinidae Dolphins x
Tursiops truncatus Bottlenose dolphin x
Lutra canadensis River otter x x
Mustela vison Marsh mink x
Lynx rufus Bobcat x x
Canis familiaris Domestic dog x
Urocyon cinereoargenteus Grey fox x
Ursus americanus Black bear x x x
Procyon lotor Raccoon x x x x
Odocoileus virginianus White tail deer x x x x x x
Aves Unidentified birds x x x
Podiceps auritus Homed grebe x x
Podilymbus podiceps Pied-billed grebe x
Phalacrocorax auritus Double-crested cormorant x x x
Phalacrocorax carbo Great cormorant x
Ardeidae Herons x x
Ardea alba Great egret x
Ardea herodias wardi Great blue heron x
Anatidae Swans, geese and ducks x
Branta canadensis Canada goose x
Bucephala clangula Common goldeneye x
Mergus culcullatus Hooded merganser x
Mergus cf., serrator Red-breasted merganser x
Buteo ineatus Red-shouldered hawk x
Colinus virginiana Bobwhite quail x
Meleagris gallopavo Wild turkey x




2005 VOL. 58(3-4)

Table 8. Comparative faunal collections from selected sites, continued.

9GN57 38JA61 Artesian 8DU5543 8DU12 8DU66 -
Well Midden
Late Archaic Early Swift Creek Swift Creek Mississippian Mississippian

Taxon CommonName Marrinan Reitzetal. Quitmyer deFrance Units 1-4 Wing 1963
1975 1987 1985 1993
Rallidae Rails x
Scolopacidae Sandpipers x
Catoptrophorus Willet x x

Larus sp. Gulls x
Lans argentatus Herring gull x x x
Larus delawarensis Ring-billed gull x
Zenaida macroura Mourning dove x
Passeriformes Songbirds x x
Stix varia Barred owl x
Colaptes auratus Northern flicker x
Corvus ossifiagus Fish crow x
Alligator mississippiensis American alligator x x
Testudines Unidentified turtles x
Chelydra serpentina Snapping turtle x x
Kinostemon sp. Mud turtle x x x x x
Terrapene carolina Eastern box turtle x x x
Malaclemys terrapin Diamondterrapn errapin x x x
Trachemys sp. Pond sliders x x x x x
Trachemys cf, floridana Pond turtle x
Deirochelys reticularia Chicken turtle x x x
Gopherus polyphemus Gopher tortoise x x
Apalone ferox Softshell turtle x x x
Cheloniidae Sea turtles x
Iguanidae Lizards x x x
Lacertilia Lizards x
Anolis carolinensis Chameleon x x
Colubridae Non-poisonous snakes x x x
Farancia sp. Mud snake x x
Nerodia sp. Water snakes x x
Coluber constrictor Black racer x x
Lampropeltis sp. Kingsnakes x
Viperidae Pit vipers x x
Amphibia Unidentified amphibians x
Amphiuma sp. Salamanders x
Sirenidae Sirens x
Siren sp. Greater siren x
Plethodon glutinosus Northern Slimy Salamander x
Anura Frogs and toads x x
Rana sp. Frog x
Osteichthyes Unidentified bony fishes x
Accipenser sp. Sturgeons x x
Lepisosteus sp. Gar fish x x x x x
Amia calva Bowfin x x x x
Elops saurus Ladyfish x x x
Anguilla rostrata American eel x


Table 8. Comparative faunal collections from selected sites, continued.

9GN57 38JA61 Artesian 8DU5543 8DU12 8DU66 -
Well Midden
Late Archaic Early Swift Creek Swift Creek Mississippian Mississippian
Taxon Common Name Marrinan Reitz et al. Quitmyer deFrance Units 1-4 Wing 1963
1975 1987 1985 1993
Clupeidae Herrings x x x x
Brevoortia sp. Menhaden x x x
Opisthonema oligmum Atlantic thread herring x
Sardinella sp. Sardines x
Esox sp. Pickerel x
Siluriformes Catfishes x x x x
Ictauridae Freshwater catfish x x
Ictalurus sp. Bullhead catfish x x x
Ariidae Marine catfishes x x x x x
Arius felis Hardhead catfish x x x x x x
Bagre marinus Gafltopsail catfish x x x x x x
Opsanus sp. Toadfish x x
Fundulus sp. Killifish or topminnow x x
Belonidae Needlefish x
Atherinidae Silversides x
Serranidae Sea basses x
Centrarchidae Sunfishes x
Lepomis sp. Sunfishes x
Micropterus sp. Bass x x x
Pomoxis sp. Crappie
Perca flavescens White perch
Lobotes surinamensis Tripletail x
Pomatomus saltatrix Bluefish x x x x
Chloroscombrus chrysurus Atlantic bumper x
Carangidae Jacks x x x
Caranx sp. Jacks x
Caranx hippos Jack crevalle x
Selene sp. Lookdown x
Anisotremus sp. Pigfish x
Sparidae Porgies x x x
Lagodon sp. Pinfish x x
Archosargus probatocephalus Sheepshead x x x x x
Sciaenidae Drums x x x
Bairdiella chrysoura Silver perch x x x x
Cynoscionsp. Seatrout x x x x x
Leistomus xanthurus Spot x x x
Menticinhus americanus Southem kingfish x x
Menticirrhus littoralis Gulf kingfish x
Menticirrhus saxatilis Northern kingfish x
Micropogonias undulatus Atlantic croaker x x x x
Pogonias cromis Black drum x x x x
Sciaenops ocellatus Redfish x x x x x x
Stellifer sp. Star drum x x x
Mugilsp. Mullet x x x x x
Peprilus alepidotus Butterfish x x
Sphyraenidae Barracudas x x




Table 8. Comparative faunal collections from selected sites, continued.

9GN57 38JA61 Artesian 8DU5543 8DU12 8DU66-
Well Midden
Late Archaic Early Swift Creek Swift Creek Mississippian Mississippian
Taxon Common Name Marrinan Reitz et al. Quitmyer deFrance Units 1-4 Wing 1963
1975 1987 1985 1993
Prionotus sp. Sea robin x
Gobionellus sp. Gobies x x
Trichiurus lepturus Atlantic cutlassfish x
Paralichthys sp. Flounder x x x x x
Tetraodontidae Puffers and burrfish x
Chondrichthyes Sharks, rays, and x
Carcharhinidae Requiem shark family x
Carcharinus sp. Requiem sharks x
Carcharhinus altimus Bignose shark x
Carcharhinus leucas Bull shark x
Carcharhinus limbatus Blacktip shark x
Galeocerdo cuvieri Tiger shark x x x
Negaprion brevirostris Lemon shark x
Sphyma sp. Hammerhead shark x
Rajiformes Rays x x x
Dasyatis sp. Stingray x
Rhinobatidae Guitarfishes x
Aetobatis nanari Spotted eagle ray x
Callinectes sapidus Blue crab x x x

expectations" for feasting, elite private consumption, and elite
animal products use. Table 9 contrasts the Jackson and Scott
criteria with the Units 1-4 data (Appendix 1).
Based on these criteria, the Kinzey Knoll data do not point
clearly to feasting nor to elite private consumption or elite
animal products use. These data also suggest that models for
identifying feasting, restricted consumption, or reserved use of
particular animals in the coastal zone must incorporate the
greater species diversity that is the environmental norm.

Summary and Conclusions

The Units 1-4 assemblage represents a dense, unusually
heavy accumulation of invertebrate and vertebrate faunal
remains occurring as a discrete deposit approximately 30x25
m in size. During excavation, Ashley (2003) noted
stratigraphically discrete, heavy concentrations of vertebrate
fauna within the invertebrate deposit. The radiocarbon dates
suggest that approximately 100 years passed during which
deposition occurred. Even with this relatively short time
period, our ability to discuss this fauna as the result of several
events or numerous events, is constrained by our inability to
discriminate time in brief episodes.
The invertebrate and vertebrate fauna are all locally or
seasonally available. In the Kinzey's Knoll Units 1-4 excava-
tion area, their deposition is dense. Exotic lithic and metal

artifacts also were recovered and over 350 ceramic vessels are
represented by rim sherd counts from an area of approximately
2.5 cubic meters (Rolland 2004).
A number of scenarios can be proposed to account for these
depositional conditions:
1. The Shields Mound midden assemblages may be purely
domestic in origin and represent occasional or sporadic use of
the area for dumping of refuse by several groups through time.
Groups may have been of various sizes, from labor engaged in
mound-related construction or families moving through the
environment in a more migratory adaptation. Although the
fauna might support such an interpretation alone, the presence
of large quantities of ceramic vessels and exotic cultural
materials suggests that this is a less tenable proposition.
2. The Shields Mound assemblage may result from the
domestic activities of a resident group over the century of
deposition. If, as Moore and Ashley speculate, the mound
served as a charnel structure or for charnel activities, this
refuse might represent the family, lineage, caste, or captives
who cared for the dead. Sears (1982:186-189, 195) has
suggested such a situation at Fort Center for Mound A midden
deposits located near the charnel pond.
3. Shields Mound may have been a substructural mound
and the midden resultant of intentional discrete discard of the
refuse from an elite family or series of families who lived on
its summit. Knight (2004) has proposed this scenario for

2005 VOL. 58(3-4)


Table 9. Archaeological expectations for feasting, elite private consumption, and elite animal products use (after Jackson
and Scott 1995:107). Shields Mound values for Units 1-4 are in italics.

Evidence for Feasting Elite Private Consumption Elite Animal Products Use

Low sample diversity High sample diversity Furbearers
Moderate sample diversity Moderate sample diversity Potential fur bearers are present in
low numbers

High meat yields Prime cuts Dangerous taxa
High estimated biomassfrom Difficult to confirm since most of Bears are present
shellfish, the
bony fishes, and mammals biomass was generated by various

Bulk meat cuts Rare taxa Rare taxa
Some bulk meat cuts (deer) There are low numbers of There are low number of mammalian
mammalian species

Bulk cooking Larger than average quantity of bird Colorful plumage bearers
Quantities suggest bulk cooking bones Most are black, white, or cryptic in
Avian remains prominent in coloration

Little butchering debris Low butchering debris High non-utilitarian bone modification
Low numbers of butchered bone Low numbers of butchered bones Highest density of bone pin fragments,
plain and decorated

Mound Q at Moundville.
4. The Shields Mound samples may result from feasting
activities that amassed large quantities of food. The vertebrate
and invertebrate fauna present are all locally or seasonally
available. Migratory waterfowl, anadromous and catadromous
fishes indicate that at least some activities occurred between
November and March. Seasonally-available plants with new
shoots, fruits, and nuts tend to have completed their reproduc-
tive cycle by November. Thus the cold months required
dependence on stored plant materials and perhaps meat
preserved by smoking or salting. It is likely that hunting and
fishing occupied more of the subsistence effort during this
period than plant gathering.
At this time, proposition 4 seems most supportable given
the faunal, ceramic, lithic, and metals evidence. In this
midden accumulation, there is clear indication of aggregation
during the cold months. Ashley (2003:113) has demonstrated
evidence of contact with the Lower Ocmulgee River valley.
Thus these deposits may have resulted from events related to
trading or other activities that required feasting. If feasting of
this sort was a cold weather activity, traded commodities
acquired or processed during the annual cycle (e.g., marine
shells, shark teeth, cassina leaves, or other locally-produced
items) would be available for trade. It is also possible that

these deposits contain the discard from different kinds of ritual
events over a century.
Given the evidence of subsistence practices in the coastal
zone since the Late Archaic Period, I contend that feasting
required large quantities of "proper" foods prepared in
prescribed ways. In the Shields Mound assemblages, there is
no significant change in "staple" species from lowest levels to
highest levels, just fluctuations in the quantity of deposition.
There is an absence in the species lists of animals that figure
in Mississippian iconography and very few animals that have
"anomalous" or "atypical" behaviors in the Muskhogean
worldview. Association with the Under World and disorder or
impurity may account for the absence of snakes, sirens, and
alligators. Given the longstanding dominance of bony fish in
the Shields Mound assemblages and in other coastal zone
middens, their appropriateness for consumption had been
culturally negotiated. The absence of dog remains is interest-
ing, and it is possible that the consumption of dog meat was
itself inappropriate for the kinds of activities performed or for
individuals occupying the mound area. Among the Natchez,
for example, dog meat was eaten preparatory to embarking on
a war party (Swanton 1979:699). The opossum, a rather
atypical mammal that is commonly a constituent of Southeast-
ern prehistoric diet, also may have been improper in this




To maintain harmony and balance among a constellation
of competing forces, the requirements of daily and ceremonial
food consumption required strict adherence to prescribed
behaviors and relationships that insured safety and survival.
Without the appropriate performance of these observances, the
world was fraught with danger. Given societal rules and
proscriptions, animal foods required appropriate rituals in
hunting, preparation, presentation, consumption, and discard.
Traditional practices, evolved over millennia, provided the
balanced course along a narrow path between purity and
pollution. In the prehistoric context, feasting might require
only more of the foods that were "proper" since this adherence
guaranteed a level of safety in a world that required constant


'Vicki Rolland (personal communication, 2004) ran ten tests on St.
Johns Period midden oyster shell and determined that the average
weight of one liter of oyster shell was 635.5 gm. The number of
liters of shell reported for each excavation unit was reduced by ten
percent, then multiplied by Rolland's average value to determine the
number of grams represented. This value was entered into an
Excel-based spreadsheet (Hale and Marrinan 1987; biomass values
in Table 2) to calculate biomass. Although a minimum number of
individuals (MNI) for oysters cannot be calculated, the biomass
values provide a relative measure of the importance of oysters.

2 The origins of the people who made St. Johns "chalky wares" is also
an issue of no small consequence. Tempered with the spicules of
freshwater sponges, St. Johns series ceramics were the dominant
wares from the midden deposits adjacent to Shields Mound. These
ceramics have a St. Johns River drainage and east coast distribution
and other occurrences in southern Florida. Outside Florida, the
distribution of spicule-tempered ceramics is sporadic. In the Western
Hemisphere, this type of tempering is known ethnographically from
Amazonia (Rolland and Bond 2003). Russo and Heide (2000:47)
report the earliest radiocarbon dates on sponge spicule-tempered
pottery in Florida (ca. 1475 B.C.) from the Joseph Reed shell ring in
Martin County. One implication of their findings may be that the
radiation of spicule-tempered pottery in Florida begins on the
southeast coast. Ashley's (2003:104) and Rolland's (2004) work in
Northeast Florida indicates that the people who constructed Shields
Mound made this kind of pottery in the Early Mississippian Period
but were later displaced to settlements further upstream (south).
James A. Ford (1969) argued that the origins of people responsi-
ble for the Late Archaic coastal shell ring sites along the southeastern
Atlantic coast lie in South America. His assertions were based on the
work of Reichel-Dolmatoff (1965, 1972) at shell ring sites such as
Puerto Hormiga, on the northern coast of Coclombia. Anna Roose-
velt's and colleagues' subsequent work in eastern Amazonia has
produced a suite of radiocarbon dates indicating that ceramic
technology was present in that area as early as 7500 B.P. (Roosevelt
1995:115; Table 10.1). Russo's and Heide's data may bear on this
question. Is it possible that we now have evidence of several waves
of migration from different points of origin (sensu Rouse 1992);
earlier waves to the Georgia and northern Florida coasts with fiber-
tempered pottery, to the South Carolina coast with gritty sand-
tempered pottery, and still later, a wave with sponge spicule-
tempered pottery to the southern Florida coast with radiation
northward and into the upper St. Johns River drainage? In this

scenario, the use of a Muskhogean-based model of human-animal
interactions may be inappropriate.


But for Vicki Rolland, I might never have been drawn into this
project. She proposed using the fauna for student projects in the
2001 Paleonutrition class at Florida State University. The students
who worked on field specimen lots from Shields Mound Units 1 and
2 that year were: Colleen Creamer, James S. Dunbar, Donna Freid,
Byran Harrell, William Morgan Hays, Rachel Horlings, Michelle
Hughes, Robert Koonce, Ellen (Triel) Lindstrom, Vicki Rolland,
Rachel Smith, Jules Straker, Brian Worthington, and Jessica Zimmer.
During the summer of 2001 and into 2002 and 2003, Vicki and I
continued to work on Units 3 and 4, then 6, then 7. In 2003, class
members again worked with Shields Mound fauna. Claire Nanfro,
Margaret Sher, Justin Stickler, and I completed work on Unit 8 and
two of the Reeves Rise Shovel Tests. I completed the Unit 5
assemblage and the third Reeves Rise shovel test in 2004.
Keith Ashley patiently answered all questions put to him andjust
as patiently waited for this paper to be written. I want to thank him
and tell him how much I believe his pursuit of the archaeology of
Northeast Florida and critical thinking will mean to the discipline.
We could not have completed this work without the assistance of
a number of colleagues at the Florida Museum of Natural History.
This project began under the curatorial oversight of Dr. Elizabeth S.
Wing and extended into that of her successor, Dr. Kitty Emery. I am
grateful to both for their assistance and access. The Environmental
Archaeology staff has been superb in their assistance to visitors: the
collections are in excellent condition and their advice is always
helpful. I commend Sylvia Scudder, Irv Quitmyer, Donna Ruhl, and
Laura Kozuch. Dr. David Steadman, Curator of Ornithology, has
never wavered in assisting with identifications ofavian remains. Dr.
Fred Thompson, Curator ofMalacology, always has been gracious in
helping with invertebrate identifications. Since my student days in
the Zooarchaeology Range (now the Environmental Archaeology
Laboratory), I have been impressed by the dedication of the museum
curators and collections staff. The research environment and
collections are unparalleled; the expertise also.
This paper was originally entitled, "Vertebrate Fauna from
Shields Mound (8Dul2)," and presented in the symposium, "The
Shields Site: New Perspectives on the Early St. Johns II Culture of
Northeastern Florida," at the Annual Meeting of the Southeastern
Archaeological Conference, Chattanooga, Tennessee, November,

References Cited

Ashley, Keith H.
2003 Interaction, Population Movement, and Political Economy:
The Changing Social Landscape of Northeastern Florida
(A.D. 900-1500). Ph.D. dissertation, Department of
Anthropology, University of Florida, Gainesville.

Bennett, Charles E. (translator)
2001 Three Voyages: Rene Laudonniere. University of Alabama
Press, Tuscaloosa.

Boas, Franz
1966 Kwakiutl Ethnography, edited by Helen Codere. Univer-
sity of Chicago Press, Chicago.

Bogan, Arthur E.
1983 Evidence for Faunal Resource Partitioning in an Eastern


2005 VOL. 58(3-4)


North American Chiefdom. In Animals and Archaeology:
1. Hunters and their Prey, edited by Juliet Clutton-Brock
and Caroline Grigson, pp. 305-324. BAR International
Series 163, Oxford, England.

Brown, James
1989 On Style Divisions of the Southeastern Ceremonial Com-
plex: A Revisionist Perspective. In The Southeastern
Ceremonial Complex: Artifacts and Analysis, edited by
Patricia Galloway, pp. 183-204. University of Nebraska
Press, Lincoln.

Cruz-Urribe, Kathryn
1988 The Use and Meaning of Species Diversity and Richness in
Archaeological Faunas. Journal ofArchaeological Science

deFrance, Susan D.
1993 Faunal material from the Greenfield Site (8Du5543), Duval
County, Florida. Ms. on file, Environmental Archaeology
Laboratory, Florida Museum of Natural History,
Gainesville, Florida.

Dietler, Michael
1996 Feasts and Commensal Politics in the Political Economy:
Food, Power and Status in Prehistoric Europe. In Food and
the Status Quest: An Interdisciplinary Perspective, edited
by Polly Wiessner and Wulf Schiefenhovel, pp. 87-125.
Berghahn Books, Providence.

Dietler, Michael, and Brian Hayden
2001 Digesting the Feast Good to Eat, Good to Drink, Good to
Think: An Introduction. In Archaeological and
Ethnographic Perspectives on Food, Politics, and Power,
edited by Michael Dietler and Brian Hayden, pp. 1-20.
Smithsonian Institution Press, Washington, D.C.

Douglas, Mary
1975 Deciphering a Meal. In Food and Culture: A Reader,
edited by Carole Counihan and Penny van Esterik, pp. 36-
54. Routledge, London.

Emerson, Thomas E.
1989 Water, Serpents, and the Underworld: An Exploration into
Cahokian Symbolism. In The Southeastern Ceremonial
Complex: Artifacts and Analysis, edited by Patricia
Galloway, pp. 45-92. University of Nebraska Press,

Ford, James A.
1969 A Comparison of Formative Cultures in the Americas:
Diffusion or the Psychic Unity of Man. Smithsonian
Contributions to Anthropology, Volume 11. Smithsonian
Institution Press, Washington, D. C.

Gilbert, Carter R.
1989 Species Profiles: Life Histories aand Environmental
Requirements of Coastal Fishes and Invertebrates (Mid-
Atlantic) --Atlantic andShortnosedSturgeons. Biological
Report 82(11.122), Fish and Wildlife Service, United
States Department of the Interior, Lafayette, Louisiana.

