Front Cover
 Front Matter
 Title Page
 Editor's page
 Northeastern Florida Swift Creek:...
 A Suwannee/Bolen artifact assemblage...
 Osteological analysis of the Manasota...
 The case for Swift Creek paddles...
 Book reviews
 About the authors
 Back Cover

Group Title: Florida anthropologist
Title: The Florida anthropologist
Full Citation
Permanent Link: http://ufdc.ufl.edu/UF00027829/00193
 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: VID00193
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
    Front Matter
        Front Matter
    Title Page
        Page 1
        Page 2
    Editor's page
        Page 3
        Page 4
    Northeastern Florida Swift Creek: Overview and future research directions
        Page 5
        Page 6
        Page 7
        Page 8
        Page 9
        Page 10
        Page 11
        Page 12
        Page 13
        Page 14
        Page 15
        Page 16
        Page 17
        Page 18
        Page 19
        Page 20
    A Suwannee/Bolen artifact assemblage from the Santa Fe River
        Page 21
        Page 22
        Page 23
        Page 24
        Page 25
        Page 26
        Page 27
        Page 28
        Page 29
        Page 30
        Page 31
        Page 32
        Page 33
        Page 34
    Osteological analysis of the Manasota period Dunwody site
        Page 35
        Page 36
        Page 37
        Page 38
        Page 39
        Page 40
        Page 41
        Page 42
        Page 43
        Page 44
        Page 45
        Page 46
        Page 47
        Page 48
        Page 49
        Page 50
        Page 51
        Page 52
        Page 53
        Page 54
    The case for Swift Creek paddles as totemic symbols: Some anthropological considerations
        Page 55
        Page 56
        Page 57
        Page 58
        Page 59
        Page 60
        Page 61
        Page 62
    Book reviews
        Page 63
        Page 64
        Page 65
        Page 66
        Page 67
    About the authors
        Page 68
    Back Cover
        Back Cover
Full Text





THE FLORIDA ANTHROPOLOGIST is publishedby 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: Dan Hughes, 2301 8th Ave. North, St. Petersburg, Florida 33713 (hughesarc@juno.com)
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
Bulk Mail: 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 0
Volume 59 Number 1 S"VCE 19 1
March 2006


Editor's Page 3
Northeastern Florida Swift Creek:
Overview and Future Research Directions. Keith H. Ashley and Neill J. Wallis 5

A Suwannee/Bolen Artifact Assemblage from the Santa Fe River. David K. Thulman 21

Osteological Analysis of the Manasota Period Dunwody Site. Melissa L. Gold 35

The Case for Swift Creek Paddles as Totemic Symbols:
Some Anthropological Considerations. Neill J. Wallis 55


Fagan: The Great Journey: The Peopling ofAncient America. Michael Wisenbaker 63

Romans: A Concise Natural History ofEast and West Florida. Simon Barker-Benfield 64

Schafer: Anna Madgigine Jai Kingsley:
African Princess, Florida Slave, Plantation Slaveowner. Simon Barker-Benfield 66
Authors 68

Cover: Swift Creek design reconstruction from a Mayport Mound vessel (enlarged). Drawing by Neill J. Wallis (see article
beginning on page 55 for more information).

Published by the
ISSN 0015-3893


This issue begins and ends with articles about the Swift
Creek culture in northeastern Florida. In the first article Keith
Ashley and Neill Wallis present a synthesis of Swift Creek in
northeastern Florida. Ashley and Wallis consider chrono-
logical issues, present known radiometric dates, site types,
relationships with neighboring areas, design contacts and
questions for future research. Researchers working in the area
will appreciate this synthetic article.
The second article, by Dave Thulman, presents an analysis
of Paleoindian and Early Archaic stone tools collected from
the Santa Fe River by river diver and collector Don Monroe.
Thulman considers the manufacture, chronology and function
of the tools, as well as how they came to be deposited in the
river. Thulman notes that the artifacts were collected as part
of the state's Isolated Finds Program, which was discontinued
in June 2005. Readers are referred to the Division of Histori-
cal Resources's webpage for more information on this pro-
gram: www.flheritage.com/archaeology/underwater/finds/

The third article is Melissa Gold's study of skeletal
remains from the Manasota Period Dunwody site. Gold's
paper adds to the knowledge of health, diet and disease in
populations of the Florida Gulf Coast and also includes a table
of radiocarbon dates from sites around Lemon Bay.
The fourth and final article is by Neill Wallis and focuses
on the interesting study of paddle design matching on Swift
Creek pottery. Wallis's paper was the winner of the student
paper prize at the 2005 Florida Anthropological Society
meeting in Gainesville.
This issue also includes three book reviews by Mike
Wisenbaker and Simon Barker-Benfield. We hope that all
readers find something of interest!

March, 2006


VOL. 59(1)


MARCH 2006



SDepartment ofLiberalArts, Savannah College ofArt and Design, Savannah, GA 31402-3146
E-mail: kashley@scad.edu

2Department ofAnthropology, University ofFlorida, Turlington Hall, Gainesville, FL 32611
E-mail: njwallis@ufl.edu

John Goggin (1952) was among the first to recognize that
the complicated stamped wares recovered by C.B. Moore and
others from sand mounds along the lower St. Johns River
belonged to the Swift Creek pottery series. Indeed, today all
it takes is a quick glance at C.B. Moore's (1894, 1895) sherd
and vessel illustrations by anyone familiar with pottery of the
Southeast to reach the same conclusion. Because few midden
excavations had been conducted in the lower St. Johns region
at the time, Goggin (1952:49-50, 70, 106) assumed that St.
Johns was the dominant Woodland period ware in the region
and that Swift Creek Complicated Stamped was not among the
"local forms of pottery" and that it was a "trade ware" re-
stricted to mound contexts. Sears's (1957, 1959) midden
testing later revealed that the occurrence of Swift Creek
pottery in northeastern Florida was not exclusive to burial
mounds, and that it was part of the local Woodland period,
sand-tempered ceramic complex. Subsequent work over the
next half century has demonstrated the dominance of Swift
Creek pottery at numerous mound and midden sites in north-
eastern Florida (e.g., Ashley 1992, 1995, 1998; Russo 1992).
Only within the past decade has our temporal understand-
ing of Swift Creek manifestations in northeastern Florida
taken a significant leap forward, with the acquisition of
numerous radiometric dates. What these dates tell us is !that
Swift Creek pottery was manufactured over a longer period of
time than originally thought and that it was produced well into
the ninth century A.D. along the Atlantic coast. As defined in
this paper, in northeastern Florida Early Swift Creek dates to
A.D. 300-600, whereas Late Swift Creek dates to A.D. 600-
850. Based on the evidence at hand, the early style sand-
and/or charcoal-tempered wares appear to have given way to
the late variety tempered with sand or grit some time during
the sixth century. With temporal refinement in place, now is
an opportune time to move beyond mere ceramic chronology
and culture history, and pursue more sophisticated and
nuanced questions of Swift Creek political economy and
cultural construction.
The complicated designs indelibly stamped on Swift Creek
pottery offer a unique opportunity not only to study contact
between groups, but also to explore social relations and
processes of cultural creation and negotiation through interac-
tion. Culture, as we see it, is not fixed and precisely bound, but

rather fluid and subject to continual reproduction and transfor-
mation at multiple scales of interaction (Wolf 1982). The
primary aim of this paper is to present new information
generated within the past decade, including the calibrated
results of 18 radiometric assays, and to formulate a new
synthesis of Swift Creek occupations in northeastern Florida
(Figure 1). In addition, we attempt to draw attention to issues
of interaction and identity formation that extend beyond the St.
Johns River region and lay a foundation for future research.

What is "Swift Creek?"

Before launching into a discussion of northeastern Florida
Swift Creek, a brief statement is necessary about our use of the
term Swift Creek. In a recent edited volume on Swift Creek
manifestations in the Southeast, the term Swift Creek was
applied in a variety of ways (Williams and Elliot 1998). In
some instances it was used to refer to an archaeological
culture, and in other cases it merely designated a pottery type.
We employ it generally to refer to Woodland-era peoples who
made and used Swift Creek Complicated Stamped pottery. In
this sense, Swift Creek is not a monolithic entity or single
Woodland culture, but rather denotes a pottery style that
covered much of the lower Southeast, yet had various regional
manifestations (Anderson 1998:275). Its popularity spread
through communication and interaction networks that linked
many Woodland groups, each of which had their own histories
and cultural traditions. Such webs of interconnectedness
undoubtedly led groups to share other cultural features, giving
Swift Creek a semblance of pan-regional cultural equivalence.
But what constitutes Swift Creek in northeastern Florida is not
the same as elsewhere in the Southeast.

Early vs. Late Swift Creek in Northeastern Florida

From a Southeastern perspective, the Swift Creek tradition
spans a lengthy period from about A.D. 100 in northwestern
Florida to A.D. 850 in northeastern Florida and southeastern
Georgia (Williams and Elliot 1998; Stephenson 2002;
Stephenson et al. 2002). However, its actual period of produc-
tion and use varied throughout the broad region. Most
archaeologists partition Swift Creek pottery into early and late




MARCH 2006

VOL. 59(1)


Figure 1. Location of Swift Creek sites mentioned in text, northeastern Florida and southeastern Georgia.

(and in some areas middle) varieties, based on certain ceramic
attributes such as rim and lip form as well as quality of design
execution and application (e.g., Kelly 1938; Kelly and Smith
1975; Phelps 1969; Willey 1949). This is true for northeastern
Florida, where both Early and Late Swift Creek Complicated
Stamped pottery have been recovered.
At present, 18 radiometric dates in the form of radiocarbon
dates on shell and AMS dates on soot have been obtained from
Swift Creek contexts in northeastern Florida (Table 1).
Specifically, the assays have come from five shell middens
(8DU81, 8DU5545, 8NA32, 8NA709, and 8NA910) and two
burial mounds (8DU68, 8DU96). As discussed below, the
earliest date from the Dent Mound (8DU68) appears to be an
outlier, but the remainder forms a consistent grouping that
extends from ca. A.D. 300 to 850. This time span dovetails
nicely with the earlier Deptford date range of 500 B.C. A.D.
300 and the later Colorinda phase currently dated to A.D. 850-
900 (Ashley 2003).

Early Swift Creek (A.D. 300-600)

In northeastern Florida, the Early Swift Creek assemblage
consists of locally produced plain and complicated stamped
wares, with rim/lip treatments that include simple round and
flat as well as hallmark forms such as notched, nicked,
scalloped, and crenulated (Ashley 1992:130-131, 1998:204;
Sears 1959:155). These wares are often charcoal-tempered,
although sand temperingalso is common. Charcoal-tempered
pottery-variously referred to as "limestone-tempered" and
"hole-tempered"' by Sears (1957, 1959) and Wilson
(1965)-contains quartz sand with charcoal inclusions that
range from common to sparse. Recent petrographic analysis
suggests that predominantly charcoal, rather than uncharred
wood, was added as temper to most charcoal-tempered vessels,
with identified species including pine, cedar, cypress, and
sassafrass (Wallis et al. 2005). In general, most charcoal
particles tend to exhibit little shrinkage from vessel


2006 VOL. 59(1)

Table 1. Calibrated radiocarbon assays for Early and Late Swift Creek contexts in northeastern Florida.

Measured C13/C12 Calibrated
C14 age ratio Conventional 1 Sigma (AD) Calibrated
Site Beta # Material (BP) (o/oo) C14 age (BP) with intercept 2 Sigma (AD) Reference

Early 8DU68 182333 Soot 1930 + 40 -24.2 1940 + 40 30 (65) 95 30 BC 135 Stephenson 2002
Swift 8DU68 182332 Soot 1690 +40 -24.7 1690 +40 330 (385)410 250 430 Stephenson 2002
Creek 8DU96 168177 Soot 1560 +40 -24.5 1570 +40 430 (460, 480, 520) 540 410-580 Stephenson 2002
8DU96 190255 Soot 1510 +40 -25.3 1510+40 530 (560) 610 440 640 Wallis 2004
460 480 and
8DU96 169421 Soot 1450 +40 -22.8 1490 +40 540 (580) 620 520 -560 Stephenson 2002
8DU68 169420 Soot 1330 +40 -18.4 1440 +40 600 (580) 650 550 660 Stephenson 2002
8DU81 181303 Oyster 1460 + 70 -3.6 1810 +70 540 (610) 670 440 710 Ashley 2003
8DU5545 163598 Oyster 1390 + 60 -3.2 1390 + 60 600 (660) 690 540- 740 Smith and Handley 2002
8DU5545 163597 Oyster 1350 + 60 -2.3 1350 + 60 620 (670) 700 560 770 Smith and Handley 2002
Late 8NA32 190666 Oyster 1340 + 60 -1.6 1340 + 60 640 (680) 720 580 -780 Handley et al. 2004
Swift 8DU5545 168176 Soot 1290 + 40 -24.5 1300 + 40 670 (690) 770 660- 790 Stephenson 2002
Creek 8NA32 190665 Oyster 1310 +60 -1.8 1310 + 60 660 (700) 770 590 850 Handley et al. 2004
8DU68 182334 Soot 1270 + 40 -25.0 1270 + 40 685 (720, 745, 760) 780 670 -870 Stephenson 2002
8DU81 182335 Soot 1250 + 40 -25.4 1240 + 40 705-815 (775) 840-855 680 -885 Stephenson 2002
8DU68 54645 Oyster 1250 +70 -2.7 1610 + 70 695 (775) 865 655 -955 Ashley 1995
8NA709 126313 Oyster 1180 + 60 0.0 1590 +60 730 (800) 875 685 -945 Dickinson and Wayne 1999
8NA910 159965 Tagelus 1120 + 60 0.0 1540 + 60 780 (870) 920 720 -1000 Hendryx and Smith 2001
8NA910 159964 Soot 1150 +40 -23.4 1180 +40 790 (880) 900 770 -970 Hendrvx and Smith 2001


Figure 2. Charcoal-tempered sherd, v. 29, Mayport Mound, exterior (left) and interior (right) views (Courtesy of Florida
Museum of Natural History).

firing and cooking, while other pieces appear to have been
burned to ash leaving holes or voids visible on the vessel
surface (Figure 2) (Ashley 1998:202; Russo et al. 1993:35-36).
As observed in petrographic thin section, the prevalence of
charcoal, crushed bone, and grog inclusions in charcoal-
tempered sherds supports the possibility that these materials
were hearth contents added to the clay during paste prepara-
tion. In addition, the consistent size of temper particles,
predominantly ranging from medium to course on the Wen-
tworth scale, suggests that charcoal was consistently pounded
and sieved before being added to the raw clay (Wallis et al
2005). In both mounds and middens, plain charcoal-tempered
wares outnumber complicated stamped versions, but sand
tempered plainwares represent the dominant ware produced
during Woodland times (Ashley 1998:200; Russo 1992:115;
Sears 1957:29).
Designs associated with Early Swift Creek pottery tend to
be sloppy and unlike the Late Swift Creek style of southern
Georgia (Frankie Snow, personal communication, 2002). This
unique Early Swift Creek variant is restricted mostly to sites
along the lower St. Johns River, although its range seems to be
expanding as more research continues in the area. Recent
excavations have yielded trace amounts of charcoal-tempered
pottery as far south along the St. Johns River as the River
Point site (8SJ4790) near the community of Switzerland, as far
south along the coast as the Shannon Road Midden site
(8SJ3169), and as far north as the Sadlers Landing site

(9CM233) in Camden County, Georgia (see Figure 1)
(Hendryx et al. 2003; Kirkland 2003; Southeastern Archaeo-
logical Research 2005). While the core production area for
charcoal-tempered pottery was restricted to areas near the
mouth of the St. Johns River, its currently known distribution
stretches from Camden County, Georgia to St. Johns County,
Florida, a straight-line distance of about 90 km (60 miles).
Prior to a few years ago, no Early Swift Creek contexts in
northeastern Florida had been radiometrically dated. As a
result, charcoal-tempered pottery was cross-dated to ca. A.D.
100-300 based on Early Swift Creek dates in northwestern
Florida (Ashley 1998:203-206). In an attempt to anchor the
production of northeastern Florida Early Swift Creek in time,
radiometric assays were obtained on soot from charcoal-
tempered plain and complicated stamped vessels from the Dent
(8DU68) and Mayport (8DU96) mounds. Calibration of these
assays suggests a time span of circa A.D. 30 to 600 for its
manufacture (Table 1). However, because the earliest date
from the Dent Mound-A.D. 30-95-is three centuries earlier
than the next earliest date, we believe this AMS date should be
considered with caution. As an anomalous outlier, this early
AMS date on soot may be a result of the old-wood problem, in
which the wood used as fuel was much older than the activity
(e.g., cooking) that caused soot to adhere to the vessel surface.
At this time we are not stating that the date is unassociated
with the production of charcoal-tempered pottery, merely that
it is problematic. If we tentatively eliminate this dubious date,


2006 VOL. 59(1)


Early Swift Creek in northeastern Florida appears to have
emerged around A.D. 300.
Based on contextually secure dates, the production of
charcoal tempered pottery seems to have ended during the
sixth century. At the McArthur Estates site (8NA32), how-
ever, radiocarbon dates on oyster shell from two features
containing charcoal-tempered pottery suggest an even later,
late seventh century A.D. date for its production (Handley et
al. 2004). Combined, however, both features yielded a total of
2 charcoal-tempered sherds, both of which contained charcoal
inclusions that made up a low percentage of the matrix. The
dominance of sand-tempered Swift Creek wares in the
McArthur Estates assemblage suggests that the radiometric
assays actually date the site's Late Swift Creek component. In
accordance with more reliable AMS dates from sherd soot, our
conservative estimation is that manufacture of charcoal-
tempered pottery ceased prior to A.D. 600.
Charcoal-tempered pottery has been recovered from both
shell middens and burial mound contexts. In the former, it is
often found associated with sand-tempered plainwares but
rarely discovered in direct association with Late Swift Creek
pottery. However, as stated above, contexts at the McArthur
Estates site (8NA32) may prove an exception (Handley et al.
2004). In burial mounds, Early and Late Swift Creek wares
frequently occur together-though in separate
contexts-lending robust support for sequential production.

Late Swift Creek (A.D. 600-850)

Late Swift Creek pottery has a much broader distribution,
and is found on sites in northeastern Florida, along the
Atlantic seaboard of Georgia as far north as the mouth of the
Altamaha River, and within the hinterland of southern-central
Georgia. With regard to temper, Swift Creek wares from sites
on the northern-most Florida coastal islands (e.g., Amelia
Island, Martin Island) and along the Georgia coast are mostly
sand/grit or grit-tempered, whereas in the lower St. Johns
region the ware is mostly fine sand-tempered (e.g., Ashley
1998:206; Dickinson and Wayne 1999:50; Espenshade
1985:304; Hendryx and Smith 2001:28; Hendryx et al.
2000:40; Smith et al. 1985:170-171). Both complete vessel
stamping and zone stamping have been identified (Figure 3).
Rim treatments associated with Late Swift Creek vessels
include hallmark rim folds and simple rounded lips that are
straight, incurvate, and excurvate. Weeden Island types, such
as Weeden Island Incised-Punctate, Keith Incised, Carabelle
Punctate, and Crooked River Complicated Stamped are
recovered in small numbers at some sites, but undecorated
wares typically dominate, particularly in middens. Along the
Atlantic coast, from the St. Johns River to the Altamaha River,
Late Swift Creek contexts have been radiometrically dated
consistently to between A.D. 500 and 850 (Ashley 2003:76-84;
Stephenson 2002; Stephenson et al. 2002:337).
In addition to the common fine sand tempering of Late
Swift Creek wares in northeastern Florida, two infrequent
variants deserve mention. First, bone tempering, which
typically consists of bone crushed into fine-sized particles, has

been observed in limited amounts in direct association with
classic, sand-tempered Late Swift Creek sherds. These bone
particles are unburned and larger than those noted in associa-
tion with charcoal-tempered vessels. At the McArthur Estates
Site, several bone-tempered sherds were recovered from a Late
Swift Creek feature dated to A.D. 640-720 (Handley et. al.
2004). In addition, apparent vessel and rim forms identify this
temper type as contemporaneous with Late Swift Creek.
Second, grog-tempering has been observed in Late Swift Creek
wares, almost always crushed into fine particles, sometimes
unnoticed without the aid of a microscopic lens (Florida
Archeological Services 1994:158; Johnson 1998b:36-37).
According to Cook (1979), grog-tempering is a characteristic
of the Late Swift Creek wares he classified as Kelvin. The use
of grog in some Late Swift Creek vessels may signify a
precursor to the conspicuously large St. Johns grog particles of
the subsequent Colorinda period.
To recap, the exact relationship between Early (including
charcoal-tempered) and Late Swift Creek pottery in northeast-
ern Florida is presently unclear, but their co-occurrence in
continuous use mounds and a few midden sites suggests that
local production of charcoal-tempered pottery eventually gave
way to the manufacture of sand-tempered wares some time
during the sixth century A.D. In fact, we should expect their
co-occurrence to some degree if there was an in situ develop-
ment of Early Swift Creek into Late Swift Creek in northeast-
ern Florida. The change from Early to Late Swift Creek
pottery may reflect a shift in the directionality of interaction
networks emanating from northeastern Florida, with early
routes having a more western-focus and later interactions
skewed to the north (Ashley 1998). The presence of Weeden
Island wares (or local copies) on Late Swift Creek sites
suggests that western routes were not abandoned.

Waning Late Swift Creek

An area of investigation that we have begun to gain ground
on is the fading years of Swift Creek pottery production along
the Atlantic coast. In Georgia, waning Late Swift Creek has
been referred to as the Kelvin Phase by Cook (1979, 1981),
based on his work on St. Simons Island. Pottery of this phase
is thought to differ from Late Swift Creek in several notable
ways. There is a decline in the workmanship and application
of the paddle stamped vessel designs, and several key motifs,
such as the barred snowshoe, are absent during the Kelvin
Phase. Common Kelvin Phase motifs include "spirals,
concentric circles, rectangular motifs, and chevrons," and
incising and punctating appear as common forms of surface
decoration (Cook 1979:84-85). Also grog tempering appar-
ently became more common, and rim folding became less
common, replaced by "unfused rims or a false rim fold" (Cook
1979:77). In the absence of radiocarbon dates, Cook (1979)
tentatively assigned the Kelvin Phase to A.D. 600-900.
Several of the pottery traits described by Cook for the
Kelvin Phase aptly characterize pottery from a few Late Swift
Creek sites in northeastern Florida, most notably the Honey
Dripper Site on Amelia Island, where nested chevron motifs



Figure 3. Characteristic Late Swift Creek vessels from the Mayport Mound. Zoned stamping (left), and complete stamping with folded rim (right) (Courtesy
of National Park Service).


outnumbered curvilinear designs 2.1 to 1 (Hendryx 2004:303).
The former were typed as Crooked River Complicated
Stamped, whereas the latter were described as poorly stamped
and lacking known paddle designs. No rim folds were
recovered and a partly reconstructed Crooked River vessel
displayed an incision below the lip giving the impression of a
fold (i.e., false fold). Also recovered were a small number of
incised and punctated sherds classified as Weeden Island.
Two calibrated radiocarbon assays date the site to ca. A.D.
780-920, and these presently represent the most recent (i.e.,
youngest) radiocarbon dates for Late Swift Creek on the
Atlantic coast (Hendryx 2004:306; Hendryx and Smith
A perusal of CRM and other reports shows that sherds
exhibiting nested chevron designs often typed as Crooked
River-though Sears (1957:27, Plate II:I) appears to have
mistyped the zigzag motif of Crooked River as St. Andrew
Complicated Stamped-are persistent minority wares on many
(and a dominant ware on a few) Swift Creek sites along the
Atlantic coast. Although more radiometric dates are needed
to substantiate this claim, Crooked River Complicated Stamp-
ed pottery may eventually serve as a reliable time marker for
waning Late Swift Creek along the Atlantic coast.
Cook's data from St. Simons Island, along with the
archaeological evidence from the Honey Dripper site, suggest
a breakdown in the Late Swift Creek pottery-making tradition
of the Atlantic coast during the late eighth and early ninth
century A.D., with other concomitant cultural changes
presumably taking place as well. This is presumably reflective
of a Late Woodland collapse in long-distance interaction
networks impacting the broader Southeast at this time (Cobb
and Nassaney 1995; Nassaney and Cobb 1991). Though Cook
(1979:67-68) contends that Kelvin is not Late Swift Creek, we
consider it representative of waning Late Swift Creek along
the Atlantic coast.