Goody, Jack

1982 Cooking, Cuisine and Class: A Study in Comparative
Sociology. Cambridge University Press, Cambridge.

Grayson, Donald K.
1984 Quantitative Zooarchaeology. Academic Press, New York.

Hale, H. Stephen
1989 Diversity and Equitability Program in GW Basic.
Zooarchaeology Lab, Florida State University, Tallahassee.

Hale, H. Stephen, and Rochelle A. Marrinan
1987 Fort Matanzas Spreadsheet for Lotus and Excel.
Zooarchaeology Lab, Florida State University, Tallahassee.

Hallowell, A. Irving
1926 Bear Ceremonialism in the Northern Hemisphere. Ameri-
can Anthropologist 28(1):1-175.

Harris, Marvin
1985 Good to Eat: Riddles of Food and Culture. Simon and
Schuster, New York.

Hayden, Brian
2001 Fabulous Feasts: A Prolegomenon to the Importance of
Feasting. InArchaeological andEthnographic Perspectives
on Food, Politics, and Power, edited by Michael Dietler
and Brian Hayden, pp. 23-64. Smithsonian Institution
Press, Washington, D.C.

Hulton, Paul
1977 The Work of Jacques LeMoyne de Morgues: A Hugenot
Artist in France, Florida, and England, 2 volumes. British
Museum Publications Limited, London.

Hudson, Charles
1976 The Southeastern Indians. University of Tennessee Press,

Jackson, H. Edwin, and Susan L. Scott
1995 The Faunal Record of the Southeastern Elite: The Implica-
tions of Economy, Social Relations, and Ideology. South-
eastern Archaeology 14(2): 103-119.

2003 Patterns of Elite Faunal Utilization at Moundville, Ala-
bama. American Antiquity 68(3): 552-572.

Jordan, Douglas F.
1963 The Goodman Mound. In Papers on the Jungerman and
Goodman Sites, Florida, by Douglas F. Jordan, Elizabeth
S. Wing, and Adelaide K. Bullen. Contributions of the
Florida State Museum, Social Sciences 10: 24-49. Univer-
sity of Florida, Gainesville.

Kelly, Lucretia S.
2001 A Case of Ritual Feasting at the Cahokia site. In Feasts:
Archaeological and Ethnographic Perspectives on Food,
Politics, andPower, edited by Michael Dietler and Brian
Hayden, pp. 334-367. Smithsonian Institution Press,
Washington, D.C.

Knight, Vernon James, Jr.
1986 The Institutional Organization of Mississippian Religion.
American Antiquity 51(4):675-687.




2001 Feasting and the Emergence of Platform Mound
CeremonialisminEastemNorthAmerica. InArchaeologi-
cal and Ethnographic Perspectives on Food, Politics, and
Power, edited by Michael Dietler and Brian Hayden, pp.
311-333. Smithsonian Institution Press, Washington, D.C.

2004 Characterizing Elite Midden Deposits at Moundville.
American Antiquity 69(2):304-321.

Lee, Chung Ho, Irvy R. Quitmyer, Christopher T. Espenshade, and
Robert E. Johnson
1984 Estuarine Adaptations During the Late Prehistoric Period:
Archaeology of Two Shell Midden Sites on the St. Johns
River. Office of Cultural and Archaeological Research,
Report of Investigations Number 5. University of West
Florida, Pensacola.

Lepionka, Larry, Donald Colquhoun, Rochelle Marrinan, David
McCollum, Mark Brooks, John Foss, William Abbott, and Ramona
1983 The Second Refuge Site: Location 22 (38Ja61), Savannah
National Wildlife Refuge, Jasper County, South Carolina.
Interagency Archaeological Service, Atlanta, Georgia.

Marrinan, Rochelle A.
1975 Molluscs, Ceramics, and Sedentism: The Late Archaic on
the Georgia Coast. Ph.D. dissertation, Department of
Anthropology, University of Florida, Gainesville.

McLane, William McNair
1955 The Fishes of the St. Johns River System. Ph.D. disserta-
tion, Department of Zoology, University of Florida,

Mintz, Sidney W.
1997 Time, Sugar, and Sweetness. In Food and Culture: A
Reader, edited by Carole Counihan and Penny van Esterik,
pp. 357-369. Routledge, New York.

Moore, Clarence Bloomfield
1894 Certain Sand Mounds of the St. Johns River, Florida, Part
II. Journal of the Academy ofNatural Sciences ofPhiladel-
phia, Second Series 10:129-246.

1895 Certain Sand Mounds ofDuval County, Florida. Journal of
the Academy of Natural Sciences ofPhiladelphia, Second
Series 10:449-502.

Pauketat, Timothy R., Lucretia S. Kelly, Gayle J. Fritz, Neal H.
Lopinot, Scott Elias, and Eve Hargrave
2002 The Residues of Feasting and Public Ritual at Early
Cahokia. American Antiquity 67(2): 257-279.

Powdermaker, Hortense
1932 Feasts in New Ireland: The Social Function of Eating.
American Anthropologist 34(2): 236-247.

Quitmyer, Irvy R.
1985 Aboriginal Subsistence Activities in the Kings Bay Local-
ity. InAboriginalSubsistence andSettlementArchaeology
of the Kings Bay Locality, edited by William Hampton
Adams, pp. 73-91. University of Florida, Department of
Anthropology, Reports of Investigations, 2.

Reichel-Dolmatoff, Gerardo
1965 Excavaciones Arquelogicas en Puerto Hormiga,
Department de Bolivar. Antropologia, Vol. 2, Bogota.

1972 The Cultural Context of Early Fiber-Tempered Pottery in
Northern Colombia. TheFloridaAnthropologist25(2-2):1-

Reitz, Elizabeth J.
1982 Vertebrate Fauna from Four Coastal Mississippian Sites.
Journal ofEthnobiology 2(1): 39-61.

Reitz, Elizabeth J., Rochelle A. Marrinan, and Susan L. Scott
1987a Survey of Vertebrate Remains from Prehistoric Sites in the
Savannah River Valley. Journal ofEthnobiology 7(2):195-

Reitz, Elizabeth J., Irvy R. Quitmyer, H. Stephen Hale, Sylvia J.
Scudder, and Elizabeth S. Wing
1987b Application of Allometry to Zooarchaeology. American
Antiquity 52(2):304-317.

Reitz, Elizabeth J. and Elizabeth S. Wing
1999 Zooarchaeology. Cambridge Manuals in Archaeology,
Cambridge University Press, Cambridge.

Rolland, Vicki L.
2004 Measuring Tradition and Variation: A St. Johns II Pottery
Assemblage from the Shields Site (8DU12). Unpublished
Master's thesis, Department ofAnthropology, Florida State
University, Tallahassee.

Roosevelt, Anna C.
1995 Early Pottery in the Amazon: Twenty Years of Scholarly
Obscurity. In The Emergence ofPottery: Technology and
Innovation in Ancient Societies, edited by William K.
Barnett and John W. Hoopes, pp. 115-131. Smithsonian
Institution Press, Washington, D.C.

Rolland, Vicki L., and Paulette Bond
2003 The Search for Spiculate Clays near Aboriginal Sites in the
Lower St. Johns River Region, Florida. The Florida
Anthropologist 56(2):91-111.

Rouse, Irving B.
1992 The Tainos: Rise and Decline of the People who Greeted
Columbus. Yale University Press, New Haven.

Russo, Michael, and Gregory Heide
2000 The Joseph Reed Shell Ring. Southeast Archeological
Center, National Park Service, Tallahassee, Florida.

Sears, William H.
1957 Excavations on Lower St. Johns River, Florida. Contribu-
tions of the Florida State Museum, Social Sciences, 2.
University of Florida, Gainesville.

1982 Fort Center: An Archaeological Site in the Lake
Okeechobee Basin. University Presses of Florida,

Shannon, Claude E., and Warren Weaver
1949 TheMathematical Theory ofCommunication. University of


2005 VOL. 58(3-4)

Illinois, Urbana.

Sheldon, A.L.
1969 Equitability Indices: Dependence on the Species Count.
Ecology 50:466-467.

Strong, John A.
1989 The Mississippian Bird-Man Theme in Cross-Cultural
Perspective. In The Southeastern Ceremonial Complex:
Artifacts and Analysis, edited by Patricia Galloway, pp.
211-238. University of Nebraska Press, Lincoln.

Swanton, John R.
1979 The Indians oftheSoutheastern UnitedStates. Smithsonian
Institution, Bureau of American Ethnology, Bulletin 137.
Washington, D.C. First Published in 1946.

2000 (1928) Creek Religion and Medicine. University of Ne-
braska Press, Lincoln.

Tagatz, Marlin
1965 The FisheryforBlue Crabs in the St. Johns River, Florida,
with Special Reference to Fluctuation in Yield between
1961 and 1962. United States Fish and Wildlife Service,
Special Scientific Report-Fisheries No. 501. Washington,

VanDerwarker, Amber M.
1999 Feasting and Status at the Toqua Site. Southeastern
Archaeology 18(1):24-34.

Wheeler, Ryan J., James J. Miller, Ray M. McGee, Donna Ruhl,
Brenda Swann, and Melissa Memory
2003 Archaic Period Canoes from Newnan's Lake, Florida.
American Antiquity 68(3):533-551.

Wing, Elizabeth S.
1963 Vertebrates from the Jungerman and Goodman sites near
the east coast of Florida. In Papers on the Jungerman and
Goodman Sites, Florida, by Douglas F. Jordan, Elizabeth
S. Wing, and Adelaide K. Bullen. Contributions of the
Florida State Museum, Social Sciences 10: 51-60. Univer-
sity of Florida, Gainesville.

1977 Subsistence systems in the Southeast. The Florida Anthro-
pologist 30(2):81-87.



Appendix 1. Kinzey's Knoll Units 1-4 composite faunal remains, 1/4-in sample.

Scientific Name Taxonomic Name NISP % Weight % Biomass % Burnt % Worked % MNI %
Mammalia, Large Probably deer or bear 378 1.45 546.5 6.21 7653.0 8.96 58 19.73 6 7.32 0.00
Mammalia, Medium Probably raccoon, bobcat, or 66 0.25 33.6 0.38 621.9 0.73 3 1.02 0.00 0.00
Mammalia, Small Probably rabbits and squirrels 55 0.21 12.7 0.14 259.1 0.30 2 0.68 0.00 0.00
Mammalia Unidentified mammals 61 0.23 25.9 0.29 492.0 0.58 7 2.38 0.00 0.00
Mammalia cf., Procyon Probably raccoon 1 0.00 1.8 0.02 44.6 0.05 1 0.34 0.00 0.00
Mammalia cf., Ursus Probably bear 1 0.00 2.3 0.03 55.7 0.07 0.00 0.00 0.00
Didelphis virginiana Eastern opossum 1 0.00 1.1 0.01 28.7 0.03 1 0.34 0.00 1 0.23
Soricidae Shrews 1 0.00 0.1 0.00 3.3 0.00 0.00 0.00 1 0.23
Scalopus aquaticus Eastern mole 1 0.00 0.5 0.01 14.1 0.02 0.00 0.00 1 0.23
Sylvilagus sp. Rabbits 25 0.10 18.3 0.21 359.9 0.42 0.00 0.00 6 1.38
Rodentia Rodents 30 0.11 5.1 0.06 114.0 0.13 0.00 0.00 0.00
Sciurus sp. Squirrels 2 0.01 0.4 0.00 11.5 0.01 0.00 1 1.22 0.00
Sciurus cf., carolinensis Probably grey squirrel 2 0.01 0.7 0.01 19.1 0.02 0.00 0.00 0.00
Sciurus carolinensis Grey squirrel 15 0.06 6.2 0.07 135.9 0.16 0.00 0.00 2 0.46
Sciurus niger Fox squirrel 57 0.22 43.6 0.50 786.2 0.92 2 0.68 0.00 6 1.38
Sigmodon hispidus Hispid cotton rat 31 0.12 6.1 0.07 133.9 0.16 0.00 0.00 5 1.15
Delphinidae Dolphins 3 0.01 3.1 0.04 72.8 0.09 0.00 3 3.66 0.00
Tursiops truncatus Bottle-nose dolphin 1 0.00 7.8 0.09 167.1 0.20 0.00 0.00 1 0.23
Urocyon einereoargenteus Grey fox 2 0.01 0.7 0.01 19.1 0.02 0.00 0.00 1 0.23
Ursus sp. Bears 23 0.09 101.3 1.15 1679.0 1.97 5 1.70 2 2.44 3 0.69
Procyonlotor Northernraccoon 21 0.08 42.5 0.48 768.3 0.90 0.00 0.00 2 0.46
Lynxrufus Bobcat 3 0.01 6.4 0.07 139.8 0.16 0.00 0.00 2 0.46
Artiodactyla Probably deer 157 0.60 280.0 3.18 4192.2 4.91 8 2.72 0.00 0.00
Odocoileusvirginianus White tail deer 155 0.59 1032.0 11.72 13561.6 15.88 3 1.02 5 6.10 6 1.38
All Mammals 1092 4.18 2178.7 24.74 31332.7 36.68 90 30.6 17 20.73 37 8.51

Aves All Bird fragments 316 1.21 158.9 1.80 2056.0 2.41 12 4.08 0.00 0.00
Podiceps auritus Homed grebe 4 0.02 1.9 0.02 36.6 0.04 0.00 0.00 1 0.23
Podilymbus podiceps Pied-billed grebe 2 0.01 1.0 0.01 20.4 0.02 0.00 0.00 1 0.23
Phalacrocorax auritus Double-crested cormorant 2 0.01 2.3 0.03 43.6 0.05 0.00 0.00 1 0.23

Appendix 1. Kinzey's Knoll Units 1-4 composite faunal remains, 1/4-in sample, continued.

Scientific Name Taxonomic Name NISP % Weight % Biomass % Burnt % Worked % MNI %
Phalacrocorax carbo European cormorant 8 0.03 11.0 0.12 181.0 0.21 0.00 0.00 1 0.23
Ardeidae Herons 1 0.00 1.0 0.01 20.4 0.02 0.00 0.00 1 0.23
Ardeaalba Great egret 1 0.00 0.4 0.00 8.9 0.01 0.00 0.00 1 0.23
Anatidae Ducks, geese, and swans 2 0.01 1.2 0.01 24.1 0.03 0.00 0.00 0.00
Anatidae Unidentified small duck 6 0.02 8.1 0.09 137.0 0.16 0.00 0.00 3 0.69
Anatidae Unidentified medium-sized 21 0.08 17.1 0.19 270.4 0.32 0.00 0.00 4 0.92
Anatidae Unidentified large duck 10 0.04 16.8 0.19 266.1 0.31 0.00 0.00 2 0.46
Anatidae Unidentified goose 20 0.08 46.2 0.52 668.1 0.78 0.00 0.00 3 0.69
Anatidae cf., Mergus Probably amerganser 2 0.01 1.2 0.01 24.1 0.03 0.00 0.00 0.00
Meleagris gallopavo Wild turkey 10 0.04 38.9 0.44 571.3 0.67 0.00 0.00 2 0.46
Larus sp. Gulls 2 0.01 0.9 0.01 18.6 0.02 0.00 0.00 0.00
Larus argentatus Herring gull 12 0.05 9.2 0.10 153.8 0.18 0.00 0.00 2 0.46
Larus delawarensis Ring-billed gull 4 0.02 1.2 0.01 24.1 0.03 0.00 0.00 2 0.46
Colinus virginiana Eastern bob white quail 1 0.00 0.1 0.00 2.5 0.00 0.00 0.00 1 0.23
Zenaida macroura Mourning dove 3 0.01 0.3 0.00 6.8 0.01 0.00 0.00 1 0.23
Passeriformes Song birds 1 0.00 0.1 0.00 2.5 0.00 0.00 0.00 1 0.23
Corvus ossifragus Fish crow 1 0.00 0.1 0.00 2.5 0.00 0.00 0.00 1 0.23
All Birds 429 1.64 317.9 3.61 4538.8 5.31 12 4.08 0 0.00 28 6.44

Alligator mississippiensis American alligator 12 0.05 33.8 0.38 398.8 0.47 0.00 0.00 2 0.46
Testudines All Turtle fragments 697 2.67 364.2 4.14 1644.7 1.93 36 12.24 1 1.22 0.00
Kinostemidae Mud or musk turtle 11 0.04 5.0 0.06 93.0 0.11 0.00 0.00 2 0.46
Terrapene carolina Eastern box turtle 31 0.12 27.5 0.31 291.3 0.34 0.00 3 3.66 2 0.46
Malaclemysterrapin Diamondbackterrapin 3 0.01 7.6 0.09 123.1 0.14 0.00 0.00 1 0.23
Trachemys sp. Cooters and sliders 4 0.02 7.8 0.09 125.2 0.15 1 0.34 0.00 2 0.46
Dierochelys reticularia Chicken turtle 56 0.21 37.1 0.42 356.0 0.42 1 0.34 0.00 2 0.46
Gopherus polyphemus Gophertortoise 288 1.10 613.2 6.96 2331.8 2.73 1 0.34 2 2.44 10 2.30
Apalone ferox Soft-shelled turtle 18 0.07 55.4 0.63 465.7 0.55 1 0.34 0.00 1 0.23
All turtles 1108 4.24 1117.8 12.70 5430.9 6.36 40 13.6 6 7.32 20 4.60

Appendix 1. Kinzey's Knoll Units 1-4 composite faunal remains, 1/4-in sample, continued.

Scientific Name Taxonomic Name NISP % Weight % Biomass % Burnt % Worked % MNI %
Colubridae Non-poisonous snakes 1 0.00 0.6 0.01 8.2 0.01 0.00 0.00 0.00
Farancia sp. Mud snake 4 0.02 0.3 0.00 4.1 0.00 0.00 0.00 1 0.23
All snakes 5 0.02 0.9 0.01 12.3 0.01 0.00 0 0.00 1 0.23

Anura Frogs and toads 2 0.01 0.2 0.00 2.7 0.00 0.00 0.00 0.00
Anura cf., Ranidae Probably frog 1 0.00 0.1 0.00 1.3 0.00 0.00 0.00 1 0.23
Siren lacertina Greater siren 1 0.00 0.2 0.00 2.7 0.00 0.00 0.00 1 0.23
All Amphibians 4 0.02 0.5 0.01 6.8 0.01 0.00 0 0.00 2 0.46
Osteichthyes All Bony Fish fragments 15172 58.12 2959.9 33.62 19130.9 22.40 43 14.63 1 1.22 0.00
Accipenser sp. Sturgeon 5 0.02 2.1 0.02 53.8 0.06 0.00 0.00 1 0.23
Lepisosteus spp. Gar 284 1.09 168.1 1.91 1873.7 2.19 7 2.38 1 1.22 5 1.15
Amia calva Bowfin 111 0.43 17.8 0.20 304.0 0.36 1 0.34 0.00 3 0.69
Elops saurus Ladyfish 14 0.05 1.2 0.01 34.2 0.04 0.00 0.00 1 0.23
Anguillarostrata Freshwater eel 138 0.53 7.9 0.09 157.4 0.18 0.00 0.00 4 0.92
Clupeidae Herrings 100 0.38 5.6 0.06 119.1 0.14 0.00 0.00 4 0.92
Esoxsp. Pikes 11 0.04 2.8 0.03 68.0 0.08 0.00 0.00 2 0.46
Silurifonnes All Catfish 433 1.66 81.1 0.92 1298.9 1.52 4 1.36 0.00 0.00
Ictaluridae Freshwater catfishes 140 0.54 65.7 0.75 1063.4 1.24 4 1.36 0.00 24 5.52
Ariidae Marine catfishes 89 0.34 25.3 0.29 429.5 0.50 4 1.36 0.00 13 2.99
Arius felis Hardhead catfish 33 0.13 10.9 0.12 193.0 0.23 1 0.34 0.00 0.00
Bagre marinus Gafftopsail catfish 6 0.02 4.0 0.05 74.5 0.09 1 0.34 0.00 0.00
Lepomissp. Sunfishes 3 0.01 0.6 0.01 11.3 0.01 0.00 0.00 0.00
Lepomis microlophus Redearsunfish 5 0.02 1.3 0.01 21.7 0.03 0.00 0.00 3 0.69
Micropterus sp. Bass 16 0.06 5.2 0.06 69.4 0.08 0.00 0.00 6 1.38
Pomatomus saltatrix Bluefish 2 0.01 0.3 0.00 11.1 0.01 0.00 0.00 1 0.23
Caranxsp. Jacks 104 0.40 40.1 0.46 1001.8 1.17 3 1.02 0.00 4 0.92
Lobotes surinamensis Tripletail 5 0.02 17.5 0.20 299.8 0.35 0.00 0.00 1 0.23
Sparidae Porgies 10 0.04 3.1 0.04 44.9 0.05 0.00 0.00 0.00
Sparidae/Sciaenidae Probably Porgy 109 0.42 4.8 0.05 67.1 0.08 0.00 0.00 0.00
Archosargusprobatocephalus Sheepshead 688 2.64 366.3 4.16 3620.2 4.24 4 1.36 0.00 42 9.66
Sciaenidae Drums 30 0.11 9.7 0.11 209.0 0.24 0.00 0.00 0.00

Appendix 1. Kinzey's Knoll Units 1-4 composite faunal remains, 1/4-in sample, continued.