Archaeological Site Types

To date, Swift Creek pottery has been recovered from
habitation sites that range from large shell middens to small
artifact scatters in northeastern Florida (Ashley 1992:130-133;
1998:208-218). Atintensively occupied multicomponent sites,
the nature and structure of Swift Creek settlement are often
masked and difficult to discern. However, this is not always
the case at single component sites or those with minimal
evidence of occupations other than Swift Creek. For instance,
at the Ocean Reach Site (8NA782) at least 22 individual
middens were identified, and each was thought to represent an
individual household refuse deposit (Johnson et al. 1997:56).
In addition, at the Honey Dripper Site (8NA910), excavators
hypothesize that Late Swift Creek occupations were marked by
discrete shell heaps transformed into a diffuse sheet midden
through site reuse by later inhabitants (Hendryx 2004: 308;
Hendryx and Smith 2001:63).
On the Greenfield Peninsula, along the south bank of the
St. Johns River immediately west of San Pablo Creek
(Intracoastal Waterway), numerous Swift Creek deposits have

been documented. At Greenfield Site #7 (8DU5543), both
Early and Late Swift Creek household middens were identified
(Florida Archeological Services 1994). Interestingly, no
deposits yielding charcoal-tempered pottery contained sand-
tempered Late Swift Creek wares and vice versa. Nearby at
Greenfield Sites #8 (8DU5544) and #9 (8DU5545)-actually
one large site separated for management purposes-several
Swift Creek shell middens and one sand burial mound (Mound
C) have been recorded (Johnson 1996, 1998a, 1998b; Smith
and Handley 2002; Smith et al. 2001).
The "Swift Creek Midden Area" at Greenfield site #8/9
(8DU5544/45) consisted of a non-mounded, horseshoe-shaped
shell midden along with two interior "household middens"
(Smith and Handley 2002). Similarly shaped shell middens in
association with Swift Creek pottery have been identified at
Kings Bay, Georgia and in other areas of Georgia and north-
western Florida (Bense 1998; DesJean et al. 1985; Milanich
1994:144; Saunders 1998:62-63; Stephenson et al. 2002).
These middens, at least in some instances, may reflect refuse
disposal associated with an arc-shaped arrangement of houses.
The Swift Creek Midden Area is located about 175 m west of
Greenfield Mound C; the latter is "cradled" by a "roughly
arcuate shell midden" referred to as Midden A (Johnson
1998b:22,71). Midden A was described as a variable density
shell midden that was only partly investigated since its western
extent extended outside project boundaries. Late Swift Creek
wares dominated, although a few charcoal-tempered sherds
were recovered (Johnson 1998b).
Fourteen Woodland-period mounds along the lower St.
Johns River have been previously attributed to the local Swift
Creek tradition (Ashley 1998:208-209). As stated above,
Greenfield Mound C has recently been labeled a Swift Creek
mound, making the total number 15. As described by Johnson
(1998b:64, 71), the tumulus was "roughly circular" (18 by 16
m) and stood only "30 cm above the surrounding terrain."
Because the "barely detectable rise" was initially thought to
represent a northern extension of Midden A, five 1 x 2-m units
were excavated. One unit exposed a partially articulatedburial
and others yielded a number of possible human bone frag-
ments. Although Swift Creek sherds were abundant through-
out mound fill, no burial pits or grave goods were encountered.
Johnson (1998a:71) surmised that the mound "once was higher
but may have been partially razed by earth-moving activities
probably relating to agricultural activities."
In general, the majority of Swift Creek mounds along the
lower St. Johns River yielded mostly complicated stamped and
undecorated pots, along with fewer St. Johns Plain, Dunns
Creek Red, and Weeden Island wares. St. Johns chalky wares
appear more prevalent in mounds than in coeval middens, but
they are typically always a minority ware in middens. A
conspicuous aspect of the Swift Creek Complicated Stamped
pottery from local mounds is that many exhibit evidence of use
in the form of wear and exterior soot and many had their bases
knocked out (Ashley 1995; Wallis 2004; Wilson 1965). Thus,
not all were pristine pots manufactured exclusively for burial.
Kirkland (2003:152) also noted that at the Sadlers Landing
Site in southeastern Georgia, which had a mortuary area and




possibly a mound prior to farming, "[a]ll of the Swift Creek
vessels were either heavily sooted or had fire clouding."
Many of the mounds appear to have been replete with
human burials. For example, it is estimated that the Dent
Mound (8DU68) contained more than one hundred human
interments, with males, females, and children represented
(Ashley 1995). Excavation of approximately 30 to 40 percent
of the Mayport Mound (8DU96) yielded more than !50
individuals-both adults and children (Wilson 1965:12-13).
Aside from pottery and human burials, non-local stone,
mineral, and metal artifacts or raw materials were interred in
local Swift Creek mounds (Ashley 1998:211-213). Absent are
true Hopewellian artifacts, which were in circulation in areas
of the Southeast during an earlier era. Thus, Swift Creek
groups along the St. Johns River were acquiring exotics well
after the heyday of Middle Woodland or Hopewellian interac-
tion and ceremonialism.
Local Swift Creek mounds appear to represent accretionary
cemeteries used by all members of local kin groups (Ashley
1998:213-214; Thunen and Ashley 1995:5). But what is
interesting is that in many instances-Greenfield Mound C
appears an obvious exception-no substantial Swift Creek
habitation sites have been identified adjacent to burial mounds.
Perhaps this is suggestive of a somewhat dispersed (and
mobile) settlement pattern along the coast and up the lower St.
Johns River, with kin groups congregating periodically at
meaningful locations along the river for feasting, ceremony,
and interment of the dead. Specific burial mounds may have
marked the territory and resources of specific kin groups. That
the mounds were used for ritual and mortuary for centuries
provided a time depth that may have reinforced a social
group's claim to a particular tract of land and resources.
In addition, the sheer substantiality and permanency of
mounds in comparison to the temporary dwellings of mobile
Woodland groups, may have created a conceptually fixed point
amid more fluid social systems. More than mere territorial
markers, mounds could also have conceptually represented the
permanency of ancestral lineage on a landscape that, aside
from mounds, demonstrated little evidence of intransigence.
Furthermore, if social identity was continually renegotiated
through relationships with other groups, the mound could have
legitimate lineal continuity throughout changes in group

Design Contacts

The potential to gain direct evidence of interregional
interaction lies in the unique artistic expression of each Swift
Creek design, created by pressing a carved wooden' paddle
onto the surface of a wet clay vessel prior to firing. Various
researchers, most notably Frankie Snow (1975, 1977, 1998),
have attempted to reconstruct complete designs from shards in
order to identify an individual design's unique character or
signature. By focusing on design flaws that result from artisan
mistake or cracks in the wooden paddle used to create the
design, archaeologists have been able to trace the geographic
distribution of some complicated stamped designs, indicating

the movement of people, paddles, or pots, depending on
circumstances (Saunders 1998; Snow 1998; Snow and
Stephenson 1998; Stoltman and Snow 1998).
Along these lines, Snow has identified matches for two
separate designs linking two sites in northeastern Florida with
one in southeastern Georgia (Ashley 1995:32; 1998:206-207).
Through direct sherd-to-sherd comparison, he has documented
a shared Late Swift Creek design that links the Dent Mound
near the mouth of the St. Johns River to the Lewis Creek site
near the mouth of the Altamaha River in Georgia, more than
100 km away (Ashley 1995, 1998). Recently dated soot from
the Dent Mound specimen dates this design to a one-sigma
calibrated date range A.D. 685-780. In addition, Snow has
verified a different shared design match linking the Mayport
Mound, across the St. Johns River from the Dent Mound, and
the same Lewis Creek site (Ashley 1995:32). A rim sherd
from each of these vessels is depicted in Figure 4.
Another Late Swift Creek design contact was recently
verified by Snow connecting the Browne Mound site (8DU62)
along the south side of the St. Johns River to the Schmidt site
(8SJ52) in northern, coastal St. Johns County (Figure 5). The
Browne Mound sherds were recovered nearly fifty years ago by
William Sears (1957:18, Plate 2B), whereas the Schmidt site
sherds were collected from a residential yard by the landowner.
Of the latter, one of the pottery fragments displays design wear
and well-developed paddle cracks, suggesting that this vessel
was stamped later-as the paddle aged-than the others
bearing this stamped design, including those from the Browne
Mound (Frankie Snow, personal communication, 2004).
Although the design has been only partially reconstructed,
Snow further notes that the design elements are remotely
similar in general layout to a reconstructed design from a
sherd recovered from the Sadler Landing site (9CM233) in
southeastern Georgia. This suggests that an unknown theme
is common to both the Brown Mound-Schmidt site design and
the Sadler Landing site design.
Although not confirmed via direct comparison, the design
on an Early Swift Creek pot (Vessel 13) from the Dent Mound
is strikingly similar to one from the nearby Alicia Mound and
illustrated by C.B. Moore (1895:plate 80). Soot from Dent
Mound Vessel 13 was AMS dated to A.D. 330-410 (see Ashley
1995:31 for depiction of Vessel 13 design and comparison to
Moore's sherd). Other sherds illustrated by Moore from the
Alicia Mound have folded rims clearly indicating a Late Swift
Creek mound component. Thus, the Dent, Alicia, and
Mayport mounds were used concurrently over a long period of
time, with Dent and Mayport demonstrating post-A.D. 700
links to groups along the Altamaha River.
To the north, in southeastern Georgia, Kirkland (2003:153-
156) with the assistance of Frankie Snow, have linked 7
complicated stamped Swift Creek designs from the Sadlers
Landing Site (9CM233) to identical paddle designs from 10
other sites in Georgia, including one from Kings Bay in
Camden County, one near the eastern edge of the Okefenokee
Swamp, two along the Satilla River, and six in the Ocmulgee
Big Bend region. In addition, Snow has confirmed several
design contacts between sites in Kings Bay and sites in the


2006 VOL. 59(1)


Figure 4. Late Swift Creek paddle match, Lewis Creek, GA, and Mayport Mound, FL (Courtesy of Frankie Snow and
Florida Museum of Natural History).

Ocmulgee Big Bend region (Kirkland 2003:153). We feel
confident that more Late Swift Creek Complicated Stamped
designs from northeastern Florida can be drawn into the
southeastern and south-central Georgia network once con-
certed efforts are made in detailed design analysis by archaeol-
ogists. Clearly, Late Swift Creek communities along the lower
Atlantic seaboard were actively engaged in intergroup interac-


Considering the broader world of Swift Creek pottery
makers, how might we begin to interpret Swift Creek manifes-
tations in northeastern Florida? Swift Creek pottery, along
with other cultural traits, appears simply to have been in vogue
along the Atlantic coast during Middle and Late Woodland
times, presumably the consequence of intensive interactions.
The presence of numerous small burial mounds containing
Swift Creek pottery sets the lower St. Johns region apart from
areas of southeastern coastal Georgia, where few have been
reported. However, while no mound was evident during his
investigation, Kirkland (2003) reports that one may have once
existed at the Sadlers Landing site in Camden County, where
burials along with exotic items such as mica were unearthed.
The Swift Creek mounds in northeastern Florida seem to
represent a continuation of and elaboration upon a mound
building and sand-tempered pottery making tradition that had

long been a part of local Woodland period life (Thunen and
Ashley 1995:3-5).
While Early Swift Creek pottery is found in appreciable
quantities in mounds and some middens in northeastern
Florida, it infrequently occurs on sites in southeastern and
southern-central Georgia. Moreover, complicated stamped
designs on these early wares tend to be unlike those stamped
on Late Swift Creek wares along the Atlantic coast and
Georgia hinterland in style and workmanship. Charcoal
tempering further distinguishes this type from coeval types to
the north. But vessel forms and rim/lip styles, such as notch-
ed, nicked, and scalloped, are clearly Early Swift Creek in
style. It has been suggested that the Early Swift Creek pottery
tradition became vogue in northeastern Florida as a result of
transpeninsular contacts that brought exotic mortuary goods
and ideas from the Gulf coast area of northwestern Florida
(Ashley 1998:218-219). To date, however, no design contacts
linking the two areas have been identified, although vessel
designs from the Block-Sterns site (8LE148) in Tallahassee
are grossly similar to those from northeastern Florida (see
Jones and Tesar 1996:440-450).
By A.D. 500, local versions of Swift Creek pottery were
being manufactured in southern-central and southeastern
Georgia (Stephenson 2002; Stephenson et al. 2002). We
suggest that as local populations became more involved in the
trafficking of exotics and interacting with Swift Creek pottery
making peoples to the north, they appropriated Swift Creek