Scientific Name Taxonomic Name NISP % Weight % Biomass % Burnt % Worked % MNI %
Bardiellasp. Perch 2 0.01 0.1 0.00 7.1 0.01 0.00 0.00 1 0.23
Cynoscion spp. Seatrout 1053 4.03 156.9 1.78 1639.7 1.92 11 3.74 0.00 63 14.48
Micropogonias undulatus Atlantic croaker 37 0.14 9.6 0.11 207.4 0.24 1 0.34 0.00 13 2.99
Pogonias cromis Black drum 215 0.82 241.0 2.74 2252.6 2.64 1 0.34 0.00 27 6.21
Sciaenops ocellatus Redfish 98 0.38 60.1 0.68 806.1 0.94 1 0.34 0.00 16 3.68
Stelliferlanceolatus Star drum 1 0.00 0.1 0.00 7.1 0.01 0.00 0.00 1 0.23
Mugilspp. Mullet 1748 6.70 199.5 2.27 2152.6 2.52 5 1.70 0.00 61 14.02
Sphyraena barracuda Great barracuda 1 0.00 1.1 0.01 31.9 0.04 0.00 0.00 1 0.23
Gobionellus sp. Gobies 78 0.30 6.9 0.08 141.1 0.17 0.00 0.00 3 0.69
Bothidae Flounder family 1768 6.77 326.5 3.71 4543.1 5.32 11 3.74 0.00 38 8.74
All bony fishes 22509 86.22 4803.1 54.55 41945.3 49.11 102 34.69 2 2.44 338 77.70
Squaliformes All unidentified shark 3 0.01 2.0 0.02 228.5 0.27 0.00 3 3.66 0.00
Carcharhinidae Requiem shark family 1 0.00 0.1 0.00 17.4 0.02 0.00 0.00 0 0.00
Carcharhinus sp. Requiem sharks 2 0.01 0.2 0.00 31.5 0.04 0.00 2 2.44 0 0.00
Carcharhinus altimus Bignose shark 1 0.00 0.2 0.00 31.5 0.04 0.00 1 1.22 1 0.23
Carcharhinus leucas Bull shark 5 0.02 1.8 0.02 208.7 0.24 0.00 3 3.66 1 0.23
Carcharhinus limbatus Blacktip shark 1 0.00 0.4 0.00 57.2 0.07 0.00 1 1.22 1 0.23
Galeocerdo cuvieri Tiger shark 4 0.02 2.2 0.02 248.0 0.29 0.00 0.00 1 0.23
Negaprion brevirostris Lemon shark 2 0.01 0.9 0.01 115.0 0.13 0.00 2 2.44 1 0.23
Rajiformes Rays 19 0.07 7.2 0.08 687.6 0.80 0.00 0.00 1 0.23
Rhinobatidae Guitarfishes 4 0.02 1.0 0.01 125.9 0.15 0.00 0.00 1 0.23
All sharks and rays 42 0.16 16.0 0.18 1751.4 2.05 0 0.00 12 14.63 7 1.61

Unidentified Vertebrate All Unidentified Fragments 904 3.46 336.1 3.82 0.00 50 17.01 45 54.88 0.00
Totals 26105 100.00 8804.8 100.00 85417.0 100.00 294 100.0 82 100.00 435 100.00

Decapoda Crabs, probably blue crabs 66 100.00 24.9 100.00 473.0 100.00 7 100.0 0 0.00 17 100.00

Crassostreavirginica Eastemoyster 2210014.8 100.00 297895.1 100.00

Appendix 2. Kinzey's Knoll West, Unit 5 composite 1/2-in sample, faunal remains.

Taxonomic Name Common Name NISP % Weight % Biomass % Burnt % MNI %
Mammalia, Large Probably deer or bear 31 8.76 17.8 27.81 351.0573 30.21 2 50.00 0.00
Mammalia, Small Probably rabbit or squirrel 3 0.85 0.3 0.47 8.9004 0.77 0.00 0.00
Rodentia Rodents 5 1.41 0.5 0.78 14.0953 1.21 0.00 0.00
Sigmodon hispidus Hispid cotton rat 1 0.28 0.2 0.31 6.1791 0.53 0.00 1 6.25
Sciurus niger Fox squirrel 1 0.28 0.5 0.78 14.0953 1.21 0.00 1 6.25
Odocoileusvirginianus White tail deer 1 0.28 8.9 13.91 188.1269 16.19 0.00 1 6.25
All Mammals 42 11.864 28.2 44.06 582.4543 50.12 2 50.00 3 18.75

Aves Unidentified birds 16 4.52 5.2 8.13 91.5299 7.88 0.00 1 6.25

Testudines Unidentified turtles 14 3.95 3.4 5.31 71.7945 6.18 0.00 1 6.25
Gopheruspolyphemus Gopher tortoise 4 1.13 2.2 3.44 53.6319 4.61 0.00 1 6.25
All turtles 18 5.08 5.6 8.75 125.4264 10.79 0 0.00 2 12.50

Serpentes Unidentified snakes 1 0.28 0.1 0.16 1.3490 0.12 0.00 1 6.25

Osteichthyes All Bony Fish fragments 99 27.97 4.8 7.50 105.1491 9.05 2 50.00 0.00
Clupeidae Herrings 19 5.37 0.2 0.31 8.0137 0.69 0.00 1 6.25
Siluriformes Catfishes 5 1.41 0.3 0.47 6.3572 0.55 0.00 1 6.25
Caranx sp. Probably jack crevelle 4 1.13 3.1 4.84 105.2927 9.06 0.00 1 6.25
Archosargus probatocephalus Sheepshead 8 2.26 1.3 2.03 20.1757 1.74 0.00 1 6.25
Sciaenidae Drums 2 0.56 0.3 0.47 15.9614 1.37 0.00 0.00
Cynoscion spp. Seatrout 6 1.69 0.8 1.25 32.9827 2.84 0.00 1 6.25
Pogonias cromis Black drum 1 0.28 0.1 0.16 7.0795 0.61 0.00 1 6.25
Mugilspp. Mullet 32 9.04 0.8 1.25 21.5170 1.85 0.00 1 6.25
Bothidae Flounder family 5 1.41 0.8 1.25 21.5650 1.86 0.00 1 6.25
All bony fishes 181 51.13 12.5 19.53 344.0939 29.61 2 50.00 8 50.00

Rajiformes Rays 1 0.28 0.1 0.16 17.3780 1.50 0.00 1 6.25

Unidentified Vertebrate All Unidentified Fragments 95 26.84 12.3 19.22 0.00 0.00 0.00
Totals 354 100.00 64.0 100.00 1162.2314 100.00 4 100.00 16 100.00

Crassostreavirginica Eastern oyster 42896.3 100.00 6507.9954 100.00

Appendix 3. Kinzey's Knoll South, Unit 6 composite 1/4-in sample, faunal remains.

Scientific Name Taxonomic Name NISP % Weight % Biomass % Burnt % Worked % MNI %
Mammalia, Large Probably deer or bear 34 0.72 34.8 3.82 641.8 5.45 4 9.09 1 7.69 0.00
Mammalia, Medium Probably raccoon, bobcat, or fox 4 0.09 1.1 0.12 28.7 0.24 0.00 0.00 0.00
Mammalia, Small Probably rabbits and squirrels 28 0.60 11.1 1.22 229.5 1.95 3 6.82 0.00 0.00
Mammalia Unidentified mammals 11 0.23 3.9 0.43 89.5 0.76 0.00 0.00 0.00
Scalopus aquaticus Eastern mole 1 0.02 0.2 0.02 6.2 0.05 0.00 0.00 1 1.12
Sylvilagus sp. Rabbits 2 0.04 0.6 0.07 16.6 0.14 0.00 0.00 1 1.12
Rodentia Rodents 3 0.06 0.5 0.05 14.1 0.12 0.00 0.00 0.00
Oryzomys sp. Rice rats 1 0.02 0.2 0.02 6.2 0.05 0.00 0.00 1 1.12
Sciurus carolinensis Grey squirrel 1 0.02 0.4 0.04 11.5 0.10 0.00 0.00 1 1.12
Sciurus niger Fox squirrel 19 0.40 16.1 1.77 320.7 2.72 0.00 0.00 2 2.25
Sigmodon hispidus Hispid cotton rat 1 0.02 0.2 0.02 6.2 0.05 0.00 0.00 1 1.12
Ursus sp. Bears 6 0.13 16.0 1.76 318.9 2.71 0.00 0.00 1 1.12
Procyon lotor Northern raccoon 3 0.06 3.3 0.36 77.0 0.65 0.00 0.00 1 1.12
Artiodactyla Probably deer 8 0.17 20.2 2.22 393.4 3.34 0.00 0.00 0.00
Odocoileus virginianus White tail deer 13 0.28 17.8 1.95 351.1 2.98 0.00 1 7.69 1 1.12
All Mammals 135 2.87 126.4 13.88 2511.4 21.32 7 15.91 2 15.38 10 11.24

Aves All Bird fragments 44 0.94 12.9 1.42 209.2 1.78 0.00 0.00 0.00
Anatidae Unidentified medium-sized duck 2 0.04 1.4 0.15 27.7 0.24 0.00 0.00 1 1.12
Anatidae Unidentified large duck 1 0.02 1.6 0.18 31.3 0.27 0.00 0.00 1 1.12
Meleagris gallopavo Wild turkey (probably male) 1 0.02 21.9 2.41 338.7 2.87 0.00 0.00 1 1.12
Larus sp. Probably L. argentatus female 1 0.02 0.5 0.05 10.9 0.09 0.00 0.00 1 1.12
Larus delawarensis Ring-billed gull 1 0.02 1.3 0.14 25.9 0.22 0.00 0.00 1 1.12
All Birds 50 1.06 39.6 4.35 643.8 5.46 0 0.00 0.00 5 5.62

Alligator mississippiensis American alligator 1 0.02 0.5 0.05 9.4 0.08 0.00 0.00 1 1.12

Testudines All Turtle fragments 143 3.04 59.8 6.57 490.2 4.16 7 15.91 0.00 0.00
Kinostemidae Mud or musk turtle 2 0.04 0.7 0.08 24.9 0.21 0.00 0.00 1 1.12
Terrapene carolina Eastern box turtle 7 0.15 9.3 1.02 140.9 1.20 0.00 0.00 1 1.12
Dierochelys reticularia Chicken turtle 2 0.04 4.7 0.52 89.2 0.76 1 2.27 0.00 1 1.12

Appendix 3. Kinzey's Knoll South, Unit 6 composite 1/4-in sample, faunal remains, continued.

Scientific Name Taxonomic Name NISP % Weight % Biomass % Burnt % Worked % MNI %
Gopherus polyphemus Gopher tortoise 14 0.30 30.0 3.29 308.8 2.62 1 2.27 0.00 2 2.25
Allturtles 168 3.57 104.5 11.48 1054.0 8.95 9 20.45 0 5 5.62

Serpentes Snakes 3 0.06 0.3 0.03 4.1 0.03 0.00 0.00 1 1.12

Caudata Probably Siren lacertina 2 0.04 0.3 0.03 4.1 0.03 0.00 0.00 1 1.12
Osteichthyes All Bony Fish fragments 3020 64.26 372.8 40.94 3571.9 30.32 13 29.55 0.00 0.00
Accipenser sp. Sturgeon 3 0.06 0.9 0.10 27.1 0.23 0.00 0.00 1 1.12
Lepisosteus spp. Gar 27 0.57 14.1 1.55 251.7 2.14 0.00 0.00 1 1.12
Amia calva Bowfin 9 0.19 1.5 0.16 41.0 0.35 0.00 0.00 1 1.12
Elops saurus Ladyfish 6 0.13 0.6 0.07 19.5 0.17 0.00 0.00 1 1.12
Anguilla rostrata Freshwater eel 39 0.83 2.4 0.26 60.0 0.51 0.00 0.00 2 2.25
Clupeidae Herings 50 1.06 2.6 0.29 64.0 0.54 0.00 0.00 2 2.25
Siluriformes All Catfish 157 3.34 31.7 3.48 532.1 4.52 0.00 0.00 0.00
Ictaluridae Freshwater catfishes 43 0.91 13.3 1.46 233.2 1.98 0.00 0.00 11 12.36
Ariidae Marine catfishes 4 0.09 0.8 0.09 16.1 0.14 0.00 0.00 0.00
Arius felis Hardhead catfish 2 0.04 0.6 0.07 12.3 0.10 0.00 0.00 1 1.12
Bagre marinus Gafltopsail Catfish 2 0.04 0.3 0.03 6.4 0.05 1 2.27 0.00 1 1.12
Lepomis sp. Sunfishes 4 0.09 1.4 0.15 23.1 0.20 0.00 0.00 2 2.25
Micropterus sp. Bass 2 0.04 0.2 0.02 4.5 0.04 0.00 0.00 1 1.12
Caranx sp. Jacks 8 0.17 2.9 0.32 99.3 0.84 0.00 0.00 1 1.12
Sparidae Porgies 22 0.47 3.4 0.37 48.9 0.41 0.00 0.00 0.00
Archosargus probatocephalus Sheepshead 37 0.79 21.6 2.37 267.7 2.27 1 2.27 0.00 4 4.49
Sciaenidae Drums 2 0.04 2.2 0.24 23.2 0.20 0.00 0.00 0.00
Cynoscion spp. Seatrout 101 2.15 12.8 1.41 117.1 0.99 0.00 0.00 8 8.99
Micropogonias undulatus Atlantic croaker 1 0.02 0.2 0.02 2.6 0.02 0.00 0.00 1 1.12
Pogonias cromis Black drum 30 0.64 20.5 2.25 180.6 1.53 0.00 0.00 3 3.37
Sciaenops ocellatus Redfish 6 0.13 4.2 0.46 42.0 0.36 0.00 0.00 1 1.12
Mugil spp. Mullet 301 6.40 27.9 3.06 437.5 3.71 0.00 0.00 16 17.98
Gobionelus sp. Gobies 12 0.26 1.0 0.11 29.5 0.25 0.00 0.00 1 1.12
Bothidae Flounder family 132 2.81 19.2 2.11 364.9 3.10 0.00 0.00 4 4.49

Appendix 3. Kinzey's Knoll South, Unit 6 composite 1/4-in sample, faunal remains, continued.

Scientific Name Taxonomic Name NISP % Weight % Biomass % Burnt % Worked % MNI %
Diodontidae Porcupine fishes 3 0.06 0.6 0.07 20.2 0.17 0.00 0.00 1 1.12
All bony fishes 4023 85.60 559.7 61.46 6496.1 55.14 15 34.09 0.00 64 71.91

Squaliformes Sharks 2 0.04 0.4 0.04 57.2 0.49 0.00 1 7.69 0 0.00
Carcharhinidae Requiem sharks 13 0.28 9.0 0.99 833.0 7.07 0.00 0.00 1 1.12
Rajiformes Rays 6 0.13 1.4 0.15 168.1 1.43 0.00 0.00 1 1.12
All sharks and rays 21 0.45 10.8 1.19 1058.4 8.98 0.00 1 7.69 2 2.25

Unidentified Vertebrate All Unidentified Fragments 297 6.32 68.5 7.52 0.00 13 29.55 10 76.92 0.00
Totals 4700 100.0 910.6 100.0 11781.3 100.0 44 100.0 13 100.0 89 100.0

Decapoda Probably blue crabs 15 0.32 5.6 0.61 139.2 1.18 4 100.00

Crassostrea virginica Eastern oyster 145275.3 100.00 21248.4 100.00

Appendix 4. Bluff Midden Units 7 and 8, composite 1/4-in sample, faunal remains.

Scientific Name Taxonomic Name NISP % Weight % Biomass % Burnt % Worked % MNI %
Mammalia, Large Probably deer or bear 153 2.25 192.6 8.27 2993.6 10.08 28 28.00 1 11.11 0.00
Mammalia, Small Probably rabbit or squirrel 22 0.32 4.2 0.18 95.7 0.32 0.00 0.00 0.00
Mammalia Mammals 12 0.18 7.7 0.33 165.1 0.56 2 2.00 0.00 0.00
Didelphis virginiana Eastern opossum 2 0.03 4.9 0.21 109.9 0.37 0.00 0.00 1 0.72
Rodentia Rodents 13 0.19 1.3 0.06 33.3 0.11 0.00 0.00 0.00
Sciurus carolinensis Grey squirrel 7 0.10 1.3 0.06 33.3 0.11 0.00 0.00 2 1.45
Sciurus niger Fox squirrel 14 0.21 14.4 0.62 290.1 0.98 0.00 0.00 3 2.17
Sigmodonhispidus Hispid cotton rat 18 0.26 3.0 0.13 70.7 0.24 0.00 0.00 5 3.62
Delphinidae Dolphins 1 0.01 1.2 0.05 31.0 0.10 0.00 1 11.11 1 0.72
Ursus sp. Bear 2 0.03 92.7 3.98 1550.1 5.22 0.00 0.00 1 0.72
Artiodactyla Probably deer 6 0.09 2.1 0.09 51.3 0.17 0.00 0.00 0.00
Odocoileus virginianus White tail deer 65 0.96 533.9 22.93 7494.0 25.23 2 2.00 2 22.22 4 2.90
All Mammals 315 4.63 859.3 36.91 12918.2 43.49 32 32 4 44.44 17 12.32

Aves All Bird fragments 37 0.54 22.8 0.98 351.3 1.18 1 1.00 0.00 0.00
Branta canadensis Canada goose 1 0.01 3.5 0.15 63.8 0.21 0.00 0.00 1 0.72
Mergus cucullatus Hooded merganser 1 0.01 0.5 0.02 10.9 0.04 0.00 0.00 1 0.72
Larus argentatus Herring gull 1 0.01 0.7 0.03 14.8 0.05 0.00 0.00 1 0.72
Strix varia Barred owl 6 0.09 1.3 0.06 25.9 0.09 0.00 0.00 1 0.72
Colaptes auratus Northern flicker 1 0.01 0.3 0.01 6.8 0.02 0.00 0.00 1 0.72
All Birds 47 0.69 29.1 1.25 473.5 1.59 1 1.00 0.00 5 3.62

Testudines All Turtle fragments 116 1.71 100.4 4.31 693.7 2.34 6 6.00 0.00 0.00
Kinosternidae Mud or musk turtle 19 0.28 4.4 0.19 85.3 0.29 0.00 0.00 2 1.45
Terrapene carolina Eastern box turtle 4 0.06 2.5 0.11 58.4 0.20 0.00 0.00 1 0.72
Trachemys sp. Pond slider 1 0.01 5.5 0.24 99.1 0.33 0.00 0.00 1 0.72
Malaclemys terrapin Diamondback terrapin 4 0.06 6.5 0.28 110.8 0.37 0.00 0.00 2 1.45

Appendix 4. Bluff Midden Units 7 and 8, composite 1/4-in sample, faunal remains, continued.

Scientific Name Taxonomic Name NISP % Weight % Biomass % Burnt % Worked % MNI %
Deirochelys reticularia Chicken turtle 1 0.01 1.0 0.04 31.6 0.11 0.00 0.00 1 0.72
Gopherus polyphemus Gopher tortoise 25 0.37 57.7 2.48 478.6 1.61 0.00 0.00 3 2.17
Apalone ferox Soft shell turtle 5 0.07 10.8 0.46 155.7 0.52 1 1.00 0.00 1 0.72
Cheloniidae Sea turtles 17 0.25 34.7 1.49 340.4 1.15 0.00 0.00 1 0.72
All turtles 192 2.82 223.5 9.60 2053.8 6.91 7 7 0 12 8.70

Serpentes Snakes 3 0.04 0.3 0.01 4.1 0.01 0.00 0.00 1 0.72

Anura Frogs and toads 1 0.01 0.1 0.00 1.3 0.00 0.00 0.00 1 0.72
Osteichthyes All Bony Fish fragments 3772 55.49 591.5 25.41 5191.4 17.48 28 28.00 1 11.11 0.00
Accipenser sp. Sturgeon 16 0.24 8.0 0.34 159.0 0.54 0.00 0.00 1 0.72
Lepisosteus spp. Gar 25 0.37 9.2 0.40 178.1 0.60 0.00 0.00 1 0.72
Amiacalva Bowfin 24 0.35 2.6 0.11 64.0 0.22 0.00 0.00 1 0.72
Elops saurus Ladyfish 28 0.41 2.2 0.09 55.9 0.19 0.00 0.00 1 0.72
Anguilla rostrata American eel 11 0.16 0.7 0.03 22.1 0.07 0.00 0.00 1 0.72
Clupeidae Herrings 65 0.96 4.2 0.18 94.4 0.32 0.00 0.00 3 2.17
Siluriformes All Catfish 175 2.57 28.7 1.23 484.2 1.63 1 1.00 0.00 0.00
Ictaluridae Freshwater Catfish 25 0.37 9.8 0.42 174.4 0.59 0.00 0.00 5 3.62
Ariidae Marine Catfish 213 3.13 58.1 2.50 946.2 3.19 0.00 0.00 10 7.25
Arius felis Hardhead catfish 6 0.09 3.0 0.13 56.7 0.19 0.00 0.00 0.00
Bagremarinus Gafftopsail catfish 17 0.25 3.4 0.15 63.8 0.21 2 2.00 0.00 0.00
Lepomis microlophis Redear sunfish 3 0.04 0.7 0.03 12.9 0.04 0.00 0.00 1 0.72
Pomatomus saltatrix Bluefish 1 0.01 0.3 0.01 11.1 0.04 0.00 0.00 1 0.72
Caranx sp. Jack 117 1.72 67.9 2.92 1592.4 5.36 0.00 0.00 4 2.90
Sparidae Porgies 9 0.13 0.4 0.02 6.8 0.02 0.00 0.00 0.00
Archosargus probatocephalus Sheepshead 291 4.28 131.4 5.64 1409.6 4.75 2 2.00 0.00 12 8.70
Sciaenidae Drums 9 0.13 1.9 0.08 62.6 0.21 0.00 0.00 0.00


Appendix 4. Bluff Midden Units 7 and 8, composite 1/4-in sample, faunal remains, continued.

Scientific Name Taxonomic Name NISP % Weight % Biomass % Burnt % Worked % MNI %
Cynoscion spp. Seatrout 284 4.18 43.8 1.88 637.8 2.15 5 5.00 0.00 12 8.70
Micropogonias undulatus Atlantic croaker 3 0.04 0.6 0.03 26.7 0.09 0.00 0.00 2 1.45
Pogonias cromis Black drum 70 1.03 48.8 2.10 690.9 2.33 0.00 0.00 6 4.35
Sciaenops ocellatus Redfish 45 0.66 14.3 0.61 278.6 0.94 0.00 0.00 7 5.07
Mugil spp. Mullet 474 6.97 42.8 1.84 618.7 2.08 5 5.00 0.00 23 16.67
Gobionellus sp. Gobies 2 0.03 0.2 0.01 8.0 0.03 0.00 0.00 1 0.72
Bothidae Flounder family 259 3.81 52.1 2.24 887.1 2.99 2 2.00 0.00 8 5.80
All bony fishes 5944 87.44 1126.6 48.39 13733.3 46.23 45 45.00 1 11.11 100 72.46

Squaliformes Sharks 1 0.01 0.4 0.02 57.2 0.19 0.00 0.00 0 0.00
Carcharhinus sp. Requiem sharks 4 0.06 1.3 0.06 157.8 0.53 0.00 0.00 1 0.72
Rajiformes Rays 7 0.10 2.8 0.12 305.2 1.03 0.00 0.00 1 0.72
All sharks and rays 12 0.18 4.5 0.19 520.2 1.75 0.00 0.00 2 1.45

Unidentified vertebrate 284 4.18 84.7 3.64 15 15.00 4 44.44 0 0.00

Totals 6798 100.00 2328.1 100.0 29704.5 100.00 100 100.0 9 100.00 138 100.00

Decapoda Crabs, probably blue 4 100.00 1.4 100.0 44.7 100.00 2 100.00

Crassostrea virginica Eastern oyster __ 529053.8 100.0 74438.0 100.00

Appendix 5. Reeves Rise shovel tests, composite 1/4-in sample, faunal remains.

Taxonomic Name Common Name NISP % Weight % Biomass % Burnt % Worked % MNI %
Mammalia, Large Probably deer or bear 13 0.57 16.0 2.56 318.9393 3.60 3 12.50 0.00 0.00
Mammlia, Small Probably rabbit or 12 0.53 2.7 0.43 64.3024 0.72 0.00 0.00 0.00
Mammalia Unidentified mammal 8 0.35 4.6 0.73 103.8681 1.17 2 8.33 0.00 0.00
Sylvilagus sp. Rabbits 1 0.04 0.6 0.10 16.6087 0.19 0.00 0.00 1 1.79
Rodentia Rodents 2 0.09 0.3 0.05 8.9004 0.10 0.00 0.00 0.00
Sciurus carolinensis Grey squirrel 3 0.13 0.9 0.14 23.9231 0.27 0.00 0.00 1 1.79
Sciurusniger Fox squirrel 6 0.26 3.4 0.54 79.1282 0.89 0.00 0.00 3 5.36
Odocoileus virginianus White tail deer 9 0.39 36.8 5.88 674.9367 7.61 0.00 1 25.00 1 1.79
All Mammals 54 2.3653 65.3 10.43 1290.6069 14.55 5 20.83 1 25.00 6 10.71

Aves Unidentified birds 27 1.18 12.8 2.04 207.7596 2.34 0.00 1 25.00 0.00
Podiceps auritus Homed grebe 1 0.04 0.2 0.03 4.7199 0.05 0.00 0.00 1 1.79
Meleagris gallopavo Wild turkey 1 0.04 0.9 0.14 18.5507 0.21 0.00 0.00 1 1.79
All Birds 29 1.27 13.9 2.22 231.0303 2.60 0 0.00 1 25.00 2 3.57

Testudines Unidentified turtles 68 2.98 42.7 6.82 391.1890 4.41 2 8.33 1 25.00 0.00
Trachemys sp. Pond slider 5 0.22 30.8 4.92 314.2889 3.54 0.00 0.00 1 1.79
Deirocehlys reticularia Chicken turtle 2 0.09 1.7 0.27 45.1233 0.51 0.00 0.00 1 1.79
Malaclemys terrapin Diamondback terrapin 8 0.35 7.9 1.26 126.3020 1.42 0.00 0.00 1 1.79
Gopherus polyphemus Gopher tortoise 23 1.01 49.9 7.97 434.2375 4.90 0.00 0.00 2 3.57
Apalone ferox Softshell turtle 2 0.09 1.7 0.27 45.1233 0.51 0.00 0.00 1 1.79
All turtles 108 4.73 134.7 21.51 1356.2640 15.29 2 8.33 1 25.00 6 10.71

Osteichthyes All Bony Fish 1098 48.09 177.4 28.33 1957.2776 22.06 9 37.50 0.00 0.00
Accipenser sp. Sturgeon 1 0.04 0.5 0.08 16.8332 0.19 0.00 0.00 1 1.79
Lepisosteus spp. Gar 8 0.35 3.3 0.53 77.6240 0.88 0.00 0.00 1 1.79
Amia calva Bowfin 5 0.22 0.7 0.11 22.1070 0.25 0.00 _0.00 1 1.79

Appendix 5. Reeves Rise shovel tests, composite 1/4-in sample, faunal remains, continued.

Taxonomic Name Common Name NISP % Weight % Biomass % Burnt % Worked % MNI %
Elops saurus Ladyfish 2 0.09 0.2 0.03 8.4686 0.10 0.00 0.00 1 1.79
Anguilla rostrata American eel 13 0.57 0.8 0.13 25.3187 0.29 0.00 0.00 1 1.79
Clupeidae Herrings 6 0.26 0.6 0.10 20.1716 0.23 0.00 0.00 1 1.79
Siluriformes Catfishes 257 11.26 44.4 7.09 732.8479 8.26 0.00 0.00 0.00
Ictaluridae Freshwater catfish 12 0.53 4.7 0.75 86.7947 0.98 1 4.17 0.00 2 3.57
Ariidae Marine catfish 110 4.82 31.8 5.08 533.7105 6.02 0.00 0.00 12 21.43
Arius felis Hardhead catfish 8 0.35 4.4 0.70 81.5230 0.92 0.00 0.00 0.00
Bagre marinus Gafftopsail Catfish 39 1.71 14.0 2.24 244.8058 2.76 0.00 0.00 0.00
Caranx sp. Probably jack crevelle 35 1.53 19.3 3.08 526.3692 5.93 0.00 0.00 2 3.57
Archosargus Sheepshead 32 1.40 11.9 1.90 154.7046 1.74 0.00 0.00 2 3.57
Sciaenidae Drums 1 0.04 2.5 0.40 76.6435 0.86 0.00 0.00 0.00
Cynoscion spp. Seatrout 81 3.55 10.6 1.69 223.2165 2.52 2 8.33 0.00 3 5.36
Micropogonias Atlantic croaker 1 0.04 0.3 0.05 15.9614 0.18 0.00 0.00 1 1.79
Pogonias cromis Black drum 20 0.88 19.8 3.16 354.4314 4.00 0.00 0.00 5 8.93
Sciaenops ocellatus Redfish 1 0.04 1.8 0.29 60.1038 0.68 0.00 0.00 1 1.79
Mugil spp. Mullet 159 6.96 19.6 3.13 382.8516 4.32 1 4.17 0.00 5 8.93
Bothidae Flounder family 104 4.56 20.4 3.26 385.0976 4.34 1 4.17 0.00 2 3.57
All bony fishes 1993 87.30 389.0 62.12 5986.8621 67.49 14 58.33 0 0.00 41 73.21

Rajiformes Rays 1 0.04 0.2 0.03 6.2794 0.07 0.00 0.00 1 1.79

Unidentified Vertebrate All Unidentified 98 4.29 23.1 3.69 0.00 3 12.50 1 25.00 0.00
Totals 2283 100.00 626.2 100.00 8871.0425 100.00 24 100.00 4 100.00 56 100.00

Decapoda Crabs 5 100.00 2.0 100.00 59.8198 100.00 2 100.00

Crassostrea virginica Eastern oyster 483297.8 100.00 68184.9260 100.00



1805 Twelve Oaks Lane, Neptune Beach, FL 32266
E-mail: vrolland@comcast.net

This paper focuses on the St. Johns II Period ceramic
assemblage recovered from activity areas located north and
west of Shields Mound (8DU12). The majority of pottery is a
spiculate-paste, St. Johns Check-Stamped ware associated with
the local St. Johns II culture (Goggin 1998; Milanich 1994).
Mineral- and grog-tempered Ocmulgee III series pottery also
is well represented (Ashley 2002, 2003, 2005, this issue;
Ashley and Rolland 2002). Ocmulgee III refers to an area
within the greater Ocmulgee River basin defined by Snow
(1977) and Stephenson (1990; also see Stephenson and Snow
2004) (Figure 1). Ocmulgee III vessels, of which most are grit
tempered and cordmarked, appear initially to have been
brought in from its principal area of manufacture in south-
central Georgia but later were locally produced (Ashley 2003,
2005, this issue; Rolland 2004). Combined St. Johns II and
Ocmulgee III wares constitute the local St. Johns II Period
ceramic assemblage of northeastern Florida, dated to A.D.
900-1250 (Ashley 2003:98-102).
From a technological standpoint, this paper focuses on
three aspects of pottery manufacture and use: paste construc-
tion, formal and stylistic characteristics, and the kinds and
frequencies of primary and secondary use-related wear.
Beyond a technofunctional analysis, I consider the role of St.
Johns II and Ocmulgee III vessels as visual markers of group
social affiliation and alliance. The maintenance of discrete
tempers or constituents (sponge spicules vs. mineral or grog
inclusions) and modes of surface modification (check carved
vs. cordage wrapped paddles) that characterize these two
distinct pottery traditions does not necessarily imply factional-
ism but rather may symbolize cooperation and alliance among
geographically disparate peoples. I contend that the continuity
and conservatism of St. Johns and Ocmulgee pottery manufac-
ture represented during the 350 year occupation at Shields
implies that pottery style, technological construction, and
firing techniques were deeply ingrained in cultural identity
and this created resistance to change.

Organization of the Data

The pottery used in this study was collected during testing
of a 300 by 250-m portion of the Shields site under the
direction of Keith Ashley (Figure 2). Sixty-nine shovel tests
were excavated on a 25-m grid, with 64 tests yielding prehis-
toric pottery. In addition, 8 1x2-m units were placed in four
areas of moderate to dense shell midden. To better explore the
distribution of pottery traits, data were organized under two

broad intrasite provenience categories each incorporating five
subareas. One analytical category consists of Shell Midden
Groups (shaded areas) or isolated, moderate to dense shell
deposits that were purposefully constructed between the mound
and bluff edge. The second category, the Shovel Test Groups,
refers to areas of far less dense shell and artifact recovery that
surround the middens. In the following tables, each of the ten
designated subareas will be referred to by initials, which are
listed below.

Shell Midden Groups

Individual Shell Midden groups contain moderate to dense
shell (i.e., more than 20 liters of shell per shovel test) and
common to abundant pottery. Based on the quantity of shell
per shovel test, Ashley (2003:162) identified three distinct
shell middens: Bluff Midden (BM), Reeves Rise (RR), and
Kinzey's Knoll (KK). Two other areas near Kinzey's Knoll
are included since they contained more than 10 liters of shell
per shovel test and frequent to common pottery. These are
designated Kinzey's South (KS) and Kinzey's West (KW).
The midden labeled Reeves Rise was located at the land-
owner's front door and was sampled with three shovel tests
(Table 1). The other four middens were tested with at least
one 1x2-m unit. The Shell Midden Groups were sampled in
disproportionate amounts with Kinzey's Knoll-the probable
focal point of ceremonial discard ca. A.D. 975-
1030-receiving the greatest attention (Table 1).

Shovel Test Groups

Shovel testing outside the five Shell Midden Groups
revealed small amounts of shell and no distinct accumulations
or layers of shell midden. Pottery frequencies varied from
occasional to abundant. The Shovel Test Groups were
designated East Group (EG), Kinzey's Group (KG), West
Bluff Group (WBG), Northwest Group (NWG), and West
Group (WG) (Table 2). The East Group, which lies immedi-
ately north of the mound produced two shovel tests that each
contained greater than 500 g of pottery. Shell quantities
ranged from a trace (less than 1 liter) to 14 liters per shovel
test. The majority of these shovel tests contained less than 3
liters of shell. Bounded unshaded areas in Figure 2 denote
contiguous shovel tests that contained greater than 70 g of
pottery. While depicted outside the core boundaries of Shovel
Test Groups, ceramic data from 23 shovel tests containing less


VOL. 58(3-4)




Figure 1. Locations of the Shields site and Ocmulgee homeland.

than 70 g of pottery were included in the Shovel Test Group
they bordered. Thus, pottery from all shovel tests is included
in this study, and areas of greater and lesser ceramic volume
are established.
Five radiocarbon dates produced from oyster shell suggest
that the focus of shell refuse accumulation shifted through time
(Table 3); however, sigma 1 date ranges overlap implying
continuous occupation of the site. Radiocarbon assays indicate
that Kinzey's Knoll represents the earliest shell midden
accumulation. For a broader discussion of the St. Johns Period
occupation of extreme northeastern Florida see Ashley (2005,
this issue).

Analysis Methods

Ceramic analysis was guided by standardized laboratory
methods, analytical categories, and definitions developed by
Ashley and Rolland (2002), Cordell (1984, 1992, 1993,2001),
Goggin (1998), Rice (1987), Saunders (2000a), Shepard
(1995), and Willey (1998). Paste analysis included identifica-
tion of tempering material, other inclusions, and particle size,
shape, and frequency (Orton et al, 1993:239; Rice 1987:38,
115-153, 349; Rye 1976; Shepard 1995:24-25, 166-168). To
ensure standardization of data, the kinds, size, and relative
abundance of aplastics were recorded using the guidelines
found in Shepard (1956:118) and Rice (1987: 38, 349) (Table


2005 VOL. 58(3-4)


Figure 2. Locations of Shell and Shovel Test Groups.

4). Eleven macroscopic and microscopic observations were
recorded for each sherd larger than 2 cm. Fresh breaks were
made on at least one edge of each sherd and the clean interior
surface was observed under 40X magnification. Data include
paste constituents; exterior and interior surface treatments
(including 4 degrees of plain surface compaction: rough,
finger smoothed, hard tooled, and burnished); body and lip
thickness; lip morphology; vessel form and orifice diameter;
check shape and size; and cordage width. Evidence of primary
(attrition, abrasion, sooting, staining) and secondary (hones,
scrapers) use wear was recorded. A detailed description of
analytical methods and results is provided in Rolland (2004).

Paste Groups

Excavations yielded 13,234 sherds weighing 71.1 kg. Of

these 5,804 sherds (44% by count and 87% by weight) were
larger than 2 cm and subject to detailed paste, formal, and
stylistic analyses. The Shields collection contains two gross
paste categories-spiculate and non-spiculate-which were
established by the presence or absence of common to abundant
sponge spicules (Table 5). A detailed description of the paste
characteristics of the six pottery types and nine paste sub-
groups identified during analysis is offered in Table 6.

Spiculate Pastes

By weight, spiculate pastes make up 75% of the assem-
blage. Four spiculate paste subgroups were identified: St.
Johns, sandy St. Johns, St. Johns-grog, and sandy St. Johns-
grog. St. Johns paste is characterized by common to abundant
sponge spicules, which I interpret as temper (Rolland and




Table 1. Summary of Shell Midden Groups.

Provenience Maximum 50-cmZ 1x2-m Pottery Pottery
size (m) Shovel Tests Units count weight (kg)
Kinzey's Knoll (KK) 50x30 1 4 5,129 34.8
Kinzey's West (KW) 50x25 2 1 570 2.5
Kinzey's South (KS) 30x25 1 1 1,290 4.8
Reeves Rise (RR) 25x25 3 0 485 3.2
BluffMidden (BM) 87x25 5 2 4,297 20.8
Totals 12 8 11,771 66.1

Table 2. Summary of Shovel Test Groups.

Provenience Maximum 50-cm2 Pottery Pottery
size (m) Shovel tests count weight (g)
East Group (EG) 125x 65 10 703 1,977.2
Northwest Group (NWG) 75x37 3 63 317.9
West Group (WG) 112x65 17 187 952.8
West Bluff Group (WBG) 75x87 8 240 917.8
Kinzey's Group (KG) 112x87 14 270 974.9
Totals 52 1463 5,140.6

Table 3. Radiometric dates from four shell midden contexts.

Provenience Beta Measured Conventional Calibrated-1 Reference
ref. # 14C (BP) C14 Age (BP) sigma
(A.D.) with
Kinzey's Knoll TU1 level 7 137818 1080 80 1450 90 865 (975) 1035 Ashley 2002
Kinzey's Knoll TU1 upper 141594 1000 80 1370 60 990 (1030) 1070 Ashley 2002
Bluff Midden TU7- level 6 165353 870 60 1250 60 1070 (1170) 1230 Ashley 2002
Reeves Rise ST-80-100 cmbs 165354 850 50 1210 60 1160 (1220) 1270 Ashley 2002
Kinzey'sSouth- TU6-110 165352 750 60 1120 60 1240 (1290) 1310 Ashley 2002
cmbs ________

Bond 2003). St. Johns paste also contains occasional very fine
to fine-sized quartz inclusions. The same paste construction
is observed in Papys Bayou and Little Manatee sherds. St.
Johns sherds are more likely to be micaceous (i.e., >3 percent
mica platelets) than the three other spiculate subgroups. The
second most frequent spiculate paste is sandy St. Johns. As
defined by Cordell and Koski (2003:120-121), it differs from
St. Johns paste by containing lower frequencies of spicules
(occasional to common), an increased volume of quartz grains
(common to abundant), and a greater range of quartz particle
sizes (very fine to medium). The remaining subgroups, St.
Johns-grog- and sandy St. Johns-grog pastes, are minor

components in this assemblage. Spiculate grog paste contains
coarse to very coarse-sized, ground, spiculate grog or sherd
inclusions that are spherical in shape and occasional to
frequent in occurrence. Two other St. Johns sherds contained
either a single fragment of shell or unidentified inclusions.
Rare, coarse- to granule-size, subrounded iron oxide inclusions
were infrequently observed in spiculate paste groups.

Non-spiculate Pastes

Non-spiculate paste subgroups reveal a wider range of
tempering material (grit, grit-grog, sand, and sand-grog),


2005 VOL. 58(3-4)


Table 4. Standardized size and abundance measurements (Rice 1987; Shepard 1956).

Wentworth Size Classification
Sand Very fine .0625-. 125 mm
Fine .125-.25 mm
Medium .25-.50 mm
Grit Coarse .50-1.0 mm
Very coarse 1.0-2.0 mm
Granule 2.0-4.0 mm
Relative Abundance Scale for Aplastics
Rare-Occasional Less than 1%-3%
Frequent 5%-10%
Common-Abundant 20%-30%

Table 5. Gross paste groups.

non-spiculate spiculate totals
total sherd count 2,261 17% 10,973 83% 13,234 100.0%
total >2cm 1,155 51% 4,649 43% 5,804 44%
total sherd weight (kg) 17.760 25 % 53.375 75 % 71.135 100.0%
total >2cm 15.421 87% 46.660 87% 62.081 87%

temper volume, and particle sizes. Of these, Ocmulgee grit-
tempered sherds were recovered in the greatest numbers (53%
of non-spiculate sherds). The volume of coarse- to very
coarse-sized quartz inclusions varies widely from occasional
to common. Under magnification, grit particles are angular
and crushed to subrounded in shape. Ocmulgee gritty paste
occasionally contains grog inclusions that were often observed
only under magnification of the fresh break. In grit-grog
paste, gritvolume remains variable with occasional to frequent
grog inclusions that are fine to coarse in size, and fractured
and angular in shape. Unlike Ocmulgee sand-grog inclusions,
larger grog particles are rare. Only two grit-grog-tempered
sherds contained spiculate-paste grog incorporated with the
non-spiculate grog inclusions. Combined grit- and grit-grog-
tempered sherds account for nearly 70 percent of non-spiculate
sherds and 17 percent of the total assemblage.
Ocmulgee sand-grog-tempered sherds, the second most
frequently recorded non-spiculate paste subgroup, contain low
volume of sand-sized quartz particles. Grog inclusions are
angular to subangular in shape and medium to granule in size,
with coarse and very coarse sizes predominating. No spiculate
grog inclusions were noted in this paste subgroup. Sand-grog-
and sand-tempered sherds are more apt to contain greater
frequencies of and larger-sized mica platelets.
Other mineral inclusions observed in the Ocmulgee series
include powdered, degraded or crushed, angular limestone,
which reacted to a 10 percent solution of hydrochloric acid.