Sl)I :2

Figure 5. Paddle match between the Schmidt site (8SJ52) and the Browne Mound (8DU62) (Courtesy of Florida Museum
of Natural History). Reconstructed design (not to scale) Frankie Snow.

pottery iconography and applied them to the surface of their
sand-tempered wares. On most sites in northeastern Florida
and southeastern Georgia, plain sherds occur more frequently
than their complicated stamped counterparts (Russo
1992:115). In this way a Swift Creek veneer was cast over
northeastern Florida and southeastern Georgia, but one with
local histories. The sharing of a common pottery style may
indicate cultural connections or alliances of some kind among
the Swift Creek pottery makers of northeastern Florida and
southeastern Georgia.
Design matches also suggest that local groups may have
acquired some pots via exchange with groups to the north.
Alternatively, it may have been the paddles that found their
way into northeastern Florida, possibly along with a female
potter who married into the local group. Based on ceramic
tempering, those groups living on the northern-most Florida
coastal islands are more similar to those of southeastern
Georgia with their grit-tempered pottery than they are to those
living near the mouth of the St. Johns River, suggesting
intraregional population groupings or simply differing clay
sources. The increase in the number of mounds and sites of
the Swift Creek period compared to earlier Woodland times in
northeastern Florida suggests a population increase, but one
that may well have involved increased local birth rates and
influxes of people from either the south or north.
One thing that has become clear as a result of radiometric
dating of Swift Creek in northeastern Florida and elsewhere in
the Southeast is that Swift Creek is post-Hopewell. Even after
the decline ofHopewell around A.D. 400, Swift Creek peoples
continued to be involved in interaction and exchange networks
with distant peoples that brought non-local paddles, people, or

pots to the tidewater hammocks of northeastern Florida, along
with exotic stone, minerals, and metals. At present, radiocar-
bon dates from northeastern Florida Swift Creek mounds
suggest that the Late Woodland lull experienced in various
areas of the Southeast occurred during the ninth century in
northeastern Florida, possibly during waning Late Swift Creek
and Colorinda times. By A.D. 900, the immigrant St. Johns
II inhabitants of northeastern Florida were once again actively
participating in interregional contact and exchange networks
(Ashley 2002, 2003).

Future Considerations

These general observations tell us little about social
relationships engendered between the different Swift Creek
pottery-making groups in northeastern Florida and Georgia.
Paddle matching combined with ceramic technological
analyses, such as petrographic thin sectioning (Stoltman and
Snow 1998), neutron activation analysis (Ashley 2003:104-
128), and sherd retiring (Rolland 2004), can help distinguish
differences in clay sources that may indicate whether paddles
or pots moved across the landscape in specific instances.
Identifying the frequency of locally produced versus imported
wares is merely a first step in understanding the social
processes that define these interactions. Whether down-the-
line exchange of pots, transfer of paddles, or movement of
paddles or people, the particular social means structuring the
dispersal of vessel designs across the landscape must be
examined from a diachronic and multiscalar spatial perspec-
tive centered on local contexts of interaction and exchange.
To undertake such an investigation, ceramic analysis must


2006 VOL. 59(1)


include large collections of vessels from contextually-secure
and well-dated sites. With respect to local Swift Creek pottery
collections, those from continuous-use mounds that contain a
variety of early and late wares offer the chance to document
changes in design styles and track interregional contact and
interaction over time. In addition, analysis also must consider
technological style, which refers to the technological choices
made during pottery production (Lemonnier 1986, 1992). In
this respect, careful attention must be given to examining
technological issues such as clay processing, firing techniques,
post-firing finishing, and methods of cooking among Swift
Creek groups in northeastern Florida. Pottery from mounds
also must be compared to those from domestic contexts to
document similarities and differences between everyday and
mortuary wares.
Variations in technological choices are important in that
they reflect the social context in which artisans learn and
subsequently perform their craft (Costin 1998; Gosselain
1998). An important concept here is habitus, which refers to
routine actions acquired through learning and reproduced
through continual practice so as to go unnoticed and become
ingrained in cultural norms (Bourdieu 1977). Habitus shapes,
frequently at a subconscious level, who people are, and has the
potential to contribute to the formation of identity differences.
Recently it has been argued that decorative styles are more
susceptible to sudden changes due to various influences,
whereas technological style is more conservative and perhaps
a better marker of social identity (Lemonnier 1986, 1992;
Stark 1998). For example, perpetuation of clay processing and
vessel manufacturing techniques with little change throughout
the introduction of a new decorative style might signify
continual production by potters who learned to craft locally.
Conversely, recognition of new ceramic technological proce-
dures within a population might indicate an influx of non-local
potters through intermarriage.
Although enculturation certainly can occur in situations of
spouse relocation after marriage, ethnographic studies of
modem ceramic-making groups, such as Gosselain's (1998)
study in Camaroon, suggest that the beginning stages of
ceramic vessel manufacture are almost never subject to the
forces of enculturation, since the results of these activities
typically are not a visible aspect of the finished product.
Rather, last order steps of manufacture in the production
sequence or operational chain, such as vessel shape, rim
treatment, and surface decoration, are the features of style
most likely to change since they are the most visible and
subject to the scrutiny of others (Carr 1995). Therefore, those
aspects of technological style least visible to the observer and
performed for the most part unconsciously would be more
likely to reveal the social identity (local or non-local) of
pottery makers.
One such technological analysis of an assemblage with
both Early and Late Swift Creek wares suggests that at the
MayportMound considerable continuity existed in the commu-
nity that produced this pottery over a period of nearly six
hundred years (Wallis 2004). While the use of charcoal
tempering ended sometime during the sixth century, the use of

very fine quartz sand as temper continued throughout the
entire period of Mayport Mound use. Combined with similar-
ity in selected clay sources and perpetuation of many vessel
forms through time, this technological continuity shows that
a local pottery tradition was maintained throughout waves of
extralocal design influence (Wallis 2004). This type of
analysis, utilizing large pottery assemblages, can begin to
explicate the social implications of Swift Creek design
incorporation. At the Mayport Mound, for example, it is
unlikely that changes in Swift Creek iconography were
accompanied by any substantial influx of foreign populations
either by intermarriage or migration. Although new vessel
forms and paste characteristics were introduced in limited
amounts during Late Swift Creek times, in general, Swift
Creek iconography seems to have been distributed through
networks of interaction that did not threaten local kinship
continuity (Wallis 2004).
Additional analysis of paddle matches between regions
may yield further information on the social mechanisms and
implications of interaction. Recognition of the paddle match
between vessels from the Mayport Mound and the Lewis Creek
site in Georgia, for example, is a positive step toward this
understanding. Recent analysis of petrographic thin sections
from both vessels revealed that they were likely made from
identical clay sources, although the location of that source
remains speculative without clay sourcing research (Ann
Cordell, personal communication, 2004). Thus, in this
particular instance, it was the vessel that was carried across the
landscape rather than the paddle, although the direction and
mechanism for movement remain uncertain. Based on the fact
that most Swift Creek wares in northeastern Florida are fine
sand-tempered, compared to the coarser tempering of coastal
Georgia wares, it seems likely that the Swift Creek vessels
with coarse temper were brought in from Georgia. In light of
the continuity demonstrated throughout the entire Mayport
Mound assemblage, we can rule out wholesale migration of
Georgia groups into the area as a mechanism for pot move-
ment, but cannot discount the movement of individuals,
perhaps through marriage and relocation. Further analysis of
more pots with paddle matches may begin to establish greater
trends in interaction.
In northeastern Florida, design similarities and paddle
matches suggest that the predominant affinities shifted from
the west (for Early Swift Creek) to the north (for Late Swift
Creek)-we argue that studying changes in local technological
traditions may better elucidate the character of these shifting
interregional relationships. Through this theoretical lens, we
may begin to gain insights into the social processes integral to
the very formation, maintenance, and decline of the local Swift
Creek culture that thrived along the lower St. Johns during the
Woodland Period.


This paper provides a brief overview of our current
understanding of Swift Creek in northeastern Florida and
offers potential directions for the future study of culture




process, a subject made imminently possible by increasing
numbers of absolute dates and the telltale signatures left by
Swift Creek paddle stamping. The intricate designs on Swift
Creek pottery offer uncommon possibilities for monitoring
dynamic processes of identity construction, cultural reproduc-
tion, and ideology appropriation and manipulation in Wood-
land period groups. The opportunities presented by
complicated-stamped ceramics should allow archaeologists to
conduct research toward an understanding of local-scale social
process that is framed within the larger social milieu of Swift
Creek pottery-making peoples. It is hoped that future research
engages these issues.


Sears (1957, 1959) originally used the designation limestone-
tempered to refer to what is today known as charcoal-tempered,
whereas Wilson used the appellation hole- tempered to denote
charcoal-tempered (see Ashley 1998:202 for a review of this ceramic
taxonomical evolution).

2 While at least one fired clay "paddle" has been recovered in
northwestern Florida (Phelps 1969:18), none has been reported in
northeastern Florida.


We would like to thank Frankie Snow for his tremendous
assistance over the years in identifying northeastern Florida Swift
Creek design contacts. Frankie is responsible for identifying the
design match between the Browne Mound and the Schmidt site, and
he also provided the design reconstruction in Figure 5. Keith
Stephenson is acknowledged and thanked for funding at least 10 of
the radiocarbon assays reported herein, as part of his broader Swift
Creek research. Archaeologists with Environmental Services, Inc.
(Greg Hendryx, Greg Smith, and Brent Handley) shared freely
information obtained from several Swift Creek sites in northeastern
Florida. Ann Cordell graciously gave insight from her petrographic
analysis of charcoal-tempered pottery. We also appreciate the
editorial comments of Ryan Wheeler, Louis Tesar, and anonymous
reviewers. Finally, special thanks to Lori Schmidt for allowing us
access to her pottery collection.

References Cited

Anderson, David G.
1998 Swift Creek in Regional Perspective. In A WorldEn-
graved: Archaeology ofthe Swift Creek Culture, edited by
M. Williams and D. T. Elliott, pp. 274-300. University of
Alabama Press, Tuscaloosa.

Ashley, Keith H.
1992 Swift Creek Manifestations along the Lower St. Johns
River. The Florida Anthropologist 45:127-138.

1995 The Dent Mound: A Coastal Woodland Period Burial
Mound Near the Mouth of the St. Johns River. The
Florida Anthropologist 48:13-35.
1998 Swift Creek Traits in Northeastern Florida: Ceramics,
Mounds, and Middens. InA WorldEngraved: Archaeology
of the Swift Creek Culture, edited by M. Williams and D.
T. Elliott, pp. 197-221. University of Alabama Press,


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.
Ph.D. dissertation, Department ofAnthropology, University
of Florida, Gainesville.

Bense, Judith
1998 Santa Rosa Swift Creek in Northwestern Florida. In A
World Engraved: Archaeology of the Swift Creek Culture,
edited by M. Williams and D. T. Elliott, pp. 247-273.
University of Alabama Press, Tuscaloosa.

Bourdieu, Pierre
1977 Outline of a Theory of Practice. Cambridge University
Press, Cambridge.

Carr, Christopher
1995 A Unified Middle-Range Theory of Artifact Design. In
Style, Society, andPerson, Archaeological andEthnologi-
cal Perspectives, edited by C. Carr and J.E. Neitzel,
pp.171-258. Plenum Press, New York.

Cobb, Charles R., and Michael S. Nassaney
1995 Interaction and Integration in the Late Woodland Southeast.
In NativeAmerican Interactions:MultiscalarAnalysis and
Interpretations in the Eastern Woodlands, edited by M. S.
Nassaney and K. E. Sassaman, pp. 205-226. University of
Tennessee Press, Knoxville.

Cook, Fred C.
1979 Kelvin: A Late Woodland Phase on the Southern Georgia
Coast. Early Georgia 7:65-86.

1981 The Kelvin Phase House: Aboriginal Domestic Structure
Design on the Southern Georgia Coast during the Late
Woodland Period. Early Georgia 9:76-87.