Angular sandstone inclusions were rarely observed. Other non-
spiculate sherds contained unidentified, rounded lithic
inclusions or white powdery material. Neither reacted to
hydrochloric acid. Ann Cordell examined the powdery
material under petrographic magnification and determined
that it was neither Fuller's Earth nor ground bone. These
materials remain unidentified.
Typological classification of the Shields sand-tempered
wares is problematic. In this collection, and those from other
St. Johns II sites, non-spiculate sand-tempered sherds contain
common to abundant, very fine to medium quartz inclusions,
similar to that seen in sandy St. Johns paste. Some Shields
sand-tempered wares exhibit deep cordmarking and thick,
sturdy, body construction also observed in grit and grit-grog-
tempered sherds, but as a whole, sand-tempered sherds exhibit
a wide variety of body thicknesses and surface treatments.
Therefore, in this study, sand-tempered sherds reflect a temper
Five sherds in the collection are associated with other
Florida cultural assemblages or time periods. One sand-
tempered Keith Incised sherd and one sand tempered, calcare-
ous clay Weeden Island Red Filmed sherd (with unidentified
crushed rock inclusions) were recovered. Three sherds
represent local pottery types from other time periods: two
sand-charcoal tempered sherds from the Middle Woodland
period (ca. A.D. 300-500) and one St. Marys Cordmarked
sherd (ca. A.D. 1250-1500) produced after the Shields





Figure 3. Locations of higher frequencies of spiculate and non-spiculate pastes.


Horizontal Distribution ofPaste Groups

Despite the differential placement, number, and size of
units excavated across the site, differing proportions of paste
subgroups are revealed (Table 7, Figure 3). Higher frequen-
cies of spiculate paste sherds are found in two Shell Midden
Groups, Kinzey's West (KW) and Reeves Rise (RR), and two
Shovel Test Groups, East Group (EG), West Bluff Group
(WBG). Higher proportions of non-spiculate pottery are
recorded in areas located most distant from the mound:
Northwest Group (NWG), West Group (WG).
Spiculate paste subgroups and formal types are unevenly
distributed across the site (Appendix A). Papys Bayou and
Little Manatee sherds, which are associated with St. Johns II
mound contexts, were restricted to Shell Midden Groups.
Sandy St. Johns and St. Johns-grog subgroups were more
frequently recovered in the later dating Bluff Midden and
Reeves Rise.
Uneven distribution of non-spiculate paste subgroups also
was noted (Figure 4, Appendix B). At Kinzey's Knoll, nearly
identical percentages of Ocmulgee grit and grit-grog paste
subgroups were recovered; percentages disproportionately
higher than those found in other Shell Midden Groups or
Shovel Test Groups. Sand-tempered sherds were found in

greater frequencies in later dating middens located farther
from the mound and Kinzey's Knoll. Fifty percent of the
combined sand-tempered and micaceous sand-tempered sherd
weight was concentrated in Bluff Midden. The East Group
held the lowest percentage of sand-tempered sherds. Sand-
grog-tempered sherds were only slightly better represented in
Kinzey's Knoll than Bluff Midden, while micaceous sand-
grog-tempered sherds were more numerous in Kinzey's Knoll.
In this case, surface treatment may contribute to the sorts of
pottery deposited in Kinzey's Knoll (e.g., sand-grog-tempered
sherds recovered from Kinzey's Knoll more often exhibited
burnished surfaces).

Surface Treatment and Paste

Vertical and horizontal pottery distributions reveal a strong
correlation between surface treatment and paste: cordmarking
and grit- and grit-grog-tempering; plain surfaces and sand;
well compacted or obliterated plain surfaces and sand-grog
tempering; and check stamping and spiculate pastes.

Non-spiculate Paste Decorated Surface Treatments

Processed fiber impressions are found on 75 percent of all
Ocmulgee pottery: 70.2 percent are cordmarked and 4.5
percent exhibit warp/weft woven elements. Ninety-four

2005 Voi, 58(3-4)


Table 6. Gross paste categories and their subgroups.

count percent weight gns percent paste description
spiculale pastes 4650 80.1%/ 46,659.7 75.2%
occasional silt-sized quartz particles
rare instances of fne to medium quartz inclusions
St. Johns 4193 72.3% 41,894 67.5% none to frequent mica; majority rare to occasional mica
abundant silt-sized sponge spicules
rare instances of rare to occasional ground iron oxide
1 sherd contained crushed shell
1 sherd contained uid inclusion
occasional to frequent very fine, fine quartz particles
common very fine to medium quartz
sandy St. Johns 403 6.9% 4,386.4 7.1% rare to frequent mica, majority rare mica
rare instances of rare to occasional iron oxide
common silt-sized sponge spicules
same as St. Johns
St. Johns-grog 20 .3% 117.2 .2% rare to occasional rounded fine to medium spiculate grog
sandy St. Johns-grog 5 .09% 60.8 .1% same as sandy St. Johns paste
rare to occasional rounded fine to medium spiculate grog
Papys Bayou 2 .03% 12.3 .02% same as St. Johns
Little Manatee 27 .5% 188.4 .3% same as St. Johns
non-spiculale pastes 1155 19.9% 15,421.6 22.8"
frequent to common subangular to angular coarse to
very coarse grit-sized particles
Ocmulgee grit 635 10.9% 8,138.9 13.1% frequent to common fine, medium quartz
rare to frequent mica, majority occasional to frequent mica
rare instances of limestone inclusions
rare instances of sponge spicules
same as Ocmulgee grit with
Ocmulgee grit-grog 163 2.8% 2,459.6 4.0% occasional to frequent fine to very coarse;
angular to sub-angular non-spiculate grog
2 sherds contained spiculate grog
common to abundant very fine to medium quartz
sand tempered 154 2.7% 1,679.6 2.7% rare instances of occasional grit-sized inclusions
none to frequent mica, majority occasional to frequent mica
abundant very fine to medium quartz particles
sand temper-charcoal 2 .03% 6.1 .01% occasional fine to medium angular voids
occasional to frequent crushed charcoal inclusions
frequent very fine, fine quarts particles
rare to occasional medium, coarse quartz
Ocmulgee sand-grog 197 3.4% 3,100.9 5.0% occasional to frequent mica, 32% frequent
rare instances of rare to occasional limestone, iron oxide, sponge
spicules, 1 sherd uid inclusions
1 .02% 5.2 .01% rare fine sand, pale St. Johns color
common very fine, fine quartz inclusions
St. Marys Cordmarked 1 .02% 10.3 .02% occasional medium quartz
occasional to frequent fine mica platelets
non-local calcareous clay
Weeden Island-like 1 .02% 17.0 .03% frequent very fine, fine sand
occasional coarse sand
rare uid, coarse, angular inclusions
common very fine, fine quartz inclusions
Keith Incised 1 .02% 4.0 .01% occasional medium quartz
rare mica
Totals 5804 100.0% 62,0813 100.0%





Bluff Midden: highest frequencies
of sandy St. Johns, St. Johns-grog,
and sand-tempered sherds

East Group: lowest
frequency of sand-
tempered sherds


Figure 4. Locations of higher frequencies of paste subgroups.

percent of grit- and grit-grog-tempered sherds are cordmarked.
Combined check stamped, incised, or punctated treatments
comprise less than 1 percent of non-spiculate surfaces.

Spiculate Paste Decorated Surface Treatments

Sixty-two percent of all spiculate paste sherds are check
stamped. Square checks predominate in both Shell Midden
Groups and Shovel Test Groups. Less frequently recovered
diamond and rectangular check shapes are unevenly distrib-
uted within the two-provenience categories (Table 8). Dia-
mond checked sherds are more frequent in Shell Midden
Groups and rectangular-checked sherds are more common in
Shovel Test Groups.
St. Johns, Papys Bayou, and Little Manatee incised and
punctated surfaces are associated with Shell Midden Groups.
In this collection incised and punctated design patterns and
applications are highly varied in the field of coverage, density
and depth of punctation, and stylus shape (e.g., round, oval,
triangular, or shell edge). Some punctated designs are
carefully executed, while the application on others is crowded,
erratic, or deep, suggesting that in some cases texturing rather
than aesthetics was the desired effect.
Other kinds of spiculate surface treatments are rare and
include simple stamping, cross simple stamping, and shallow

parallel or erratic scoring. Faint woven or knotted impressions
were recorded in the Shovel Test Group samples. Because the
majority are basal fragments, impressions are more likely
evidence of woven support mats or baskets used during
manufacture than intentional surface decoration. Cordmarking
on St. Johns paste is extremely rare (n = 8) and does not
resemble the tightly wrapped, deep impressions observed on
Ocmulgee wares. The few St. Johns cordmarked sherds were
recovered in and near Bluff Midden.

Plain and Burnished Plain Sherds

Non-spiculate. Twenty percent of non-spiculate paste groups
bear plain surfaces (Figure 5, Table 9). Plain surfaces are
minor components in gritty paste subgroups (3.6 percent) but
the principal surface treatment for sand- and sand-grog paste
subgroups (60 percent of sand-tempered pastes; 65 percent of
sand-grog-tempered pastes). Purposeful obliteration of a
previously impressed surface, often to the point of haphazard
burnishing, is more frequent in the sand-grog paste subgroup.
Plain sand-tempered sherds are found in greater frequency
in Shovel Test Groups along the south and northwest areas of
the site: West Group (15.2 percent of its non-spiculate sam-
ple), Kinzey's Group (22.7 percent), and Northwest Group

Kinzey's Knoll: highest frequencies
of Ocmulgee grit- and grit-grog-tempered and
micaceous sand-grog tempered sherds


2005 Voi. 58(3-4)

Table 7. Distribution and frequencies of spiculate and non-spiculate pastes. Percentages are calculated by total weights
within each area (% area) and the site (% site or 62,081.1 g).

spiculate non-spiculate totals
wt (g) % area % site wt (g) % area % site wt (g) % area % site
Kinzey's West 1697.1 85.4% 2.7% 289.5 14.6% 0.5% 1,986.6 100.0% 3.2%
East Group 1368.4 85.2% 2.2% 238.6 14.8% 0.4% 1,607.0 100.0% 2.6%
Reeves Rise 2467.6 84.2% 4.0% 462.8 15.8% 0.7% 2,930.9 100.0% 4.7%
W. Bluff Group 643.1 81.1% 1.0% 149.5 18.9% 0.2 % 792.6 100.0% 1.3%
Bluff Midden 13,417.7 77.0% 21.6% 4,007.8 23.0% 6.5% 17,428.5 100.0% 28.1%
Kinzey's Group 655.4 76.0% 1.1% 206.5 24.0% 0.3% 861.9 100.0% 1.4%
Kinzey's Knoll 22,826.2 74.4% 36.8% 8,686.8 25.6% 13.8% 31,454.2 100.0% 50.6%
Kinzey's South 2871.5 73.3% 4.6% 1,044.8 26.7% 1.7% 3,916.3 100.0% 6.3%
West Group 614.5 70.1% 1.0% 262.1 29.9% 0.4% 876.6 100.0% 1.4%
N'west Group 98.5 36.3% 0.2% 172.7 63.7% 0.3% 271.2 100.0% 0.4%
totals 46,660.0 75.2% 15,421.1 24.8% 62,081.1 100.0%

(26.7 percent). In the Shell Midden Groups, plain sand-
tempered sherds are significant only in later-dated Reeves Rise
(25.0 percent of its non-spiculate sample). The lowest
percentages of plain sand-tempered sherds are found in Shell
Midden Groups nearest the mound: Kinzey's Knoll (4.0
percent) and Kinzey's South (7.3 percent).
A different discard pattern is revealed for plain sand-grog-
tempered sherds. These vessels are poorly represented in the
areas containing higher frequencies of plain sand-tempered
sherds. Plain sand-grog-tempered sherds constitute a signifi-
cant part of the East Group's non-spiculate sample (36.4
percent) while in nearby Shell Midden Groups they are
uncommon: BluffMidden (18.2 percent), Kinzey's West (20.0
percent), and Kinzey's Knoll (5.3 percent).

Spiculate. Plain sherds comprise 36 percent of the spiculate
assemblage (33.3 percent of St. Johns, 38.1 percent of sandy
St. Johns). The ratios of St. Johns Check Stamped to St. Johns
Plain sherds range from 3:1 (West Group) to 1:0.9 (East and
Kinzey's groups). Bluff Midden (BM) and Kinzey's Knoll
(KK)-each 32.2 percent-contain the lowest percentages of
St. Johns Plain and sandy St. Johns Plain sherds in the Shell
Midden Group samples. St. Johns Plain and sandy St. Johns
Plain are better represented in the two Shovel Test Groups
nearer the mound: East Group (EG) and Kinzey's Group
Concentrations of discarded plain pottery, regardless of
paste, suggest that particular functions or activities better
served by plain vessels were located in selected areas. For
example, refuse from activities that utilized plain pottery was
uncommon in the ceremonial deposits in Kinzey's Knoll but
more common in areas of low shell surrounding that midden.
The frequency of interior and exterior burnished surfaces

observed on St. Johns sherds decreases in locations farthest
from the mound (Table 10). Reeves Rise (RR) and Kinzey's
Knoll (KK) contain higher percentages of burnished plain St.
Johns sherds. Kinzey's Group (KG), the Shovel Test Group
nearest the mound, contains a percentage nearly equal that of
the large, dense shell BluffMidden (BM) sample. Comparing
the frequencies of burnished St. Johns sherds found within the
two provenience categories reveals that a lower percentage are
present in Shovel Test Group (14 percent) than those recov-
ered in the Shell Midden Group (24 percent).

Primary and Secondary Utilization

Evidence of primary vessel use includes additive (e.g.,
sooting, staining, iron oxide as ground powder or red film) and
subtractive (e.g., abrasions, attrition, and pitting) information.
Secondary utilization was recorded for a small number of
sherds that exhibit evidence of recycling as hones or as
finishing or scraping tools.

Sooted Exterior and Carbonized Interior Surfaces

Overall, soot accumulations observed on sherd surfaces are
thin and intermittent. This suggests that vessels were either
infrequently placed directly into fire or were used over
smokeless, glowing embers. Soot was observed on simple and
globular rim forms and all four non-spiculate sub-groups.
Heavy accumulations of soot are rare; 239 sherds (0.04
percent) had heavily sooted exteriors and 301 sherds (0.05
percent) retained carbonized residues on interior surfaces.
Only 36 of798 (4.5 percent) grit- or grit-grog tempered sherds
exhibit heavy sooting and 50 (6.5 percent) exhibit thin interior
lenses of carbonized material. Bluff Midden contains the





Figure 5. Significant frequencies of plain surfaces.

majority of heavily sooted grit- (15 percent) and grit-grog-
tempered (24 percent) sherds (Figure 6). Just four percent of
Kinzey's Knoll non-spiculate sherds exhibit heavily sooted
exteriors and eight percent retain interior carbonized material.
Heavily sooted St. Johns sherds comprise 5 percent of that
paste subgroup, but account for 73 percent of the heavily
sooted sample. Soot was observed on both checked and plain
bowls. Higher frequencies of heavily sooted surfaces were
recovered from the Shell Midden Groups: exterior soot in
Kinzey's South, Kinzey's West, and BluffMidden and interior
carbonized residues in Reeves Rise, Kinzey's Knoll, Kinzey's
South, and Bluff Midden samples.
Heavily sooted sandy St. Johns surfaces (7 percent of
sooted sample) also were found in lower shell contexts:
exterior soot in West Bluff Group, Kinzey's West, and
Kinzey'sKnoll and interior carbonized residue in Reeves Rise,
Kinzey's West, and Kinzey's Group. Little Manatee sherds
rarely exhibit heavy exterior sooting (n = 1 of 12) or interior
carbonized residues (n = 2); all were recovered from Kinzey's

Iron Oxide Residues

It is not clear that the presence of red film is the result of
purposeful decorative surface finishing or incidental residue

from ritual or ceremonial solutions.Partial or complete surface
coverage of red film is present only on interior surfaces of 120
sherds: spiculate pastes (n = 108) and non-spiculate pastes (n
= 12) (Table 11). No sand-tempered sherds exhibited redfilm.
The majority of red-filmed sherds were recovered from
Kinzey's Knoll (KK), including one Little Manatee sherd.
Red filming is observed on four sandy St. Johns sherds (1.1
percent of that paste subgroup). The non-local Weeden Island
Red Filmed sherd is not included in these data.
An additional 29 sherds retained finely ground iron oxide
powder.' Iron oxide powder was observed on interior basal
surfaces, in pitting, or adhering on broken edges of sherds that
may have functioned as scoops. Kinzey's Knoll (n = 8, 28
percent of powdered sample) and Reeves Rise (n = 6, 20
percent) contained the highest frequencies of sherds exhibiting
powered iron oxide. Ground iron oxide residues also were
collected in lower numbers in all but West Group, and two
areas near Kinzey's Knoll: Kinzey's Group, and Kinzey's

Surface Abrasion andAttrition

The Shields assemblage displays little evidence of heavily
damaged surfaces. The number of sherds exhibiting signifi-
cant but really restricted abrasion or surface attrition is shown


2005 VOL. 58(3-4)


Table 8. Distribution of check shapes.

Check Shape Shell Midden Groups Shovel Test Groups
diamond 11% 5%
rectangular 9% 16%
square 80% 79%
multiple shapes <1%
100% 100%

Table 9. Distribution of plain spiculate, sand-, and sand-grog-tempered sherds.

speculate 25% 27% 30% 46% 47% 32% 35% 35% 32% 34%
sand, sand-grog 34% 33% 14% 45% 32% 26 % 30% 23% 9% 39%

Table 10. Distribution of burnished St. Johns Plain sherds.

# plain 11 3 21 56 31 356 45 90 546 64 1223
# burnished 1 0 2 8 6 73 7 21 143 18 279
% burnished 9% 0 10% 14% 19% 21% 16% 23% 26% 28% 23%

Table 11. Distribution of red-filmed sherds.

paste subgroups WG NWG WBG EG KG BM KW KS KK RR totals
St. Johns 1 1 3 13 1 1 82 2 104
sandy St. Johns 1 2 1 4
grit, grit-grog 1 4 1 6
sand-grog 1 3 2 6
totals # 1 2 5 20 1 4 84 3 120
% <1% 2% 4% 17% <1% 3% 70% 3% 100%

Table 12. Frequencies of heavily abraded surfaces, selected middens.

paste subgroup Kinzey's Knoll Bluff Midden
St. Johns 5% (severe abrasion .3%) 16% (severe abrasion 2%)
sandy St. Johns 10% 16% (severe abrasion .5%)
grit, grit-grog 3% 4%
sand 10% 16%
sand-grog 11% 12%

in Table 12. The frequencies of sherds exhibiting more severe
areal surface damage are included. Only data from the two
largest shell midden samples, Kinzey's Knoll and Bluff
Midden are presented.
Overall, surface damage suggesting extended or more

rigorous use is more prevalent in the Bluff Midden sample.
This is most dramatic for the St. Johns, sandy St. Johns, and
sand subgroups. Only 5 percent of the Kinzey's Knoll St.
Johns sherds exhibit significant wear and tear v. 16 percent of
that paste group deposited in the Bluff Midden. On the whole




Figure 6. Distribution of heavily sooted sherds.

Ocmulgee grit- and grit-grog-tempered vessel surfaces were in
pristine conditions with cordage twists and deep impressions
clearly visible. The percentages of abraded sand-grog surfaces
are nearly identical.

Secondary Use Wear

Fifty-four sherds display evidence of recycling as tools.
Twenty-six sherds of these sherds bore hone scars either on
exterior surfaces or perpendicularly across the edge-break
(Figure 7). Hone scars only appear on St. Johns and sandy St.
Johns sherds; the finer textures of these pastes resemble
modem fine-gauge sandpaper. Kinzey's Knoll contained the
highest frequency of hones (42 percent) as well as a significant
number of sandstone abraders. This might suggest that not
only were ceremonial materials deposited within that midden,
but also the tools used to produce them. Bluff Midden (23
percent) and Kinzey's South (8 percent) contained fewer
examples. Hones were recovered from two Shovel Test
Groups: West Bluff (11 percent) and East Group (8 percent).
The second kind of sherd-tool exhibits at least one well-
rounded edge caused by scraping or smoothing against a soft
or yielding surface; most commonly all four edges and covers
are worn. There are 18 of these tools; they are primarily
spiculate (n = 16) and come in a variety of sizes. The majority

was recovered from Shell Midden Group proveniences: Bluff
Midden (33 percent of sample), Kinzey's Knoll (27 percent),
and Kinzey's South (22 percent). One was recovered from
Kinzey's Group, the Shovel Test Group nearest the mound.
Kinzey's South, a later dated midden, contained one grit- and
one grit-grog-tempered sherd worn in a similar manner.
The third type of recycled sherd-tool has one beveled edge
along the narrow width of the coil break (Figure 8). Two of
these were recovered from levels nine and ten in Kinzey's
Knoll; both are St. Johns paste. The larger is a hand tool with
thumb eroded checks and the smaller bears oblique hafting
cordage scars.

Vessel Forms

Technofunctional study of ceramic assemblages couples
analysis of paste composition and surface treatment with vessel
form and size.2 The underlying premise is that, for practical
reasons, potters manipulate vessel construction based on
intended use. Through training, observation, and experience,
aboriginal potters obtained a clear understanding of the
construction, drying, firing, and use-related properties (i.e.,
thermal or impact stress) of local raw clays and aplastic
constituents. Thus, researchers propose that analysis of the
combined attributes of paste, vessel shape, wall thickness, and


2005 VoL. 58(3-4)



Figure 7. Examples of utilized sherds as homes or smoothing or scraping tools.

size can contribute to our understanding of prehistoric con-
tainer needs and functions.
The Shields collection contains three vessel forms: globu-
lar, open, and simple. Table 13 provides a breakdown of all
rim sherds by vessel form, including the counts of those too
small to assign to a specific vessel form category. Because of

the small number of non-spiculate rims assignable to vessel
form categories, the following section focuses on St. Johns and
sandy St. Johns Check Stamped and Plain vessel forms (see
Rolland 2004:75-89 for a discussion on non-spiculate vessel




Figure 8. Utilized St. Johns sherds from Kinzey's Knoll. Top: TU3, level 10; Bottom: TU2, level 9.

Globular Vessels

Rims exhibiting constricted or flattened in-turning orifices
were found in higher proportion in the Shovel Test Groups.
Arnold (1999) suggests that globular containers are the most
versatile and transportable; lending themselves well to a
variety of storage, cooking, and traveling situations. Perhaps
non-residents attending events at Shields may have found
globular bowls better fit their travel and temporary occupation
needs. St. Johns Check Stamped globular rims recovered from
Shell Midden Groups are larger (median = 28 cm) and
constructed in a wider range of sizes than those recovered from
the Shovel Test Groups (median = 21 cm) (Table 14, data by
subareas reported in Appendix C). Check stamped globular
rims from Shell Midden Groups (median = 28 cm) are nearly
70 percent larger than globular plain rims (median = 16 cm).
Plain globular rims from Shell Midden Groups include more
examples of vessels with orifices smaller than 10 cm in
Larger-sized globular check-stamped bowls were recovered

in Kinzey's Knoll (median = 30 cm), the focus of ceremonial
and feasting activities, and Bluff Midden (median = 27 cm).
East Group globular bowls (median = 18 cm) are smaller and
this may reflect vessel construction aimed at individual or
family v. group feasting needs. Restricted and smaller size
ranges (Reeves Rise range = 18-28 cm, and West Bluff Group
range = 19-26 cm) may reflect family oriented assemblages.
Sandy St. Johns Plain globular bowls recovered from Shell
Midden Groups are smaller than both St. Johns Check
Stamped and Plain.

Open Vessels

Open rims (expanding wall/rim orientation) were recov-
ered in the lowest frequencies (Table 13). Open vessels may
have been used for ceremonial presentation or as individual
serving bowls. The differences in sizes may reflect vessel
function as conspicuous display, holding either larger cuts or
larger portions of food, or serving multiple or single attendees.
Check stamped open bowls recovered in Shell Midden Groups

2005 VOL. 58(3-4)



Table 13. Spiculate and non-spiculate rim fragments from Shell Midden Groups (SMG) and Shovel Test Groups (STG)
(percentages read across).

globular open simple uid+<2cm irregular totals
# % # % # % # % #% # %
SMG spiculate 177 21% 66 8% 421 50% 179 21% 6 <1% 849 100%
SMG non spiculate 11 9% 3 3% 81 69% 23 20% 118 100%
SMG subtotal 188 20% 69 7% 502 52% 202 21% 6 <1% 967 100 %
STG spiculate 24 28% 21 25% 36 42% 4 5% 85 100%
STG non-spiculate 2 13% 2 13% 6 40% 5 33% 15 100%
STG subtotal 26 26% 23 23% 42 42% 9 9% 100 100%
total rims 214 20% 92 9% 544 51% 211 20% 6 <1% 1067100 %

Table 14. Summary of spiculate globular rim dimensions.

Proveniences St. Johns St. Johns sandy St. Johns sandy St. Johns
Check Stamped Plain Check Stamped Plain

# range cm # range cm # range cm # range cm
median avemedian average rage median average median average
Shell Midden
Group 75 9-62cm 47 5-52cm 4 16-35cm 6 9-28cm
28cm 28cm 16cm 16cm 26cm 27cm 22cm 20cm
Shovel Test
Group 7 16-32cm 4 10-26cm 1 26cm 3 18-33cm
21cm 25cm 20cm 20cm 30cm 27cm
82 9-66cm 51 5-52cm 5 16-35cm 9 9-33cm
_25cm 26cm 18cm 18cm 26cm 27cm 26cm 24cm

(median = 28 cm) are larger than those from Shovel Test
Group samples (median = 22 cm) (Table 15, Appendix D). In
contrast, Shovel Test Group plain open bowls (median = 25
cm) are larger than Shell Midden Group plain open bowls
(median = 20 cm). Again, the sample in Kinzey's Knoll
contains a larger range of vessel sizes.
Few sandy St. Johns open bowl rims were recovered.
Although a small sample, sandy St. Johns open check stamped
(median = 36 cm) and plain (median = 28 cm) rim diameters
appear constructed with larger orifices than St. Johns Check
Stamped (median = 28 cm) and Plain (median = 25 cm) open
bowls. Large diameter sandy St. Johns open rims were
recovered in areas at the bluff edge: Bluff Midden (median =
30 cm) and West Bluff Group (single rim = 43 cm).

Simple Vessels

Simple or direct rims are the most frequently recovered
vessel form (Table 16, Appendix E) and were constructed in
a wide range of sizes. In the Shields collection, no complete
vessel walls could be reconstructed, therefore bowl and jar
forms could not be distinguished. For this study, simple rims
are defined by vertical or slightly incurring or excurving
rim/wall orientation. Kinzey's West and Kinzey's Knoll
contain greater numbers of large-sized simple St. Johns Check
Stamped and St. Johns Plain vessels and the smaller Reeves
Rise midden contains the smallest simple bowls. In the Shovel
Test Groups, large-sized simple vessels are recorded for
Kinzey's Group (median = 31 cm), which lies near the
ceremonial center of the site, and West Group (median = 32



Figure 9. Canid adorno recovered from Reeves Rise.

cm). Sandy St. Johns Check Stamped simple bowls are
concentrated in Kinzey's Knoll (median = 29 cm), East Group
(median = 27 cm), Reeves Rise (median = 23 cm), and Bluff
Midden (median = 18 cm).

Unusual Vessel Modifications

Recovery of special appendages or more elaborate lip
modifications was extremely rare. Those recovered are from
four Shell Midden Groups (Table 17). The most unique is a
simply formed canid adorno of St. Johns paste that was
recovered from Reeves Rise (Figure 9). The head is well
smoothed with incised eyes and open mouth. The adoro and
a single burnished crossmending rim fragment were fired in a
reduced atmosphere. The head faces the interior of the vessel
similar to the Early Weeden Island effigy vessel illustrated in
Milanich (1994:180).

Ocmulgee Applique Rims

Snow (1977:39) reported that 28 percent of the Ocmulgee
III vessels in his study collection were finished with a folded
rim modification (Figure 10). In the Shields assemblage, 21
rims (or 15 percent) of three non-spiculate paste subgroups
exhibit a rim fold, or more accurately, an applique strip added
at the lip after the vessel had been formed and cordmarked.
Each of the applique strips is highly individualistic. Although
some are little more than a vigorous smear to the exterior,
most are three dimensional. Only one applique is plain; the
others display cordmarking at oblique or perpendicular angles.
One rim was finished with a double applique; and three rims
have ticked lips. The applique may have functioned as a ledge
with which to secure a hide or woven cover during the trip to
the St. Johns region.
The majority of applique rims (n = 18, or 86%) are found

on grit-tempered pastes. Two sand-tempered and one grit-
grog-tempered vessels exhibit applique strips. No sand-grog
tempered rims bore an applique strip. Applique lengths (lip to
base of flattened coil) vary from 5.6 to 20.8 mm (median =
12.8mm); applique thickness (coil dimension rising above the
vessel exterior) varies from nearly flush to 4.5 mm (median =
2.5mm). Sixty-six percent of the applique rims were recovered
in Kinzey's Knoll. Of these, 61 percent came from middle to
upper unit levels (levels 2-5). Six were recovered in Bluff
Midden (50 percent above level 5) and one was recovered from
a shovel test in West Bluff Group.

Kinzey's Knoll: A St Johns H Ceremonial Assemblage

The artifact assemblage from Kinzey's Knoll contains
exotic lithic and mineral materials, a variety of shell and
copper beads, uniquely incised designs on bones (some infused
with powdered iron oxide), and abundant subsistence and
possibly ritual-related faunal remains (Ashley 2005, this
issue; Marrinan 2005, this issue; Penders 2005, this issue;
Bland 2001). Kinzey's Knoll ceramic assemblage also
exhibits a series of formal, stylistic, functional, and use-related
traits that distinguish it from assemblages in other areas of the
site. The density, variety, and distribution of artifacts com-
bined with use-related evidence strongly suggest that Kinzey's
Knoll functioned as a repository for intermittent ritual or
special event discard.
Kinzey's Knoll contains the highest proportions of
Ocmulgee coarse-wares (grit- and grit-grog-tempered
cordmarked pottery) and fine-textured St. Johns and Little
Manatee pottery. Burnished sand-grog-tempered sherds were
more common in this midden, while sandy pastes, both
spiculate and non-spiculate, were more frequently discarded in
other areas of the site. Kinzey's Knoll contains greater
numbers of open/serving vessels and a high frequency oflarge-



2005 VOL. 58(3-4)


Table 15. Summary of spiculate paste open rim dimensions.

Proveniences St. Johns St. Johns sandy St. Johns sandy St. Johns
Check Stamped Plain Check Stamped Plain

# range cm # range cm # range cm # range cm
median average median average median average median average
Shell Midden
Group 25 6-44cm 20 10-32cm 3 24-44 1 26cm
28cm 27cm 24cm 23cm 30cm 33cm
Shovel Test
Group 9 16-32cm 5 12-38cm 1 43cm 1 30cm
22cm 24cm 25cm 26cm
open summary 34 6-44 25 10-38 4 24-44cm 2 26-30cm
23cm 25cm 25cm 24cm 36cm 38cm 28cm 28cm

Table 16. Summary of spiculate paste simple rim dimensions.

Proveniences St. Johns St. Johns sandy St. Johns sandy St. Johns
Check Stamped Plain Check Stamped Plain

# range cm # range cm # range cm # range cm
median average median median median average median average
Shell Midden
Group 162 5-62cm 105 5-68 15 11-45cm 16 8-52cm
29cm 29cm 24cm 23cm 23cm 24cm 31cm 30cm
Shovel Test
Group 16 10-53cm 8 3-55cm 3 9-44cm 2 10-36cm
32cm 30cm 20cm 23cm 27cm 27cm 23 23cm
178 10-62cm 113 5-68cm 18 9-45cm 18 8-52cm
_16 10-53cm 8 3-55cm 3 9-44cm 2 10-36cm

sized vessels.3 Plain pottery is less common in Kinzey's Knoll,
but those examples present are more apt to be burnished.
Evidence of single use vessels (lack of surface damage and
heavy sooting) and iron oxide residues are strongest in the
Kinzey's Knoll sample.
Researchers have proposed that one type of special event
assemblage is revealed by the recovery of large-sized or
specific use vessels and high frequencies of smaller, probably
individual, serving vessels found in discrete areas (Blitz
1993a, 1993b; Deal 1998; De Boer 2001; Hally 1986; Junker
2001; Mills 1999; Renfrew 2001; Steponaitis 1983; Wills

2001). Kinzey's Knoll rim data conform well with those
findings, while other areas of the site, especially the Shovel
Test Group samples, often contain smaller vessel sizes geared
more to individual or family needs.

Stratigraphic Comparisons

Some ceramic studies undertaken in the American South-
west have used stratigraphic variation in the frequencies of
ceramic, faunal, or construction refuse to identify "event
layers" (Toll 2001:73, also see Pepper 1996; Renfrew 2001;





,.. ".1 i W, .4.i4

,, .l

'i fI '

Figure 10. Samples of Ocmulgee III applique rims.
Wills 2001; and Windes 1987:612-667). These researchers

propose that intermittent levels of dense artifact accumulation

And, as a result of their inability to reconstruct even partial
.. t -... ^."t ^-,

vessel walls, Wills (2001) and Toll (2001) suspect that some
;,- "? .. *,. .' ,-i .i ',bl ':' k r ,r I-l

layers of ceramic debris may indicate annual village-wide
cleaning of domestic areas. 612-667). T

propose that intermittent levels of dense artifact regional-scale
ceremonial/ritual gatherings may be present at Shields
required a closer analysis of the stratigraphic data collected
from the four units excavated in Kinzey's Knoll. The frequen-
cies of discard within the levels were variable, and similar to
And, as a routhwestern studies, all efforts to reconstruct even partialcom-
plete vessel wall Wis (2001) and Toll (2001) suspect that some
layers of ceramic debris may indicate annual village-wide
cleaning of domestic areas.
That similar evidence of intermittent regional-scale
ceremonial/ritual gatherings may be present at Shields

present a closer analysis of the stratigraphic distribution of St. Johns and

Ocmulgee grit-tempered sherds recovered from 10-cm levels
withifrom the four x2-units excavated in Kinzey's Knoll. The frequen-tages
cies of discarlculated within the totalevels weight of eah type retr, and similar to

each unit.
the Southwestern studies, all efforts to reconstruct one com-
plete vessel wall section went unrewarded. Figures 11-14
present the stratigraphic distribution of St. Johns and
Ocmulgee grit-tempered sherds recovered from 10-cm levels
within the four 1x2-m units at Kinzey's Knoll. Percentages
are calculated by the total weight of each type retrieved from
each unit.

St. Johns. Unit 1 reveals slight variation in the frequencies of
St. Johns in its lower levels (5-9). The volume of deposition
increases in level 4, followed by decreasing accumulations in
levels 2 and 3 (Figure 11). Units 2 (levels 10, 8, 6, 3) and 3
(levels 12, 9, 7, 2) each contain four peaks. In Unit 2, in-

creased deposits are observed in level 8. A more sustained
period of discard is revealed in levels 3 through 6, then volume
decreases in levels 1 and 2 (Figure 12). In Unit 3, heightened
use is recorded in levels 7, 9, and 12. In the middle levels (3-
6), the rate of discard is somewhat even until level 2 (Figure
13). Unit 4 reveals less activity in the lower levels until level
8. Dense accumulations suggest that another gathering
occurred during levels 2 through 4 (Figure 14).

Ocmulgee grit-tempered. Two peaks are observed in each of
the four units, which roughly coincide with episodes of greater
St. Johns was deposition. In Unit 1, there was an initial (level
9) and second (levels 4 and 5) event concluding with the
volume of grit-tempered pottery deposits overtaking St. Johns
in levels 1-3 (Figure 11). A similar pattern is observed in Unit
2, where there is an early period of significant discard (level 8)
that mirrors the initial St. Johns peak. Little activity occurred
until levels 3 and 4 when deposits overshadow St. Johns in
levels 1 and 2 (Figure 12). A different pattern is revealed in
Unit 3, with low activity until level 6. Later in the upper level
(levels 1-3), the quantity grit-tempered pottery again eclipses
St. Johns deposits (Figure 13). In Unit 4, grit-tempered
pottery is discarded in greater volume than St. Johns (levels 6
and 7); little activity is noted during a later St. Johns peak;
then again grit tempered volume surpasses St. Johns deposits



2005 VoL. 58(3-4)


Table 17. Unusual vessel forms or appendages.

Special form and paste Area Test Unit, level
podal, St. Johns Plain Bluff Midden TU7. level 2
podal or node, grit-tempered, plain Kinzey's Knoll TU3, level 7
rim projection St. Johns Checked-Stamp Bluff Midden TU8, level 2
rim projection St. Johns Incised Kinzey's South TU6, level 6
Canid adorno, St. Johns Plain Reeves Rise ST1032N 1000E, 60-80 cmbs


20.0% -
15.0% -
10.00/% -

Levels 1

2 3 4 5 6 7 8 9

Figure 11. Distribution of St. Johns and Ocmulgee grit-tempered sherds in Test Unit 1. Percentages based on weight total
for each type.


20.0% -

15.0% -

10.0% -

5.0% --

0.0% 4
Levels 1

O TU2-SJ: 4094.9 gms

O TU2-grit: 1163.1 gms



S4 5 6 7 8 9 10
2 3 4 5 6 7 8 9 10

Figure 12. Distribution of St. Johns and Ocmulgee grit-tempered sherds in Test Unit 2. Percentages based on weight total
for each type.


' '

, , , ,





25.0% 3 TU3-SJ: 5970.8 gms -

20.0% --- TU3-grit: 1089.8 gms -




0.0% vdiI4T_ I
Levels 1 2 3 4 5 6 7 8 9 10 11 12

Figure 13. Distribution of St. Johns and Ocmulgee grit-tempered sherds in Test Unit 3. Percentages based on weight total
for each type.







0.0% --L
Levels 1

OTU4-SJ: 3658.8 gms -

O TU4-grit: 859.2 gms

2 3 4 5 6 7 8 9 10 11

Figure 14. Distribution of St. Johns and Ocmulgee grit-tempered sherds in Test Unit 4. Percentages based on weight total
for each type.

in levels 1 and 2 (Figure 14).

Implications of Stratigraphic Accumulations

Stratigraphic ceramic data suggest that Kinzey's Knoll was
partly formed as a result of initial and terminal refuse deposi-
tions with high pottery content, perhaps the byproduct of
regional aggregation and ritual. Between those events, the
midden continued to be used, although pottery accumulation
was less intensive. The lower volume noted in some levels
may represent the domestic refuse of a permanent residential
population and/or deposits from small-scale ritual events. In
Units 1-3, nearly half of the weight of the grit-tempered
samples was recovered in the upper three levels.
It is not clear whether the dominance of grit-tempered

pottery in the upper levels at Shields marks a shift in the use
of St. Johns to Ocmulgee pottery during ritual or if Ocmulgee
pottery was better integrated into society, therefore playing a
greater role in site activities. The retiring of a sample of St.
Johns and Ocmulgee pottery from upper, middle and lower
levels of these units suggests that, by the middle levels,
Ocmulgee pottery was being manufactured with local clays
(Rolland 2004:132-150).4 Thus, the increased frequency of
Ocmulgee vessels, as revealed in the units' upper levels, may
reflect more intensive local production of this southern
Georgia pottery type. The fact that locally and non-locally
produced Ocmulgee wares in the Shield collection are indistin-
guishable in shape, style, and firing technique suggests that
Ocmulgee potters, probably women, may have made extended
visits to the site, taken up residence, or married into the local

2005 VOL. 58(3-4)



St. Johns population.