Costin, Cathy L.
1998 Introduction: Craft and Social Identity. In Craft and Social
Identity, edited by C. L. Costin andR. P. Wright, pp. 3-16.
Archaeological Papers, No. 8. American Anthropological
Association, Arlington, VA.

DesJean, Thomas, Irvy Quitmyer, and Karen Jo Walker
1985 A Coastal Swift Creek Community at King's Bay, Georgia.
Florida Journal of Anthropology, Special Publication

Dickinson, Martin F., and Lucy B. Wayne
1999 Island in the Marsh: An Archaeological Investigation of
8NA59 and 8NA709, The Crane Island Sites, Nassau
County, Florida. Report on file, Division of Historical
Resources, Tallahassee.

Espenshade, Christopher
1985 Ceramic Technology of the Kings Bay Locality. In Aborig-
inal Subsistence and Settlement Archaeology of the Kings
Bay Locality, Vol. 1, edited by W. H. Adams, pp. 295-226.


University ofFlorida, Department of Anthropology Reports
of Investigations No. 1, Gainesville.

Florida Archeological Services, Inc.
1994 Phase II Archeological Investigations at Sites 8DU5541,
8DU5542, and 8DU 5543 at the Queen's Harbour Yacht
and Country Club, Duval County, Florida. Report on file,
Division of Historical Resources, Tallahassee.

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

Gosselain, Olivier P.
1998 Social and Technical Identity in a Clay Crystal Ball. In The
Archaeology of Social Boundaries, edited by Miriam T.
Stark, pp. 78-106. SmithsonianInstitutionPress, Washing-
ton, D.C.

Handley, Brent M., Steve Ferrell, and Neill Wallis
2004 Addendum To: Data Recovery at 8NA32, The McArthur
Estates Tract, Nassau County, Florida. Report on file,
Division of Historical Resources, Tallahassee.

Hendryx, Greg S.
2004 The Honey Dripper Site (8NA910): A Late Swift Creek
Encampment inNortheasternFlorida. TheFloridaAnthro-
pologist 57:299-310.

Hendryx, Gregory S., and Greg C. Smith
2001 Archaeological Data Recovery and Mitigation at 8NA910
(TheHoneyDripperSite), Nassau County, Florida. Report
on file, Division of Historical Resources, Tallahassee.

Hendryx, Gregory S., Ryan Seip, and Jennifer Nash
2003 Cultural Resource Management at River Point P.U.D.
Parcel: Survey and Site Testing at 8SJ4790 and 8SJ4791.
Report on file, Division of Historical Resources, Tallahas-

Hendryx, Gregory S., Greg C. Smith, and Sidney Johnson
2000 An Intensive Archaeological and Historical Assessment and
Site Evaluation at 8NA703, Martin's Island, Nassau
County, Florida. Report on file, Division of Historical
Resources, Tallahassee.

Johnson, Robert E.
1996 An Intensive Archeological Survey of Florida Inland
Navigation District Tract DU7, Duval County, Florida.
Report on file, Division of Historical Resources, Tallahas-

1998a Phase II Archeological Investigations of Sites 8DU5544
and 8DU5545, Queen'sHarbour Yacht and Country Club,
Duval County, Florida. Report on file, Division of Histori-
cal Resources, Tallahassee.

1998b A Phase II Archeological Investigation of Florida Inland
Navigation District Tract DU7, Greenfield Peninsula,
Duval County, Florida. Report on file, Division of Histori-
cal Resources, Tallahassee.

Johnson, Robert E., Myles C. P. Bland, B. Alan Basinet, and Bob
1997 An Archeological Investigation of the Ocean Reach Site
(8NA782), Nassau County, Florida. Report on file,
Division of Historical Resources, Tallahassee.

Jones, B. Calvin, and Louis D. Tesar
1996 Emergency Archaeological Salvage Excavation within the
Swift Creek Subarea of the Block-Sterns Site (8LE148),
Leon County, Florida: A Public Archaeology Project.
Report on file, Division of Historical Resources, Tallahas-

Kelly, Arthur R.
1938 Preliminary Report on Archeological Explorations at
Macon, Georgia. Smithsonian Institution, Bureau of
American Ethnology, Bulletin 19. Washington, D.C.

Kelly, Arthur R., and Betty A. Smith
1975 The Swift Creek Site, 8-Bi-3, Macon, Georgia. Report
submitted to the Southeast Archeological Center, National
Park Service, Atlanta.

Kirkland, S. Dwight.
2003 Human Prehistory at the Sadlers Landing Site, Camden
County, Georgia. Early Georgia 31:107-192.

Lemonnier, Pierre
1986 The Study of Material Culture Today: Toward an Anthro-
pology of Technical Systems. Journal ofAnthropological
Archaeology 5:147-186.

1992 Elements ofan Anthropology ofTechnology. Anthropolog-
ical Papers of the Museum of Anthropology, vol. 88.
University of Michigan, Ann Arbor.

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

Moore, Clarence B.
1894 Certain Sand Mounds of the St. Johns River, Florida. Part
II. Journal of the Academy of Natural Sciences ofPhila-
delphia, SecondSeries 10:129-246.

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

Nassaney, Michael S., and Charles R. Cobb (editors)
1991 Stability, Transformation, and Variation: The Late Wood-
land Southeast. Plenum Press, New York.

Phelps, David S.
1969 Swift Creek and Santa Rosa in Northwest Florida. Insti-
tute ofArchaeologyandAnthropology, University ofSouth
Carolina Notebook 1:14-24..

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



Russo, Michael
1992 Chronologies and Cultures of the St. Marys Region of
Northeast Florida and Southeast Georgia. The Florida
Anthropologist 45:107-126.

Saunders, Rebecca
1998 Swift Creek Phase Design Assemblages from Two Sites on
the Georgia Coast. InA WorldEngraved: Archaeology of
the Swift Creek Culture, edited by M. Williams and D. T.
Elliott, pp. 154-189. University of Alabama Press,

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

1959 Two Weeden Island Period Burial Mounds, Florida.
Contributions of the Florida State Museum 5, Gainesville.

Smith, Greg C., and Brent M. Handley
2002 Addendum To: ArchaeologicalData Recovery and Mitiga-
tion at 8DU5544/5545, Queen's Harbour Yacht and
Country Club, Duval County, Florida. Report on file,
Division of Historical Resources, Tallahassee.

Smith, Greg C., Brent M. Handley, Keith H. Ashley, and Gregory S.
2001 Archaeological Data Recovery and Mitigation at
8DU5544/45, Queen's Harbour Yacht and Country Club,
Duval County, Florida. Report on file, Division ofHistori-
cal Resources, Tallahassee.

Smith Robin L., R. Bruce Council, and Rebecca Saunders
1985 Three Sites on Sandy Run: Phase II Evaluation at Sites
9Cam183, 184, and 185 at Kings Bay, Georgia. Report
submitted to the U.S. Department ofNavy, Kings Bay, GA.

Snow, Frankie
1975 Swift Creek Designs and Distributions: A South Georgia
Study. Early Georgia 3:38-59.

1977 An Archaeological Survey of the Ocmulgee Big Bend
Region: A Preliminary Report. Occasional Papers from
South Georgia, Number 3. South Georgia College,

1998 Swift Creek Design Investigations: The Hartford Case. In
A World Engraved: Archaeology of the Swift Creek
Culture, edited by M. Williams and D. T. Elliott, pp. 61-
98. University of Alabama Press, Tuscaloosa.

Snow, Frankie, and Keith Stephenson
1998 Swift Creek Designs: A Tool for Monitoring Interaction. In
A World Engraved: Archaeology of the Swift Creek
Culture, edited by M. Williams and D. T. Elliott, pp. 99-
111. University of Alabama Press, Tuscaloosa.

Southeastern Archaeological Research, Inc.
2005 Phase IIAssessment of 8SJ3149 Located on the Marshall
Creek DR[Development Site, St. Johns County, Florida.
Report on file, Division of Historical Resources, Tallahas-

Stark, Miriam T. (editor)
1998 The Archaeology of Social Boundaries. Smithsonian
Institution Press, Washington, D.C.

Stephenson, Keith
2002 A Regional Evaluation of Deptford and Swift Creek
Chronologies from Radiometric and AMS Dates. Paper
presented at the 591 annual Southeastern Archaeological
Conference, Biloxi.

Stephenson, Keith, Judith A. Bense, and Frankie Snow
2002 Aspects ofDeptford and Swift Creek on the South Atlantic
and Gulf Coastal Plains. In The Woodland Southeast,
edited by D. G. Anderson and R. C. Mainfort, Jr., pp. 318-
351. University of Alabama Press, Tuscaloosa.

Stoltman, James B., and Snow, Frankie
1998 Cultural Interaction Within Swift Creek Society: People,
Pots, and Paddles. In A World Engraved: Archaeology of
the Swift Creek Culture, edited by M. Williams and D. T.
Elliott, pp. 130-153. University of Alabama Press,

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

Wallis, Neill J.
2004 Perpetuating Tradition on the Lower St. Johns River:
Pottery Technology and Function at the Mayport Mound
(8DU96). The Florida Anthropologist 57:271-298.

Wallis, Neill J., Ann S. Cordell, and Lee A. Newsom
2005 Petrographic Analysis of Charcoal-Tempered Pottery from
Northeastern Florida. Poster presented at the 701 annual
meeting of the Society for American Archaeology, Salt
Lake City.

Williams, Mark, and Daniel T. Elliott (editors)
1998 A World Engraved: Archaeology of the Swift Creek
Culture. University of Alabama Press, Tuscaloosa.

Willey, Gordon R.
1949 Archeology of the Florida Gulf Coast. Smithsonian
Miscellaneous Collections, Volume 113.

Wilson, Rex L.
1965 ExcavationsattheMayportMound, Florida. Contributions
of the Florida State Museum 13, Gainesville.

Wolf, Eric R.
1982 Europe and the People Without History. University of
California Press, Berkley.

2006 VOL. 59(1)

A video on lorida's native peoples

"Shadows and Reflections:
Florida's Lost People"
Produced by the Florida
Funded by the
jFlorida Department
rA I of State

-W 19fMEMP--w, W --
Produced and Directed by Chaos Productions
Executive Producer: Brent Weisman
Written by Marshall Riggan
Artwork by Theodore Morris

1998 Florida Anthropological Society and the Florida Department of State

To obtain copies send $23.62 [$18.81 plus $1.31 (sales tax) and $3.50 (S&H) to:
Terry Simpson, 9907 High Meadow Ave., Thonotosassa, FL 33592

Florida Anthropological Society Chapters

1) Archaeological Society of Southern Florida C ... .
2495 NW 35th Ave., Miami, FL 33142

2) Broward County Archaeological Society f 5
481 S. Federal Highway,.Dania Beach, FL 33004

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

4) Central Gulf Coast Archaeological Society 13 11
P.O. Box 82255, St. Petersburg, FL 33682

5) Indian River Anthropological Society
3705 S. Tropical Trail, Merritt Island, FL 32952

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

7) Emerald Coast Archaeological Society
333 Persimmon Street, Freeport, FL 32435

8) Panhandle Archaeological Society at Tallahassee ,
c/o The Tallahassee Trust for Historic Preservation
423 E. Virginia Street, Tallahassee, FL 32301 -.

9) Pensacola Archaeological Society
P.O. Box 13251, Pensacola, FL 32591 13) Time Sifters Archaeology Society
P.O. Box 25883, Sarasota, FL 34277-2883
10) St. Augustine Archaeological Association
P.O. Box 1301, St. Augustine, FL 32085 14) Volusia Anthropological Society
P.O. Box 1881, Ormond Beach, FL 32175
11) Southeast Florida Archaeological Society
P.O. Box 2875, Stuart, FL 34995 15) Warm Mineral Springs Archaeological Society
P.O. Box 7797, North Port, FL 34287
12) Southwest Florida Archaeological Society
P.O. Box 9965, Naples, FL 34101



Department ofAnthropology, Florida State University, Tallahassee, FL 32304
E-mail: dkt7819@fsu.edu

In 1999 Don Munroe, a conscientious' avocational artifact
collector and river diver, found a concentration of lithic tools
in the Santa Fe River, near High Springs, Alachua County,
Florida (Figure 1). Over the next few weeks, Munroe collected
over 100 complete stone tools, including lanceolate-based and
Bolen projectile points, a nearly completed Simpson point,
unifaces, bifaces, bola stones, and adzes. Although the
assemblage contains three Archaic stemmed points, a
Hernando point, a probable Columbia point, and a pottery
sherd, the rest of the tools are consistent with Paleoindian and
Early Archaic lithic assemblages from Florida and other
eastern North American sites. The number ofPaleoindian and
Early Archaic artifacts represents an opportunity to enhance
our understanding oflithic tool variation in Florida during this
time period.