Two distinct pottery traditions, St. Johns and Ocmulgee,
were maintained by potters from the initial occupation of the
Shields site until its abandonment. At Shields, pottery types
can be distinguished by choice of tempering material, surface
treatment, body thickness, and firing techniques. Through
time, potters within each tradition continued to choose sets of
stylistic and formal traits that served both as functioning
containers and visual cues signifying separate yet allied social
identities within the St. Johns region. The preservation of
clearly distinguishable traits suggests that the perception of a
particular social identity and heritage was an active ingredient
in vessel production. The replacement of broken domestic or
ritual ceramic containers with like vessels provided St. Johns
and Ocmulgee potters the opportunity to express and reinforce
material and symbolic aspects of their daily lives and separate
histories. The conservation and perpetuation of specific
technological styles, formal, and stylistic traits were so
ingrained that even after St. Johns area clays were used,
traditional Ocmulgee construction conventions were upheld.4
Examples of vessels that were produced utilizing the other
ware's traits are limited to a handful of St. Johns cordmarked
sherds or the rare spiculate grog inclusions in Ocmulgee grit-
grog-tempered sherds. The significant absence of blending of
St. Johns and Ocmulgee ceramic traits reflects the durability
of traditional craft production. At the domestic level, ceramic
production incorporates technological style and cultural values
that reinforce social identity (Lechtman 1977:3-8;
Pfaffenberger 2001). Domestic production brings together the
act of teaching and craft production in kin or gender related
task groups (Bunzel 1972; Crown 1999; Gosselain 1992,
2000:190-192; Hagstrum 1995; Rice 1987; Tschopik 1941),
ethnic identification, social memory, and the continuity of
ideological symbols and ritual-related objects (Costin 1998;
David et al. 1988; Emberling 1999; Flannery 1976; Hardin
and Mills 2000; Hegmon et al 1995; Hendon 1999, 2000;
Johnson 2000; Sassaman 1998; Saunders 2000a, 200b, 2001;
Spielmann 1991; Speth 1991; Walker 1995, 2001).
At different times during the early occupation of Shields
(ca. A.D. 975-1030), the focus of activities at Kinzey's
Knoll-an area containing ritual-related deposits-may have
been second only to the events occurring on or within the
mound. At times, Kinzey's Knoll may have been the center of
attention. Data recovered from Kinzey's Knoll suggest that
social identification and social inclusion played a role in those
events. The collection of sherds recovered from Kinzey's
Knoll contained the highest proportions of decorated sherds of
both spiculate and non-spiculate paste groups. This suggests
that identification of the gathered participants was an impor-
tant aspect of the activities in that arena (also see Boone 2000;
Bowser 2000; Mills 1999; Neill 2000; Pertulla 2000; Renfrew
2001; Schortman and Urban 1998; Weissner 1983, 2001;
Wobst 1977, 1999). Further, there may have been special
significance to the materials transported to Shields in

Ocmulgee Cordmarked vessels that merited their discard in the
ceremonial area of the site.
St. Johns and Ocmulgee vessels were an integral part of
both domestic and ritual activities that took place at the
Shields site, as well as other areas throughout northeastern
Florida (Ashley 2003). The reproduction of traditional wares
was not a fragile concept to the St. Johns and Ocmulgee
potters who, through 350 years of interaction, used this
medium to reinforce and relay messages of identity and
function, heritage and alliance.


It has been suggested that quantities of iron oxide powder were
loose in the midden matrix and that it's presence on sherds was
incidental to the environment. In response, it was noted that as the
Shields large faunal assemblage was prepared for analysis that none
bore traces of iron oxide powder. It seems unlikely that natural
processes would have adhered loose powder exclusively to one kind
of midden material.

2 For studies examining the relationships of paste construction, vessel
form, and wall thickness to thermal and impact stresses see Aronson
et al. 1994; Hally 1983a, 1983b, 1986; Mills 1999; Rice 1987;
Sassaman 1993, 1998; Shepard 1995; Schiffer and Skibo 1987.

3Fifty percent of the spiculate rim sherds recovered from Kinzey's
Knoll (n = 212 of 428) could not be used to record lip morphology,
vessel form, and orifice diameter. Of those excluded, most rim
fragments were not large enough to accurately measure orifice
diameter. Several rim fragments bore little arc suggesting that more
vessels constructed with diameters greater than 50-60cm are present
but remain undocumented.

4 The light yellow oxidized hues achieved on St. Johns vessels
contrasted with the reduced or deep reddish-brown clay colors of
imported Ocmulgee pots. Because of Ocmulgee reduction firing
methods, it was only after retiring a variety of pottery types to
optimal oxidized color that the later use of local St. Johns River clays
for Ocmulgee construction was revealed (Rolland 2004).


This must begin with grateful thanks to Keith Ashley, who
conceived of this project that has provided enough material to keep
several of us busy for years. Deepest thanks go to Mr. and Mrs.
Kinzey Reeves, who have protected the site and allowed us to return
again and again with shovels and tarps intent upon ripping up their
beautiful yard. During the excavation ofKinzey's Knoll we dug in the
shadow of their rose garden, and finer backdrop cannot be imagined.
Thank you to Keith Ashley, Rochelle Marrinan, Ken Sassaman, and
Rebecca Saunders who read earlier versions of this paper and, as
always, made helpful suggestions. Thank you to Ann Cordell and
Ryan Wheeler for their final editing comments and their patience.
Lastly, thank you to Jeff Rolland who, along with all other archaeo-
logical spouses, puts up with a whole lot difficult behavior and hours.

References Cited

Arnold, Phillip J. m
1999 Tecomates, Residential Mobility, and Early Formative
Occupation in Coastal Lowland Mesoamerica. In Pottery
and People, edited by James Skibo, pp. 159-171. Univer-




sity of Utah Press, Salt lake City.

Aronson, M., James Skibo, and Miriam Stark
1994 Production and Use Technologies in Kalinga Pottery. In
Kalinga Ethnoarchaeology: Expanding Archaeological
Method and Theory, edited by W. Longacre and J. Skibo,
pp. 83-113. Smithsonian Institution Press, Washington

Ashley, Keith H.
2002 On the Periphery of the Early Mississippian World:
Looking Within and Beyond Northeastern Florida. South-
eastern Archaeology 21(2):162-178.

2003 Interaction, Population Movement, and Political Economy:
The Changing Social Landscape of Northeastern Florida
(A.D. 900-1500). Ph.D. dissertation, Department of
Anthropology, University of Florida, Gainesville.

2005 Introducing Shields Mound (8DU12) and the Mill Cove
Complex. The Florida Anthropologist (This issue).

Ashley, Keith H., and Vicki Rolland
2002 St. Marys Cordmarked Pottery (Formally Savannah Fine
Cord Marked of Northeastern Florida and Southeastern
Georgia: A Type Description. The Florida Anthropologist

Bland, Myles C.P.
2001 Moore to the Point. Paper presented at the 58" annual
meeting of the Southeastern Archaeological Conference,
Chattanooga, Tennessee.

Blitz, John
1993a Big Pots for Big Shots: Feasting and Storage in a Missis-
sippian Community. American Antiquity 58(1):80-96.

1993b Ancient Chiefdoms of the Tombigbee. University of Ala-
bama Press, Tuscaloosa.

Boone, James L.
2000 Status Signaling, Social Power, and Lineage Survival. In
Hierarchies inAction CuiBono?, Occasional PaperNo. 27,
edited by M. Diehl, pp. 84-113. Southern Illinois Univer-
sity, Carbondale.

Bowser, Brenda J.
2000 The Role of Ceramics in Political Action. From Potter to
Politics: An Ethnoarchaeological Study of Political Faction-
alism, Ethnicity, and Domestic Pottery Style in the Ecua-
dorian Amazon. Journal of Archaeological Method and
Theory 7(3), edited Brenda J. Bowser, pp. 219-249.
Kluwer Academic Publishers, Norwell, Massachusetts.

Bunzel, Ruth L.
1972 The Pueblo Potter. Dover Publications, New York.

Cordell, Ann
1984 Ceramic Technology at a Weeden Island Period Archaeo-
logical Site in North Florida. In Ceramic Notes No.2.
Occasional Publications of the Ceramic Technology
Laboratory, Florida State Museum, Gainesville. 1992
Technological Investigation of Pottery Variability in

Southwest Florida. In Culture and Environment in the
Domain of the Calusa, pp. 105-191. Monograph Number
1. Institute of Archaeology and Paleoenvironmental
Studies, University of Florida. Gainesville.

1993 Chronology Variability in Ceramic Paste: A Comparison of
Deptford and Savannah Period Pottery in the St. Marys
River Region of Northeast Florida and Southeast Georgia.
Southeastern Archaeology 12(1):33-58.

2001 Continuity and Change inApalacheePottery Manufacture.
University of South Alabama Center for Archaeological
Studies, Mobile.

Cordell, Ann, and Steven Koski
2003 Analysis of a Spiculate Clay from Lake Monroe, Volusia
County, Florida. The Florida Anthropologist 56(2):113-

Costin, Cathy Lynne
1998 Introduction: Craft and Social Identity. In Craft and Social
Identity, edited by C. Costin and R.Wright, pp. 3-19.
Archeological Papers of the American Anthropological
Association Number 8, Washington D.C.

Crown, Patricia
1999 Socialization in American Southwest Pottery Decoration.
In Pottery and People, edited by J. Skibo and G. Feinman,
pp. 24-43. University of Utah Press, Salt Lake City.

David, Nicholas, Judy Sterner, and Kodzo Gavua
1988 Why Pots are Decorated. CurrentAnthropology 29(3):365-

De Boer, Warren R.
2001 The Big Drink: Feast and Forum in the Upper Amazon. In
Feasts, edited by M. Dietler and B. Hayden, pp. 215-240.
Smithsonian Institution, Washington D.C.

Deal, Michael
1998 Pottery Ethnoarchaeology in the CentralMaya Highlands.
University of Utah Press, Salt Lake City.

Emberling, Geoff
1999 The Value of Tradition: The Development of Social
Identities of Early Mesopotamian States. In Material
Symbols, Occasional Paper No. 26, edited by J. Robb, pp.
277-302. Southern Illinois University, Carbondale.

Flannery, Kent V.
1976 Interregional Religious Networks. In The Early
Mesoamerican Village, edited by K. Flannery, pp. 329-345.
Academic Press, Orlando.

Goggin, John M.
1998 [1952] Space and Time Perspective in Northern St. Johns
Archeology, Florida. University Press of Florida,

Gosselain, Olivier P.
1992 Technology and Style: Potters and Pottery Among the Bafia
of Cameroon. Man 27:559-586.


2005 VOL. 58(3-4)


2000 Materializing Identities: An African Perspective. In Journal
ofArchaeologicalMethod and Theory Vol.7(3), edited by
B. Bowser, pp. 187-219. Kluwer Academic Publishing,
Norwall, Massachusetts.

Hagstrum, Melissa
1995 Creativity and Craft: Household Pottery Traditions in the
Southwest. In Ceramic Production in the American
Southwest, edited by B. Mills and P. Crown, pp. 281-301.
University of Arizona Press, Tucson.

Hally, David J.
1983a Use Alteration of Pottery Vessel Surfaces, An Important
Source of Evidence for the Identification of Vessel Func-
tion. North American Archaeologist 4(1):3-26.

1983b The Interpretive Potential of Pottery from Domestic
Contexts. Midcontinental Journal ofArchaeology 8:163-

1986 Identification of Vessel Function: A Case Study from
Northwest Georgia. American Antiquity 51:267-295.

Hardin, Margaret A., and Barbara Mills
2000 A Social and Historical Context of Short-term Stylistic
Replacement: A Zuni Case. Journal of Archaeological
Method and Theory, Vol.7(3), edited by B. Bowser, pp.
139-165. Kluwer Academic Publishing, Norwall, Massa-

Hegmon, Michelle, Winston Hurst, and James R. Allison
1995 Production for Local Consumption and Exchange: Compari-
sons of Early Red and White Ware Ceramics in the San
Juan Region. In Ceramic Production In the American
Southwest, edited by Barbara Mills and Patricia Crown, pp.
30-63. University of Arizona Press, Tucson.

Hendon, Julia
1999 Multiple Sources of Prestige and the Social Evaluation of
Women in Prehistoric Mesoamerica. InMaterial Symbols,
Occasional Paper No. 26, edited by J. Robb, pp. 257-277.
Southern Illinois University, Carbondale.

2002 Social Relations and Collective Identities: Household and
Community in Ancient Mesoamerica. In The Dynamics of
Power, Occasional Paper No. 30, edited by M. O'Donovan,
pp. 273-301. Southern Illinois University, Carbondale.

Johnson, Eric S.
2000 The Politics of Pottery: Material Culture and Political
Process Among Algonquians of Seventeenth-Century
Southern New England. In Interpretations ofNative North
American Life, edited by M. Nassaney and E. Johnson, pp.
118-146. University Press of Florida, Gainesville.

Junker, Laura Lee
2001 The Evolution of Ritual Feasting Systems in Prehispanic
Philippine Chiefdoms. In Feasts, edited by M. Dietler and
B. Hayden, pp.267-311. Smithsonian Institution Press,
Washington D.C.

Lechtman, Heather
1977 Style in Technology-Some Early Thoughts. In Material

Culture, Styles Organization, and Dynamics of Technol-
ogy. edited by Heather Lechtman and Robert Merrill, pp.
3-21. West Publishing Co., St. Paul.

Marrinan, Rochelle
2005 Early Mississippian Faunal Remains from the Shields
Mound (8DU12). The FloridaAnthropologist (This issue).

Milanich, Jerald T.
1994 Archaeology ofPrecolumbian Florida. University Press of
Florida, Gainesville.

Mills, Barbara
1999 Ceramics and Social Contexts of Food Consumption in the
Northern Southwest. In Pottery andPeople, edited by J.M.
Skibo and G.M. Feinman, pp. 99-115. University of Utah
Press, Salt Lake City.

Neill, Susan M.
2000 Emblems of Ethnicity: Ribbon Garments from the Great
Lakes Region. In Interpretations ofNative North American
Life, edited by M. Nassaney and E. Johnson, pp. 146-173.
University Press of Florida, Gainesville.

Orton, Clive, and Paul Tyers, and Alan Vince
1993 Pottery in Archeology. Cambridge University Press,

Penders, Thomas E.
2005 Bone, Antler, Dentary, and Shell Artifacts from the Shields
Site (8DU12). The Florida Anthropologist (This issue).

Pepper, George H.
1996 [1920] Pueblo Bonito. University of New Mexico Press,

Perttula, Timothy
2000 Functional and Stylistic Analyses of Ceramic Vessels from
Mortuary Features at a 15t and 16~ Century Caddo Site in
Northeast Texas. Midcontinental Journal ofArchaeology

Pfaffenberger, Bryan
2001 Symbols DoNot Create Meanings-Activities Do: Or, Why
Symbolic Anthropology Needs the Anthropology of Tech-
nology. In Anthropological Perspectives in Technology,
edited by M. Schiffer, pp. 77-87. University of New
Mexico Press, Albuquerque.

Renfrew, Colin
2001 Production and Consumption in a Sacred Economy: The
Materials Correlates of High Devotional Expression at
Chaco Canyon. American Antiquity 66(1): 14-26.

Rice, Prudence M.
1987 Pottery Analysis: A Sourcebook. University of Chicago
Press, Chicago

Rolland, Vicki L.
2004 Measuring Tradition and Variation: A St. Johns II Pottery
Assemblage from the Shields Site (8DU12). Unpublished
Master's thesis, Department ofAnthropology, Florida State
University, Tallahassee.




Rye, O.S.
1976 Keeping your Temper Under Control, Materials and the
Manufacture of Papuan Pottery. Archeology and Physical
Anthropology in Oceania 11(2):106-137.

Sassaman, Kenneth E.
1993 Early Pottery in the Southeast. University of Alabama
Press, Tuscaloosa.

1998 Crafting Cultural Identity in Hunter-Gather Economies. In
Craft and Social Identity, edited by C. Costin and R.
Wright, pp. 93-109. Archeological Papers of the American
Anthropological Association Number 8, Washington D.C.

Saunders, Rebecca A.
2000a Stability and Change in Guale Indian Pottery A.D. 1300-
1702. University of Alabama Press, Tuscaloosa.

2000b The Guale Indians of the Lower Atlantic Coast: Change
and Continuity. In Indians of the Southeast, edited by B.
McEwan, pp. 26-57. University Press of Florida,

2001 Negotiated Tradition? In Archaeology of Traditions:
Agency and History Before and After Columbus, edited by
T. Pauketat, pp.77-93. University Presses of Florida,

Schiffer, Michael, and James Skibo
1987 Theory and experiment in the Study of Technological
Change. Current Anthropology 28:595-622.

Shepard, Anna O.
1995 [1956] Ceramics for the Archaeologist. Publication 609.
Carnegie Institution of Washington, Washington D.C.

Schortman, Edward M., and Patricia A. Urban
1998 Culture Contact and Structure and Process. In Studies in
Culture Contact, Occasional Paper No. 25, edited J.
Cusick, pp. 102-126. Southern Illinois University, Carbon-

Snow, Frankie
1977 An Archeological Survey of the Ocmulgee Big Bend
Region. Occasional Papers from South Georgia, No.3.
South Georgia College, Douglas.

Speth, John D.
1991 Some Unexpected Aspects of Mutualistic Plains-Pueblo
FoodExchange. In Farmers, Hunters, and Colonists, edited
by Katherine Spielmann, pp. 18-36. University of Arizona
Press, Tucson.

Spielmann, Katherine A.
1991 Interaction Among Nonhierarchical Societies. In Farmers,
Hunters, and Colonists, edited by K. Spielmann, pp. 1-18.
The University of Arizona Press, Tucson.

Stephenson. Donald Keith
1990 Investigation of Ocmulgee Cord-Marked Pottery Sites in
the Big Bend Region of Georgia. Unpublished Master's
thesis, Department ofAnthropology, University of Georgia,

Stephenson, Keith, and Frankei Snow
2004 Swift Creek to Square Ground Lamar: Situating the
Ocmulgee Big Ben Region in Calibrated Time. Early
Georgia 32(2):127-160.

Steponaitis, Vincas P.
1983 Ceramics, ChronologyandCommunityPatterns. Academic
Press, New York.

Toll, Wolcott H.
2001 Making and Breaking Pots in the Chaco World. American
Antiquity 66(1) 56-78.

Tschopik, Jr., Harry
1941 Navajo Pottery Making. Peabody Museum of American
Archaeology and Ethnology 22(1), Harvard University,

Walker, William H.
1995 Ceremonial Trash? In Expanding Archeology, edited by J.
Skibo, W. Walker, and A. Nielsen, pp. 67-80. University
of Utah Press, Sale Lake City.

2001 Ritual Technology in an Extranatural World. In Anthropo-
logicalPerspectives on Technology, edited byM. Schiffer,
pp. 87-107. University ofNewMexico Press, Albuquerque.

Weissner, Polly
1983 Style and Social Information in Kalahari San Projectile
Points. American Antiquity 48(2):253-276.

2001 Of Feasting and Value: Enga Feasts in a Historical Per-
spective (Papua New Guinea). In Feasts, edited by M.
Dietler and B. Hayden, pp. 115-144. Smithsonian Institu-
tion Press, Washington D.C.

Wills, W.H.
2001 Ritual and Mound Formation During the Bonito Phase in
Chaco Canyon. American Antiquity 66(3):433-451.

Willey, Gordon R.
1998 [ 1948] Archeology of the Florida Coast. University Press
of Florida, Gainesville.

Windes, Thomas C.
1987 Investigation at the Pueblo Alto Complex, Chaco Canyon,
Vol. II, Part 2. National Park Service, Santa Fe, New

Wobst, H.M.
1977 Stylistic Behavior and Information Exchange. In Papersfor
the Director: Research Essays in Honor of James B.
Griffin, edited by C. Clelan, pp. 317-342. University of
Michigan Museum of Anthropology Anthropological
Papers 61, Ann Arbor.

1999 Style in Archaeology or Archaeologists in Style. In Mate-
rial Meanings, edited by E. Chilton, pp. 118-133. Univer-
sity of Utah Press, Salt Lake City.


2005 VOL. 58(3-4)


Appendix A. Distribution of spiculate pastes. Percentages based on total weight
for each paste subgroup (read down).

St. Johns Papys Bayou Little Manatee s. St. Johns St. Johns-grog s. St Johns-grog
WG 543.6 1% 59.1 1% 11.8 10%
NWG 63.9 <1% 24.5 <1% 10.1 16%
WBG 505.0 1% 132.7 3% 5.4 5%
KG 537.7 1% 110.9 3% 6.8 6%
EG 1126.1 3% 239.4 6% 3.0 3%
KW 1588.2 4% 4.9 40% 104.0 2%
KS 2715.6 7% 155.9 4%
KK 21295.7 51% 7.4 60% 119.7 64% 1368.5 31% 12.0 10% 22.6 37%
RR 2025.7 5% 35.5 19% 346.6 8% 48.6 41% 11.5 20%
BM 11496.0 27% 33.2 17% 1845.1 42% 29.7 25% 16.6 27%
total 41,897g 100% 12.3g 100% 188.4g 100% 4286.4g 100% 117.2g 100% 60.8g 100%

Appendix B. Distribution of non-spiculate pastes. Percentages based on total weight for each paste group (read down).

micaceous micaceous
Ocmulgee sand sa Ocmulgee Ocmuee
Ocmulgee grit sand Ocmulgee
grit-grog tempered tempered sand-grog s -grg
tempered sand-grog
WG 108.3 1.3 45.5 1.8 41.8 2.5 66.1 2.1
NWG 81.8 1.0 21.8 1.3 4.2 1.2 69.1 2.2
WBG 82.8 0.9 42.4 1.7 15.9 0.9 7.9 2.2 18.4 0.6
KG 141.5 1.7 6.9 0.3 30.0 1.8 3.3 0.9 22.4 0.7 22.4 2.0
EG 46.1 0.6 12.0 0.5 11.7 0.7 158.5 5.1
KW 60.8 0.7 27.3 11 38.9 2.3 162.2 5.2
KS 492.5 6.1 226.1 9.2 79.7 4.7 246.5 7.9 82.2 7.3
KK 5278.8 64.9 1593.4 64.8 459.3 27.3 110.7 31.2 1229.1 39.6 5827.0 51.7
RR 172.7 2.1 21.2 0.9 129.8 7.7 40.9 11.5 139.6 4.5 24.9 2.2
BM 1683.6 20.7 484.8 19.7 850.7 50.6 187.8 52.9 988.7 31.9 413.8 36.7
total 8138.9g 100% 2459.6g 100% 188.4g 100% 354.8g 100% 4386.4g 100% 1126.0.8g 100%

Appendix C. St. Johns globular rims.

St. Johns St. Johns sandy St. Johns sandy St. Johns
check Stamped Plain Check Stamped Plain
# range cm # range cm # range cm # range cm
median average median average median average median average
Kinzey's South 7 12-48 cm 5 6-12cm 1 26cm
23 cm 24cm 8cm 9cm
Kinzey's West 1 32cm

Bluff Midden 32 12-66 17 8-50cm 3 16-35 2 12-16cm
27cm 28cm 24cm 25cm 21cm 24cm 14cm
Reeves Rise 3 18-28cm 2 7-27cm 1 30cm
28cm 24cm 17cm
Kinzey's Knoll 32 9-62cm 23 5-52cm 3 9-28cm
30cm 30cm 14cm 16cm 22cm 20cm
midden subtotals 75 9-62cm 47 5-52cm 4 16-35cm 6 9-28cm
__28cm 28cm 16cm 16cm 26cm 27cm 22cm 20cm


East Group 3 16-52cm 2 10-26cm 1 26cm 1 33cm
18cm 29cm 18cm
Kinzey's Group 1 32cm 2 20-22cm 1 30cm
West Bluff Group 2 19-26 1 18cm
West Group 1 14cm

Northwest Group

group subtotals 7 16-32cm 4 10-26cm 1 26cm 3 18-33cm
21cm 25cm 20cm 20cm 30cm 27cm
globular summary 82 9-66cm 51 5-52cm 5 16-35cm 9 9-33cm
27c 26cm 18cm 18cm 26cm cm 8m 4cm 24cm

Appendix D. St. Johns open rims.

St Johns St. Johns sandy St. Johns sandy St. Johns
Check Stamped Plain Check Stamped Plain
# range cm # range cm # range cm # range cm
median cm average median cm average cm median cm average cm m median cm average cm
Kinzey's South 3 26-38cm 4 20-32cm
38cm 34cm 27cm 26cm
Kinzey's West 1 24cm

BluffMidden 7 13-23cm 1 20cm 3 24-44cm
20cm 19cm 30cm 33cm
Reeves Rise 1 20cm

Kinzey's Knoll 15 644cm 13 10-32 1 26cm
28cm 27cm 25cm 24cm
midden subtotals 25 6-44cm 20 10-32cm 3 24-44 1 26cm
28cm 27cm4c24cm 23cm 0cm 33cm
East Group 2 25-32cm
1 26cm
Kinzey's Group 1 24cm 3 24-30cm 1 30cm
24cm 26cm
West Bluff Group 2 16-27cm 2 12-38cm 1 43cm
22cm 25cm
West Group 3 22-24
22cm 23cm
Northwest Group 1 22cm

group subtotals 9 16-32cm 5 12-38cm 1 43cm 1 30cm
22cm 24cm 25cm 26cm
open form totals 34 6-44 25 10-38 4 24-44cm 2 26-30cm
23cm 25cm 25cm 24cm 36cm 38cm 28cm 28cm

2005 VoI. 58(3-4)


Appendix E. St. Johns simple rims.

St. Johns St Johns sandy St Johns sandy St. Johns
Check Stamped Plain Check Stamped Plain
# range cm # range cm # range cm # range cm
median cm average cm median cm average cm median cm average cm median cm average cm
Kinzey's South 14 12-46cm 6 14-36cm
24cm 26cm 26cm 24cm
Kinzey's West 9 5-51cm 9 10-51cm
29cm 30cm 30cm 30cm
Bluff Midden 57 12-53cm 34 6-42cm 9 11-45cm 12 8-35cm
29cm 30cm 19cm 20cm 18cm 24cm 22cm 21cm
Reeves Rise 4 24-38cm 5 10-28cm 3 18-27
32cm 32cm 18cm 18cm 23cm 22cm
Kinzey's Knoll 79 10-62cm 51 5-68cm 3 17-34cm 4 24-52cm
28cm 29cm 24cm 25cm 29cm 27cm 40cm 39cm
midden subtotals 162 5-62cm 105 5-68 15 11-45cm 16 8-52cm
29cm 29cm 24cm 23cm 23cm 24cm 31cm 30cm
East Group 2 28-40cm 2 3-23cm 3 9-44cm 1 10cm
34cm 13cm 27cm 27cm
Kinzey's Group 5 11-53cm 2 5-22cm
31cm 30cm 14cm
West Bluff Group 5 10-40cm 2 14-55cm
16cm 20cm 35cm
West Group 4 20-51cm 1 20cm 1 36cm
32cm 34cm
Northwest Group 1 34cm

group subtotals 16 10-53cm 8 3-55cm 3 9-44cm 2 10-36cm
32cm 30cm 20cm 23cm 27cm 27cm 23 23cm
simple form totals 178 10-62cm 113 5-68cm 18 9-45cm 18 8-52cm
___29cm 30cm 22cm 23cm 25cm 25cm 29cm 27cm



Florida Anthropological Society Chapters









9 7- -:. 1 0


) Archaeological Society of Southern Florida
495 NW 35th Ave., Miami, FL 33142

) Broward County Archaeological Society 5
81 S. Federal Highway, Dania Beach, FL 33004

) Central Florida Anthropological Society
.O. Box 947544, Maitland, FL 32794-7544 4 -

) Central Gulf Coast Archaeological Society 13 6
.O. Box 82255, St. Petersburg, FL 33682

) Indian River Anthropological Society
705 S. Tropical Trail, Merritt Island, FL 32952

) Kissimmee Valley Archaeological and Historical Conservancy 2
95 Huntley Oaks Blvd., Lake Placid, FL 33852 12

) Emerald Coast Archaeological Society
33 Persimmon Street, Freeport, FL 32435

) Panhandle Archaeological Society at Tallahassee
/o The Tallahassee Trust for Historic Preservation"
,23 E. Virginia Street, Tallahassee, FL 32301 -
./ -q"-

9) Pensacola Archaeological Society
P.O. Box 13251, Pensacola, FL 32591

10) St. Augustine Archaeological Association
P.O. Box 1301, St. Augustine, FL 32085

11) Southeast Florida Archaeological Society
P.O. Box 2875, Stuart, FL 34995

12) Southwest Florida Archaeological Society
P.O. Box 9965, Naples, FL 34101

13) Time Sifters Archaeology Society
P.O. Box 25883, Sarasota, FL 34277-2883

14) Volusia Anthropological Society
P.O. Box 1881, Ormond Beach, FL 32175

15) Warm Mineral Springs Archaeological Society
P.O. Box 7797, North Port, FL 34287



Thomas Penders & Associates, 3365 Heather Drive, Titusville, FL 32796
E-mail: pendarch@yahoo.com

Bone and shell artifacts have been recovered from numer-
ous archaeological sites throughout Florida. This is particu-
larly the case along the eastern and southwestern coasts of
Florida, where shell and bone served as useful raw materials
in the absence of locally available stone. Recent excavations
at the Shields site (8DU12) near the mouth of the St. Johns
River in northeastern Florida yielded 114 bone and shell
artifacts from five different loci (Ashley 2003:158-178). This
paper summarizes functional and stylistic analysis of these
artifacts, the results of which were compared to those from
other St. Johns II sites. Special attention was paid to design
elements on bone, with some iconography possibly represent-
ing native cosmology. Based on the quality and quantity of
faunal artifacts across the sampled section of the site, both
daily living areas and at least one special use loci (Kinzey's
Knoll) are suggested. Finally, consideration is given to the
role whelk shell played as a raw material and exchange item
in Mississippian economic systems that extend beyond
northeastern Florida.

Excavation Areas

The Shields site (8DU12) is a St. Johns II (A.D. 900-1250)
mound center located east of downtown Jacksonville along the
lower St. Johns River. It consists of a large platform mound
excavated by C.B. Moore (1895) in the 1890s and a series of
middens. Bone and shell artifacts used in this study are from
excavations conducted over a three-year period, 1999-2002
(Ashley, 2005a, this issue). The majority of artifacts came
from three distinct shell midden areas designated Kinzey's
Knoll, Reeves Rise, and BluffMidden (Figure 1). A minority
of the artifacts were recovered from two peripheral areas
designated Kinzey's South and Kinzey's West, and from
shovel tests placed outside these areas (herein referred to as
Shields Other) (Figure 1).
In 1999, 2 1x2-m units (Units 1, 2) were excavated atop
Kinzey's Knoll, a dome shaped shell midden (30x25 m)
located about 50 m northwest of the Shields Mound. The
following summer, 2 additional 1x2-m units (Units 3 and 4)
were excavated, so that the four formed two adjacent 2-m
squares. Radiocarbon assays on shell from Kinzey's Knoll
provided ranges of A.D. 865-1035 and A.D. 990-1070 (Ashley
2005a, this issue). Two more 1x2-m units, Unit 5 (Kinzey's
South) and Unit 6 (Kinzey's West), were excavated the
following year and placed south and west of Units 1-4,
respectively. A radiocarbon date with the range of A.D. 1240-

1310 was obtained from shell from a deep feature in Unit 5
(Ashley 2005a, this issue).
Reeves Rise was a circular shell midden, approximately 20
m in diameter, located 25 m north of Kinzey's Knoll. Testing
of this midden consisted of three closely-spaced 0.5-m square
shovel tests. Radiocarbon dating places the sampled portion
of the shell midden at Reeve's Rise at A.D. 1160-1270 (Ashley
2005a, this issue). The Bluff Midden was a linear shell
midden along the bluff edge that measured 90 x 25 m. Two
1x2-m units (Units 7 and 8) were combined to form a 2-m
square, located within the western section of Bluff Midden.
Shell from the Bluff Midden was radiocarbon dated to A.D.
1070-1230 (Ashley 2005a, this issue).


All modified faunal materials were removed and separated
from zooarchaeological samples and organized by general
material category (e.g., bone, antler, tooth, shell). Dr. Ro-
chelle Marrinan (Marrinan 2005, this issue) identified the
species and element of most of the bone artifacts. Bone
specimens not analyzed by Marrinan and all the shell artifacts
were identified by the author, based upon morphological
characteristics. Anatomical terminology was utilized when the
landmarks on bone, antler, teeth or shell were identifiable.

Other physical attributes noted were the presence of designs,
exfoliation (i.e., loss of the outer surface), root etching (i.e.,
scars from acids released by plants), shape, and completeness
(Klein and Cruz-Uribe 1984; Loy and Powell 1977; Semenov
1964). Each artifact was assigned to an artifact type category
(e.g., antler tine flaker, bone pin, shell bead, etc.) based on
faunal species, element, and other morphological characteris-
tics, as described in Penders (1997:35-39). Table 1 and 2
provide a description of the morphological terms used during
analysis (Campana 1989; Fenner 1974; Kelly et al. 1981; Loy
and Powell 1977; McComb 1989).
For most artifacts, shape was considered for three main
sections: base, shaft (or body), and tip (or tang or end). Base
is the bottom or proximal end of a tool and it is the point of
articulation where the object attaches to a handle or body of
another component. The tip is the distal end of a pointed tool
that is modified for use and may have the highest amount of
wear. The shaft (or body) of an object is the main portion
located between the tip and the base.
Faunal artifacts were analyzed for gross modification,
profile, and cross section. Gross modification refers to the


VOL. 58(3-4)




Figure 1. Excavation areas at the Shields Site (adapted from Ashley 2003).

general alteration of the skeletal element. Profile is the shape
of the entire object from a side view. Cross-section is the
shape of an object when viewed from one or more ends.
Profile and cross-section morphologies use the same descrip-
tive terminology (see Table 2).
The length, width, thickness and diameter of each artifact
was measured in millimeters using Muyoto digital calipers.
Length (or x-axis) is the measurement of greatest dimension
or along the line of intended alignment (not naturally since
they are artifacts). Width (or y-axis) is the second largest
measurement and is taken from the sides. Thickness (z-axis)
is typically the smallest measurement. Each measurement is
typically at right angles to the other two.
Microscopic analysis focused on striation patterns and
polish from manufacture and use. This analysis was conducted
using a Bausch and Lomb binocular microscope with a
minimum magnification of 4X and maximum of 1 IX. The
light source consisted of a dual fiber optic light that allowed
the adjustment of light angles and intensity. Movement of the
fiber optic lights was crucial to the identification of fine
striations, visible only if the light is at a low, sharp angle to
the object and at maximum brightness. This relationship
creates shadows permitting the striations to be viewed easily.

The location and direction of these alterations were recorded,
since it has been demonstrated that the location of striations or
polish is indicative of use (Campana 1989; McComb 1989).
In addition, wear also can obscure evidence of manufacturing
on the artifact surface.
Striations can be described as longitudinal, oblique (or
diagonal), random, transverse (or perpendicular), or parallel.
Longitudinal striations are the markings on the surface of an
artifact oriented from tip to base along the long axis of a tool
that do not deviate more than 250 to 300 from the long axis.
Oblique or diagonal striations are markings oriented at an
angle approximately 30 to 600 from the long axis. In addition,
oblique striations can bisect each other to form a crosshatched
pattern or random pattern. Transverse or perpendicular
striations are lines at an angle approximately 900 from the
long axis and can deviate from 60' to 120'. Parallel striations
are typically two or more striations oriented along the same
line or direction that do not diverge more than 25' to 30".
Terms to describe striations include short, shallow, deep,
thin, and thick. Deep striations are those markings that
penetrate more than 0.5 mm beneath the artifact surface.
Shallow striations penetrate less than 0.5 mm below the
surface and appear as light scratches. Thin striations are less



2005 VOL. 58(3-4)


Table 1. Artifact Traits and Morphological Descriptions.

Trait Description
Whole, modified Complete skeletal element altered and utilized
Whole, Complete skeletal element not altered but utilized
Half split Element that has been split into two pieces; typically result of primary groove and
Quarter split Element that has been split into four separate pieces; result of secondary groove
and splintering
Perforated Hole in surface created by drilling or coring
Grooved Object with a cut or series of cuts less than 1 mm. wide; typically deep and show
relief under magnification
Notched Deep, broad cut or series of cuts greater than 1 mmun
Socketed A hole created for hafting; usually found at the base
Indeterminate Object fragmented and modification cannot be determined

than 1 mm wide, and thick striations are more than 1 mm
wide. The terms long and short striations are subjective and
based on their relationship to tool size and each other. How-
ever, Chomko (1975:29) has defined long striations as being
4 mm or longer and short striations are less than 4 mm.
Polish can be identified by the presence of very fine minute
striations over the entire surface of an object. Polish from use
as a tool appears as a smooth glossy surface and is classified as
unintentional and intentional. Unintentional polish is not
localized and may encompass the entire artifact, but inten-
tional polish is typically localized. Polish from manufacture is
typically found on all surfaces of an object. The entire surface
will have a smooth glossy appearance and may obscure any
traces of the manufacturing striations (Campana 1989:63,65;
McComb 1989:18; Chomko 1975:29; Penders 1997:41-42).

Bone, Tooth, and Antler Artifacts

Eighty-five bone, tooth and turtle shell artifacts were recov-
ered, representing ten classes. These type names, in many
cases, indicate artifact function. Faunal material that does
exhibit modification (e.g., burned, butchering marks, etc.) but
has no evidence of tool use or function and was not indetermi-
nate was placed in a general category and briefly described.

Bone Pins

Forty-nine pins (or pin fragments) were recovered from the
Shields site, which represents 57.7% of the identified bone
artifact types (Table 3). Pins are present in the archaeological
record throughout North and Central America (e.g. Lewis and
Lewis 1961:75-81; Chomko 1975, n.d; Dalley 1970:106-113;

Gilliland 1975; Hudson and Blackburn 1985:200,244-245;
Fenner 1974:40-49; Olsen 1979:354-357, 365-370; Wheeler
1994). Pins were used as hair or clothes fasteners, sewing tools
or as general perforation devices (Loy and Powell 1977:42;
Olsen 1979:355, 1980:49-53; Fenner 1974:40-49). Pins from
the nearby Goodman Mound site were recovered from the head
area in several burials suggesting their use as hairpins (Jordan
1963). Walker (1989, 1992:229-232) has speculated that some
objects classified as pins were actually a type of projectile point
related to fishing.
All pins from Shields were slender pieces of large mammal
long bone (possibly deer metapodials) that were cut, scraped,
and polished. They all exhibit longitudinal striations typical of
the manufacture of bone tools and were identified by the
presence of polish over 75% or more of the surface. Replica-
tion analysis of other artifact assemblages by the author and
other researchers consistently report localized polish, abra-
sions, or discoloration as a result of use on morphologically
similar tools such as awls, projectile points, or fishing tools
(Penders 1997:41-42, 100; Mitchell 1995; Campana 1991;
Chomko n.d; 1975; Olsen 1979:355, 1980:49, 53). Of the 49
bone pins from Shields, 35 (71.4%) are highly polished or
exhibit polish suggesting use not typical of patterns observed
in fishing or fabric working tools. Thus, the pins from Shields
may represent hairpins or clothing fasteners (Figure 2). Eight
pins (16.3%) have localized polish, suggesting use as a tool,
but due to their fragmentary nature exact function is undeter-
mined. Six pins (12.2%) are either burned or too eroded to
determine the presence or absence of polish; their fragmented
condition further obfuscates interpretation of use.
Five pins or pin fragments were either morphologically
unique or decorated (Figure 3). One specimen (15.001) was




Table 2. Artifact Profile and Cross Section Description Traits.

Trait Description
Straight Profile or cross section which does not curve or bend; has two or more linear
Tapered Profile or cross section that decreases gradually on one surface
Double tapered Profile or cross section that gradually decreases on two surfaces and may form
a point
Spatulate Flares out to a broad square-like surface
Bio-convex Two convex curves that meet to form a oval or round-like profile
Plano-convex One flat side and one curved side that meet
Concave Curved profile or cross section
Beveled Surface or plane that sharply slopes and forms a sharp edge
Double beveled Two surfaces or planes that meet to form a sharp edge
Pointed Distal or proximal end (or both) that forms a point from tapering or beveling
Symmetrical Two or more surfaces identical or equal in form
Asymmetrical Two or more surfaces that are not identical or equal in form
Rounded Circular or oval in shape
Unmodified Not altered and has the elements natural shape
Indeterminate Fragmented and unable to determine cross section or profile
Rectangular Has a shape rectangular in form; has four sides and all the angles are
approximately 900
Triangular Has a shape triangular in form; has three sides and at least two angles are less
than 900
Irregular Cross section that has no distinctive shape
Hollow Object drilled or cored completely through

identified as a cap-headed pin (Figure 3). Its surfaces were
highly polished, its cross-section is circular, and its base was
carved to form a cap. Another specimen (189.001) is a pin
fragment with a concave cross-section. The convex side is
decorated with a series of four curvilinear lines with a circle in
the center. Two lines were on each side of the circle. This
decoration may represent an eye or bird head (Figure 3).
Specimen 204.001 is an almost complete tenon-headed pin
(Figure 3). It has a stem extending from the body and is flat
rectangular in cross section. Its surfaces are highly polished.
The stem or tenon that extends from the shaft often noted on
bone pins from sites in Florida and the Southeast. Examples of
this type of pin were recovered from the nearby Goodman
Mound site in Jacksonville (Jordan 1963). The stem at the base
also may have been used to attach feathers or other decorative
items. Tenon-headed pins have been associated with rings of
bone or shell called ferrules. deBry engravings based on
LeMoyne watercolors from the French explorations of the New

World American Indians wearing feathers and feather holders
(Jordan 1963:43; Wheeler and Coleman 1996:50).
Specimen 227.001 was broken into two sections and
represents the base and shaft of a pin (Figure 3). Both sections
are highly polished and are decorated. The base has a series of
cross hatched lines. The shaft contains a set of two curvilinear
"ladders". These ladders have a pair of incised lines oriented
diagonally to the longitudinal axis of the shaft. Between the
pairs of diagonal lines are short incised lines oriented perpen-
dicular to the longitudinal axis. Finally, specimen 335.001 is
a pin shaft containing a set of three curvilinear "ladders"
(Figure 3). These ladders have a two pair of incised lines
oriented diagonally to the longitudinal axis of the shaft.
Between the inner pairs of diagonal lines on each "ladder" are
short incised lines oriented perpendicular to the longitudinal
axis. Wheeler (1992:45, 112) has described the decorative
geometric motifs from sites in these areas as "Peninsular
Geometric Style" of the St. Johns and Indian River Area.


2005 VOL. 58(3-4)

University of Florida Home Page
© 2004 - 2010 University of Florida George A. Smathers Libraries.
All rights reserved.

Acceptable Use, Copyright, and Disclaimer Statement
Last updated October 10, 2010 - - mvs