The earliest evidence for people in Florida dates to the
early Paleoindian period (ca. 11,200-10,800 B.P.). An ivory
tool from the Sloth Hole site (8JE121), which also produced
Clovis points, on the Aucilla River has been radiocarbon dated
to 11,050 50 B.P. (Hemmings 2004). The working hypothe-
sis in Florida is that Suwannee/Simpson followed Clovis in the
middle Paleoindian period (ca. 10,800 10,500 B.P.), and
Bolen follows Suwannee/Simpson some time after that. Single
component Bolen levels have been confidently dated at ca.
10,000 B.P. at 8LE2105 in Leon County (Hornum etal. 1995),
the Page-Ladson site in Jefferson County (Dunbar et al.
1988:Table 1), and Wakulla Springs Lodge site (Tesar and
Jones 2004). The place of Suwannee and Simpson points in
this chronology is still unresolved, however, because no sites
producing these artifacts have been dated. Further, the
relationship of Bolen and Suwannee tool assemblages is
unclear. The similarities of the Bolen and Suwannee tool
assemblages support an inference of temporal continuity
(Milanich 1994:54), but whether they represent chronologi-
cally distinct traditions is uncertain. The Bolen and Suwannee
points showed no statistically significant stratigraphic separa-
tion at Harney Flats, which may have been due to a hiatus in
geological deposition at that time or occupation of the same
surface at different times (Daniel and Wisenbaker 1987:33-
38). Regardless, no Suwannee or Simpson points have been
recovered from the radiocarbon-dated Bolen sites in Florida or
from the Greenbrier-age site of Warm Mineral Springs
(Cockrell and Murphy 1978; Clausen et al. 1975), and the

inference is that people were no longer making Suwannee and
Simpson points by 10,000 B.P. The Munroe collection
included six lanceolate bases, a Simpson preform, several
projectile point tips that may have been Simpson or Suwannee
reforms, and three Bolen points. Based on the presence of
the projectile points, it may represent a Suwannee/Bolen
assemblage like that from Harney Flats.

The Site

The site is located in a wide section of the Santa Fe River
approximately 2 km upriver from the Munroe Quarry (Smith
et al. 1997) and 1.5 km down river from the Highway 27
bridge. The tools were found in a shallow, 3 x 2 m depression
in the limestone bed of the river about 3 m below a lower river
stage. The area is located in a wide part of the river, in which
a limestone shelf slopes from east to west at a shallow angle.
The edges of the depression are about 20 cm deep. In the
depression is a layer of rocks of various sizes mixed with
organic debris which overlies mousse-like sediments of
indeterminate depth. The west side of the river is about 4 m
from the western edge of the depression. To the east is low
floodplain, and to the west the bank separates the river from a
dry flood channel. About 15 m to the west of the bank, a steep
river bank rises from the dry flood channel. The assemblage
likely did not deflate in place, but gathered, at least in part, in
the depression. The lack of "river polish" on the artifacts,
however, indicates their origin must have been nearby.
The assemblage does not include all the tools from the site.
Other than projectile point parts, it appears Munroe did not
collect broken scrapers, biface fragments (unless they were
either large or parts of projectile points), slightly modified
flakes, or microliths. In order to get a sense of what else was
in the depression we screened eight buckets of material from
the depression and surrounding area through a /4 inch screen
and recovered debitage, cores, and one core-tool. Most of the
artifacts were made from a gray Suwannee chert of varying
quality that is locally available. Several tools were made of
chert from the Ocala formation. The variation in raw material
from our eight bucket samples matches the variation seen in
the tools.
Although the assemblage is mixed with younger material,
several lines of evidence support an inference of likely assem-
blage integrity. The Suwannee and Simpson reforms and
Bolen points are made of the same gray chert from the
Suwannee chert cluster as the majority of the presumptively


VOL. 59(1)


MARCH 2006


Figure 1. Map showing the location of the Munroe assemblage.

associated tools. In contrast, the Archaic points were made
from a yellow chert and are less patinated than most of the
presumptive Suwannee/Bolen tools. Finally, the descriptions
and dimensions of the tools fall within the tool categories at
the Suwannee/Bolen Harney Flats site (8HI507) (Daniel and
Wisenbaker 1987), and the Bolen sites of 8LE2105 (Hornum
et al. 1995) and Jeanie's Better Back (8LF54) (Austin and
Mitchell 1999).

The Artifacts

In this analysis, artifacts were divided into four main
groups: unifaces, bifaces, pounding tools, and adzes.2 The
quality of the raw material of the tools ranged from grainy

silicified limestone to a smooth texture, high-quality chert.
Most of the artifacts were patinated to some extent, obscuring
the original color of the chert and some of the finer details of
the scar patterns. Some of the unpatinated, higher-quality
chert was yellow or black, but the majority was gray, brown,
tan, gray-banded, and yellow-banded. Some of the lower
quality material had inclusions, geodes, voids, fossils, and
bands of lower quality chert. The breaks in most of the broken
bifaces could be attributed to fossil or geode inclusions, which
appears to be a common problem with Florida cherts (Purdy
Tool use was inferred from edge damage or use-wear. A
general dichotomy exists between proponents of low- and
high-magnification use-wear analysis (Odell and Odell-


2006 VOL. 59(1)


Vereecken 1980); both approaches produce distinct but
complimentary information (Odell 2001:50). In this analysis,
a 10x loupe was used to identify retouch locations. Inferences
about use wear were made using other use-wear studies and
experiments, especially George Ballo's (1985) use-wear
experiments and analysis of a sample of the Harney Flats
assemblage. Ballo used a low-power scope ranging from 20x-
50x magnification (Ballo 1985:104). He did not analyze any
tools on which the edges were modified or distorted by
weathering or patination, which eliminated most of the
artifacts from analysis (Ballo 1985:104). In addition, the
graininess of the lithic material made some analyses problem-
atic (e.g., Ballo 1985:107). The Munroe assemblage also has
some tools made of very grainy chert. Like Ballo's problem
with edge-patination, differentiating use-wear from post-
depositional damage to tool edges caused by water and
transport along the riverbed could be problematic. My
research has not revealed any methodology for the systematic
assessment of the post-depositional effects of river transport
and water erosion, but it appears such damage would be most
pronounced at a microscopic level, thereby prejudicing high-
power use-wear analysis.
I relied on Odell (1981) for a basic description of how
stone tools are used. Scraping with the planar (ventral)
surface leading (Figure 2a) would take flakes off the dorsal
surface (Odell 1981:200-202), whereas scraping with the
dorsal surface leading (Figure 2b) would take flakes off the
planar side. However, because of the peculiarities of flake
mechanics, the flaking damage in the first instance would be
greater than in the second, all other things being equal.
Scraping also will round the edge of the tool. Planar surface
leading is designated in this analysis as Type 1 scraping and
dorsal surface leading as Type 2. Whittling is the process of
pushing the tool away from the user transverse to the material
being worked (Figure 2c-d). Whittling produces flake scars on
both surfaces with feather terminations on the lower and hinge
terminations on the upper surface (Odell 1981:202-203).
Cutting or sawing results in flakes removed from both sides of
the edge. It may also result in a denticulated edge (Odell
1981:203). At 10x magnification I only observed Type 1 and
Type 2 scraping, whittling, and cutting. The scar pattern on
a tool is a function of the nature of the material of which the
tool is made, the material worked, and the method of use. The
degree of step flaking on the edge of the tool indicated the
relative hardness of the material that was worked; more step
flaking and stacked step flaking indicated that the tool had
been applied to a relatively harder material.


Unifaces are chipped stone artifacts with flake scars on the
dorsal surface and no, or almost no, flake scars on the opposite
ventral, or planar, surface. Artifacts that had some flake scars
removed from the ventral surface were included in the uniface
category if the scars could be attributed to use wear or to small,
marginal flakes that manifested an intent to remove the bulb
of percussion in order to flatten the bottom of the tool. I tried

not to be too dogmatic about the uniface/biface dichotomy
because the tool-maker may have intended to flatten a flake
with a large bulb of precussion, which would require the
removal of at least one flake from the ventral surface. Re-
moval of the bulb may have been done to facilitate hafting
(Daniel and Wisenbaker 1987:68). Cross sections ofunifaces
were defined as piano-convex, triangular, trapezoidal, and
irregular. Most flakes were struck from unprepared cores.
There appears to be little uniformity among archaeologists
in defining uniface categories in Paleoindian and Early
Archaic assemblages. Daniel and Wisenbaker (1987:62-81)
created categories loosely based on plan-view morphology to
describe the unifaces. Their uniface categories include end
scrapers, discoidal scrapers, oblong scrapers, adzes, thick
unifaces, thin unifaces, and miscellaneous, but each of these
categories is further subdivided by shape, such as irregular,
circular, oval, and ovoid (Daniel and Wisenbaker 1987:67,
Table 7). These designations confuse the broader categories
and create awkward sub-categories where end, discoidal,
oblong, thick, and thin scrapers can all be oval. Other site
reports list additional unifacial categories (e.g., Coe 1964;
Deller and Ellis 1992). In an effort to maintain a means to
compare the Munroe assemblage with Harney Flats I relied on
plan-view descriptions and added one category, ovoid scrapers,
to those used at Harney Flats. My uniface categories include
end (meaning triangular), discoidal, ovoid, oblong, adzes, and
side scrapers. Side scrapers were subdivided into thick and
thin scrapers. The term scrapers and adze are heuristic names
and are not intended to infer use, only to facilitate compari-

End scrapers. Daniel and Wisenbaker (1987:63) define the
end scraper as "roughly triangular or tear-dropped shaped
flake with a rounded working edge outline usually opposite the
platform end" with a tapering stem and steep and extensive
retouch. A review of their sample (1987:64, Figure 21)
shows a multitude of forms some of which do not appear either
triangular or tear-dropped. Ballo (1985:105) finds that the
Harney Flats end scrapers were used to scrape with the planar
(ventral) surface as the leading edge (Type 1) in contact with
the material, although some of the tools had damage to both
planar and dorsal surfaces indicating they also were used in a
back-and-forth motion to saw or cut. The degree of damage
indicates the end-scrapers were used to work medium to hard
substances. The mean edge angle of the Harney-Flats end
scrapers was 69, which Wilmsen (1970:70-71) infers means
the tools would have been used for bone and wood working
and shredding.
I defined end scrapers as having a broad working edge
opposite the proximal end of the flake (or where the proximal
end likely was on broken tools), although the working edge of
the tool may extend down one or both of the lateral sides. End
scrapers were trianguloid or teardrop-shaped, usuallyunifacial
but could include bifacial tools as long as some of the ventral
face remained, and it appeared the bifacial thinning was done
to make a flat bottom on the tool. All of the end scrapers were
used in Type 1 scraping. Some also were used for Type 2




Scraper motion

Use wear
flakes -

C Initial whittling motion


Scraper motion

Use wear

\ Vectorforce

Figure 2. Idealized tool motions: a) planar or ventral edge leading (Type I scraping); b) dorsal edge leading (Type I
scraping); c) initial whittling motion; d) secondary whittling motion. From Odell (1981:200-203).

scraping or whittling. One may have been used for cutting
across the bit edge.
Four complete specimens in the assemblage meet the
classic "tear-drop" shape described in other assemblages
(Figure 3a-c). The quality of the stone varies greatly on the
end scrapers, which may indicate that they were intended for
different uses. The specimen in Figure 3b is made on fine-
grained black chert, whereas the specimen in Figure 3a is a
very coarse material that did not retain small flake scars on the
working edge. Figure 4a illustrates one of two end scrapers in
the collection that have retained their platform, with narrowed
distal ends and worked lateral edges.4 This artifact meets the

criteria for a Hendrix scraper as defined by Purdy (1981) and
were recovered at Harney Flats (Daniel and Wisenbaker
Three additional artifacts appear to be the broken bit ends
of hafted end scrapers. Figure 3d is the broken bit end of a
large end scraper and is made of coarse chert. One broken end
scraper has a feather termination that matches the bending
snap fracture pattern found on hafted tools (Grimes and
Grimes 1985:41).
Several complete end scrapers and snapped end scrapers
were recovered at 8LE2105 (Hornum et al. 1995:213-214,
Figure 42) and Jeanie's Better Back (Austin and Mitchell


2006 VOL. 59(1)


0 1

4 cm

Figure 3. Four endscrapers of different sizes: a) and d) are made from coarse chert; d) is missing the haft end.


Ovoid scrapers. I define ovoid scrapers as unifaces that are
ovoid in plan view but also may be pointed at one or both ends
(Figures 4b-c). Ovoid scrapers are hump-backed, triangular,
or piano-convex in cross-section. Daniel and Wisenbaker
(1987) would have included ovoid scrapers in the end scraper
category. As with all end scrapers, Daniel and Wisenbaker
(1987) assume these were hafted.5 Similarly shaped unifacial
scrapers were recovered at Jeanie's Better Back (Austin and
Mitchell 1999:Figure 16) and 8LE2105 (Hornum et al.
1995:Figure 44).
The Munroe assemblage contains eleven ovoid scrapers
that range from 45.8 75.7 mm in length and to 33.6 50.0
mm in width. Nine of these are triangular or plano-convex in
cross-section, and two are relatively thin. All are retouched
around their entire circumference except two, which are
retouched on about 80% of the circumference. Two bifacial
tools match the shape of the ovoid scrapers with flat bottoms
and triangular cross sections.
The ovoid scrapers show evidence of Type 1 scraping,
slight Type 2 scraping, whittling, and cutting. Both the thin
ovoid scrapers were used for light scraping or cutting. The
rest were used for heavier scraping. Artifacts of similar size
and shape are known as "Dalton adzes" (Goodyear
1974:Figure 14) or "Clear Fork gouges" (Bullen and Benson
1964:161). Gaertner (1994) determined through use-wear
experimentation that Dalton adzes from Arkansas were hafted
or socketed and used exclusively for woodworking.

Discoidal scrapers. Discoidal scrapers are described by
Daniel and Wisenbaker (1987:69-70) as similar to end
scrapers but more circular and likely hand-held.5 Ballo

(1985:106-107) found that most discoidal scrapers at Harney
Flats were used for Type 1 scraping, although one specimen
showed flaking consistent with Type 2 scraping. They were
likely used on medium hard to hard materials based on the
degree of edge damage.
Five small (< 6 cm diameter) discoidal scrapers in the
Munroe collection appear clearly differentiated from the ovoid
scrapers. These tools are not significantly modified, and the
working edges are thin. The width to thickness ratio is about
3:1, whereas in the ovoid scrapers the ratio is about 2:1. All
five retain at least some of the original flake platform and are
worked around the entire circumference except for the plat-
form. They range in diameter from 36.9 56.8 mm. Two of
the three larger scrapers have flakes removed from the ventral
side to thin the pronounced bulb of percussion. Four are
nearly circular in plan view with a low plano-convex or
trapezoidal profile. The discoidal scrapers appear to havebeen
used for light Type 1 scraping.
The assemblage also includes one large (118 mm diameter)
discoidal scraper (Figure 5a). This tool is retouched around
the entire circumference and was not made from a core. It fits
neatly in one hand, and judging by the multiple step fractures,
it was probably used for heavy Type 1 scraping. A similar-
looking artifact was found at the Sloth Hole site in the Aucilla
River, Jefferson County (Hemmings 1999).

Oblong scraper. Oblong scrapers have parallel sides and
either rounded or snub-nosed ends. Purdy (1981:18-20)
describes two oblong scrapers, the Hendrix scraper and a snub-
nosed or oblong scraper (Purdy 1981:20-21), but Daniel and
Wisenbaker (1987:70-74) view both types as variations of a
general oblong scraper-type that changes shape as the tool is
resharpened. The variation of the working ends of the oblong




0 1 4cm

Figure 4. Three scrapers: a) narrow-end Hendrix scraper; b) and c) are two ovoid scrapers.

A 0 1 4 cm

Figure 5. Two larger scrapers: a) large discoidal scarper; b) core-scraper.


2006 VOL. 59(1)


0 1 4cm
0 0 1 0

Figure 6. Four oblong scrapers.

scrapers in the Munroe collection support an inference that the
tools were created for specific purposes, however. Figure 6
shows four examples from the eight complete specimens from
the Munroe collection. Figure 6b appears to match Purdy's
(1981:19) examples of a Hendrix scraper, but it has minimal
use wear (i.e., only polish) on the tip. Ballo (1985:107-109)
analyzed a limited sample of oblong scrapers because most had
excessive edge damage. The scrapers he examined indicated
that both the dorsal and planar surfaces were used as leading
Along with the eight complete oblong scrapers, the
collection also has five broken specimens. The complete
scrapers range from 61.7 115.5 mm long and 30.9 51.0
mm wide. They are generally rounded to slightly pointed on
the ends, although the example in Figure 6a has a straight bit
at one end. The broken specimens are all of similar length,
indicating they broke at the haft. Use wear indicates that some
of the ends were used for whittling. Type 1 scraping is
indicated on some of the ends and lateral margins, and some
of the lateral margins were used for cutting. Most of the
scraping appears to have been against harder material.
Grimes and Grimes (1985:50) analyzed oblong scrapers
from Bull Brook, which they termed spokeshavers, and
determined these tools were likely socketed, hand-held
scrapers. The Bull Brook sample showed a regular breakage
pattern consistent with hafting. Some of the spokeshavers
showed a deliberately narrowed proximal end as if the haft was
trimmed to fit a handle. In the Munroe assemblage, the likely
haft-end of the oblong scrapers tend to fall in three widths:

approximately 26 mm (25, 26, 26), 30 mm (30), and 43 mm
(42, 43, 44, 44) wide, which indicates an intent to fit the tool
to a particular haft.

Side scrapers. The side scraper category is typically a catchall
for unhafted unifacial tools that do not meet other criteria.7 At
Harney Flats Daniel and Wisenbaker (1987:74-79) include all
other unifacially retouched flakes that did not meet the end,
discoidal, adze, or oblong scraper definitions into either a thin
or thick scraper category. Thin and thick scrapers are differ-
entiated on the basis of the width of their working edges: thin
unifaces have working edges less than 10 mm thick at the
location of retouch, and thick unifaces have working edges
thicker than 10 mm. The Munroe assemblage has twelve side
scrapers. All are unifacial and all but two of the specimens
retain the original platform. In addition to differentiating the
side scrapers into thick and thin, they can be generally
grouped into four sub-categories: backed steep-sided scrapers,
wide distal-end scrapers, low-angle scrapers, and miscella-
neous. On the five backed steep-sided scrapers (Figure 7), the
working edge is opposite a flat edge that was either formed by
the original platform or by snapping or removing the edge of
the flake. Two specimens are wide distal-end scrapers formed
on expanding angle flakes. Eight specimens are low-angle
scrapers on which one or more edges are retouched.

Thick scrapers. Ballo (1985:109-111) found that thick
unifaces were used for both Type 1 and 2 scraping. Some
showed bifacial damage consistent with sawing. The Munroe




0 1 4

Figure 7. Steep-sided, thick side-scraper.

assemblage has four thick scrapers, which all had use wear
consistent with Type 1 and 2 scraping.

Thin scrapers. Most of the thin unifaces were used as scrapers
with the planar surface as the leading edge (Ballo 1985:111-
113). Thin unifaces were used on material of medium hard-
ness. The Munroe assemblage has eight thin scrapers, which
all had use-wear consistent with heavy Type 1 scraping. Some
showed light Type 2 scraping and cutting, and one may also
have a graver spur.


Daniel and Wisenbaker (1987:79-81) define adzes
as unifacial heavy-duty tools that were apparently
hafted. Ballo (1985) did not analyze adzes as a sepa-
rate category. Odell and Odell-Vereecken (1980:100)
state that adzes are used transverse to the working
surface and will show unifacial flake scars and

UnifacialAdzes. I included four large unifacial tools
in this category. Figure 5b is a unidirectional core that
was recycled as an adze. Daniel and Wisenbaker
(1987:82) identified similar recycled cores at Harney
Flats.8 The tool is made on poor quality material and
the ventral face is particularly irregular. Although it
fits neatly in one's hand or could have been hafted,
there is little evidence of retouch or heavy use-wear.
The artifact pictured in Figure 8a was apparently hafted
at the waist. The heaviest retouch is along the rounded
bit end. None of the unifacial artifacts in the collection
fall within the dimensions of a "typical" Aucilla adze
as defined by Gerrell et al. (1991:Figure 5).

Bifacial Adzes. Daniel (1998:63-66) describes the
bifacial adze as a somewhat oval biface with a convex
bit at the broadest end of the tool. The adzes have
evidence of hafting on the rounded butt-end. Bifacial
adzes were found at the Dalton-age Brand site in
Arkansas (Goodyear 1974) and at Harney Flats (Daniel
and Wisenbaker 1987:79-81), where one bifacial and
five unifacial adzes are described. Two bifacial adzes
are in the Munroe collection (Figure 8b).


Bifaces form the other large category of tools. The
bifaces in the Munroe collection are categorized as
points, Type I, II, and III bifaces, and adzes (Daniel
1998:50-66). As biface manufacture progresses, the
cm tool generally gets thinner and a more definite shape.
The three biface types represent apparent forms along
that reduction continuum (Daniel 1998:64). This is not
to suggest that the three types are just preliminary
stages of a final formal biface type; at any of the stages
the biface may be used as a tool and any of the stages
could be the intended final form of that tool. However, these
types are a convenient way to conceptualize and group bifaces
for comparison and analysis.

Projectile points. Although the term "projectile point" is
laden with functional implications, it is commonly used as a
category of bifaces, and I use it here to refer to a biface that
could have been used as a projectile, knife, or both. The
Munroe collection does not contain a complete Suwannee
point but has five lanceolates with concave bases, one of which
exhibits lateral grinding (Figure 9a). One Simpson perform


2006 Voi. 59(l)


A 4 cm B

Figure 8. Two adzes: a) unifacial; b) bifacial.

O 1 4 cm


Figure 9. Lanceolate point bases: a) has ground basal edges; b) burinated and reworked after the tip broke; c) snapped at
the location of a large geode.




0 1 cm

notch with a straight base; a partially beveled, corer-
notched, with an incurvate base; a partially beveled,
corner-notched, with an excurvate base; and an
unbeveled with parallel sides), one Hernando point, one
complete and two partial Archaic stemmed points, and
what may be a Columbia point.

Type I bifaces. Type I are early stage bifaces and
typically called "blanks" with irregular or roughly oval
shapes and flake patterns, higher width:edge ratios, and
sinuous edges (Daniel 1998:60-61). The Munroe
collection has ten complete tools that meet this defini-
tion. Step fracturing on all these artifacts indicates they
were used as tools.

Type II bifaces. Type II may be referred to as "per-
forms" and are thinner, more symmetrical, have
biconvex profiles, and less sinuous edges (Daniel
1998:61). The Munroe collection contains one broken
and nine complete Type II bifaces, all of which show
evidence of tool-use (Figure 11). One is significantly
larger and has evidence that its base was used for
pounding; it may have been used as a wedge. Odell and
Odell-Vereecken (1980:100) state that wedges will have
well-defined hinged and stepped fractures on the work-
ing edge and evidence of pounding on the opposite edge,
and this tool shows evidence of bifacial flaking consis-
tent with these criteria.

Type IIIbifaces. Type III bifaces typically have pointed
ends, squared bases, and closely spaced flake patterns
Daniel (1998:62). The Munroe collection has 17
examples of this biface type, but none are complete. Six
broken tips are large enough to have been part of
Suwannee or Simpson performs, but two others are
narrower and may have been from Bolen points. Most
appear to have been broken prior to completion of the
point because they lack a final fine edge. Five of the six
show evidence of a material flaw in the break such as a
geode or fossil. However, one shows an impact fracture
at the tip and a bending fracture in middle of the point.
Three of the artifacts are rounded bases that appear to
have been snapped. Figures 9a-c appear to be Suwannee
reforms. Only Figure 9a shows evidence of basal


Figure 10. Simpson perform that was abandoned before complete

was recovered (Figure 10). The Simpson preform was
apparently abandoned when the knapper was unable to remove
a large mass on the lateral edge. It appears the knapper tried
to remove the problem with at least four overshot flaking
attempts (Andy Hemmings, personal communication 2002).
The collection also includes four Bolen points (a beveled E-

Groundstones. The assemblage included two dimple, or
Sbola, stones. The function of these artifacts, one of
which was recovered from the Bolen level at Page-
Ladson, are unknown but the subject of great speculation
(Tesar 1994; Rachels and Knight 2004).

Hammer stones. Four artifacts exhibit evidence of use as
hammer stones. All are made of cryptocrystalline materials
that exhibit one or more surfaces that are rounded by pecking.


2006 VOL. 59(1)


Figure 11. Two type-H bifaces.

The surfaces are pitted and cracked (Odell and Odell-
Vereecken 1980:100). Three appear to be recycled cores, one
of which was also reused as a scraper.

Discussion and Conclusion

From the foregoing analysis it is safe to conclude that the
lithic assemblage (but for the younger projectile points) is
consistent with other lithic assemblages from the
Paleoindian and Early Archaic periods in eastern North
America. The site is deflated, but it is likely located, or was
located, in the immediate area of the depression. The tools
did not show any obvious effects of river transport or polish-
ing; some of the edges were still sharp and small flake scars
were apparent on the working edges. In addition, there was
no evidence of the random patterns of flake scars that would
be expected if the scars were produced by natural forces
(Tringham et al. 1974). Finally, because of the topography

of the riverbed, the tools could not have moved more than
50 meters down river. Whether the Santa Fe River was
flowing during the Late Pleistocene/Early Holocene in this
location is unknown, however, it was not flowing during the
late glacial maximum when trees grew in mid-channel. One
such tree has been dated at 18,550 200 B.P. (Smith et al.
The site is not single-component, but whether it repre-
sents a mixture of Paleoindian and Early Archaic period
assemblages is unknowable because many of the these arti-
facts are found in both Bolen and Suwannee sites in Florida
and from Paleoindian and Early Archaic-age sites through-
out eastern North America. Other than the large discoidal
scraper (Figure 5a), versions of all the tools were also found
at Harney Flats. The presence of the Suwannee and
Simpson points argues for at least a mixed Paleoindian and
Early Archaic assemblage, but other than the large scraper,
the lanceolate points, and large projectile point tips, all of




the tool forms also are present at the Bolen site of 8LE2105.
Assuming that Bolen and Suwannee/Simpson represent
temporally distinct cultures there may be temporal continu-
ity in the lithic tools. Other than the diagnostic projectile
points, the lithic assemblages are too difficult to distinguish.
Based upon the use-wear analysis it appears that tasks
were undertaken at the site that involved the modification of
hard materials like wood, bone, and antler. The number of
ovoid scrapers that look very similar to Dalton adzes, oblong
scrapers, and larger scrapers indicates that wood working
was a significant activity. In fact, most of the larger tools
look as if they were used for scraping harder material as
opposed to scraping softer material or activities involving
cutting. Whether significant tool manufacture and raw
material reduction was conducted at the site cannot be
known, but the presence of broken, unfinished projectile
points indicates that some tool manufacture was undertaken.
Notwithstanding the sampling issues, because most of the
tools were finished and used, we can tentatively conclude
that tool production was a relatively minor focus of the site.
The number and variety of tools indicates the site may have
been a sizable activity area. Hopefully, parts of the site are
still intact, either in the river bottom or on the adjacent
shore, and await further investigation.
Although these tools were found out of stratigraphic
context, there is always some useful information that can be
gained if the artifacts are made available for study. Ques-
tions of morphological variation, tool use, tool and site
distributions, and site formation processes are but a few of
the issues that can be addressed by "river finds." Don
Munroe is an avocational collector who understands that his
hobby can enhance academic archaeology to the benefit of
everyone and should be commended for reporting his finds
and making them available for study.


He reported the finds to the Isolated Finds program of the Divi-
sion of Historical Resources, Florida Department of State. The
Isolated Finds program has been discontinued.

2 The weight (0.1 gm), and maximum length, width, and thickness
(0.1 mm) were measured on all tools. On all tools that still had a
platform the maximum length was measured from the platform to
the distal end, and the maximum width and thickness were mea-
sured perpendicular to that axis. On unifaces that did not have a
platform the maximum length was first measured and the other
two dimensions were measured perpendicular to that axis. On
whole bifaces the maximum length was measured from the tip to
the base. On broken bifaces the maximum length was measured
along the axis that would have intersected the tip and base. Raw
material was measured as coarse or not coarse. Retouch was
defined as at least three flakes removed in a row.

3 There is no general agreement on the definition of an end
scraper. Deller and Ellis (1992:56) limit end scrapers to steep-
bitted tools that were apparently hafted, whereas Goodyear
(1974:43-46, Figure 15) identifies both a hafted and unhafted end
scraper. Storck (1997:78) defines them in regards to the location
of the steep, convex working edge opposite the platform instead of

the by their unifacial nature. Daniel (1998:66) defines them as
having steep bits on the relatively narrow end of a uniface. Kraft
(1973:70-71) differentiates nine end scraper types, Deller and Ellis
(1992:55) identify eight, Storck (1997:78) identifies four with
three subtypes, and MacDonald (1968:90-95) describes six. End
scrapers are differentiated on the basis of retouch location (e.g.,
Storck 1997), plan-view shape (e.g. Deller and Ellis 1992), spurs
(MacDonald 1968), and combinations of these attributes. Most
authors surmise that end scrapers were likely inserted in either an
antler or bone handle or attached to a wooden handle.

4 These match the narrow end scrapers found at the Thedford II
site, a Paleoindian site in Ontario (Deller and Ellis 1992).

5 At the Paleoindian Plenge site in New Jersey (Kraft 1973:71)
ovoid scrapers appear to be types 28 and 29 of the end scrapers.
At the Paleoindian/Early Archaic Brand site in Arkansas, they
appear to be characterized as adzes (Goodyear 1974:40-42). Purdy
(1981:11, 23, 35-38) would probably describe them as general
unifacial or unifacial hump-backed scrapers. Storck (1997:65-66)
describes scrapers with a thick cross-section as "beaked" from the
Paleoindian Fisher site in Ontario. Coe (1964) and Daniel (1998)
would include them in Type II or Type III end scrapers from the
Early Archaic Hardaway site in North Carolina. MacDonald
(1985:94) would describe these as humpback end scrapers from the
Paleoindian Debert site in Nova Scotia that were not likely hafted
because they were retouched around the entire circumference.

6 Purdy (1981) does not differentiate a discoidal scraper. Kraft
(1973:71) shows a circular end scraper that is similar to the
Munroe artifacts.

7 A tabular core scraper was found at the Paleoindian Wells Creek
Crater site in Kentucky (Dragoo 1973:41, 44), and similar artifacts
were found at the Paleoindian/Early Archaic Stansfield-Worley site
in Alabama (DeJarnette et al. 1962) and the Early Archaic Eva site
in Tennessee (Lewis and Lewis 1961).

8 Kraft (1973:100) includes utilized flakes and flakes that were
only modified on the working edge in this category. Dragoo
(1973:27) defines side scrapers as having straight or nearly straight
working edges. MacDonald (1968:95) defines them as unifacial
tools with steep retouch. Daniel (1998:83-84) defines side scrap-
ers as having retouch on any edge other than the end.

9 Similar core-tools are described by Dragoo (1973:39-42) at the
Wells Creek Crater site.


I would like to thank Don Monroe for generously lending me
his collection of artifacts for several months for study and for the
example he sets for all avocational collectors and Jim Dunbar who
provided thoughtful and incisive comments. This paper was
originally written for Dr. Michael Faught, whose input and sugges-
tions were helpful. Without Michael's infectious enthusiasm, I
would probably be off doing other things and not having as much

References Cited

Austin, Robert J., and Scott E. Mitchell
1999 Archaeological Investigations at Jeanie 's Better Back
(8LF54), an Early Archaic Site in Lafayette County,


2006 VOL. 59(1)


Florida. Southeastern Archaeological Research, Inc,

Ballo, George R.
1985 Experiments in Use-Wear Formation on Stone Tools
Made from Florida Cherts: A Study Supporting a
Microwear Analysis ofPaleoOlndian Lithic Artifacts
from the Harney Flats Site, Tampa, Florida. Unpub-
lished Master's thesis, Department of Anthropology,
University of South Florida.

Bullen, Ripley P., and Carl A. Benson
1964 Dixie Lime Caves Numbers 1 and 2, a Preliminary Re-
port. The Florida Anthropologist 17:153-164.

Clausen, Carl J., H. K. Brooks, and Al B. Wesolowsky
1975 The Early Man Site at Warm Mineral Springs, Florida.
Journal ofField Archaeology 2:191-212.

Cockrell, W. A., and Larry Murphy
1978 Pleistocene Man in Florida. Archaeology ofEastern
North America 6:1-13.

Coe, Joffre L.
1964 The Formative Cultures of the Carolina Piedmont.
Transactions 54. American Philosophical Society, Phila-

Daniel, Jr., I. Randolf
1998 Hardaway Revisited: Early Archaic Settlement in the
Southeast. University of Alabama Press, Tuscaloosa.

Daniel, R. I., and M. Wisenbaker
1987 Harey Flats: A Florida Paleo-Indian Site. Baywood
Publishing, Farmingdale.

DeJarnette, David L., Edward B. Kurjack, and James W. Cambron
1962 The Stansfield-Worley Bluff Shelter. Journal ofAla-
bama Archaeology 8:1-111.

Deller, D. Brian, and Christopher J. Ellis
1992 Thedford II: A Paleo-Indian Site in the Ausable River
Watershed of Southwestern Ontario. Memoirs of the
Museum of Anthropology, University of Michigan No.
24. Museum of Anthropology, Ann Arbor.

Dragoo, Don W.
1973 Wells Creek An Early Man Site in Stewart County,
Tennessee. Archaeology ofEastern North America

Dunbar, James S., Michael K. Faught, S. David Webb III
1988 Page/Ladson (8JE591): an Underwater Paleo-Indian Site
in Northwestern Florida. The Florida Anthropologist

Gaertner, Linda M.
1994 Determining the Function of Dalton Adzes from North-
east Arkansas. Lithic Technology 19:97-109.

Gerrell, Phillip R., John F. Scarry, and James S. Dunbar
1991 Analysis of Early Archaic Unifacial Adzes from North
Florida. The Florida Anthropologist 44:3-16.

Goodyear, Albert C. mI
1974 The Brand Site: A Techno-Functional Study of a Dalton
Site in Northeastern Arkansas. Arkansas Archaeological
Survey Research Series No. 7. University of Arkansas,

Grimes, John R., and Beth G. Grimes
1985 Flakeshavers: Morphometric, Functional and Life-cycle
Analysis of a Paleoindian Unifacial Tool Class. Archae-
ology ofEastern North America 13:35-57.

Hemmings, C. Andrew
1999 The Paleoindian and Early Archaic Tools of Sloth Hole
(8Je121): An Inundated Site in the Lower Aucilla River,
Jefferson County, Florida. Unpublished Master's thesis,
Department of Anthropology, University of Florida,

2004 The Organic Clovis: a Single Continent-Wide Cultural
Adaptation. Ph.D. dissertation. Department of Anthro-
pology, University of Florida, Gainesville.

Hornum, Michael B., Donald J. Maher, Clifford Brown, Julian
Granberry, Frank Vento, Arlene Fradkin, and Michele Williams
1995 Phase III Data Recovery at Site 8LE2105for the pro-
posed Florida Gas Transmission Company Phase III
Expansion Projectd, Leon County, Florida: Draft Re-
port, v. I. R. Christopher Goodwin & Associates, Inc.,
New Orleans.

Kraft, Herbert C.
1973 The Plenge Site: A Paleo-Indian Occupation Site in New
Jersey. Archaeology ofEastern North America 1(1):56-

Lewis, Thomas M., and Madeline K. Lewis
1961 Eva: An Archaic Site. University of Tennessee Press,

MacDonald, George E.
1968 Debert: a Paleoindian Site in Central Nova Scotia.
Persimmon Press, Buffalo.

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

Odell, George H.
1981 The Mechanics of Use-Breakage of Stone Tools: Some
Testable Hypotheses. Journal ofFieldArchaeology

2001 Stone Tool Research at the End of the Millennium: Clas-
sification, Function, and Behavior. Journal ofArchaeo-
logical Research 9:45-100.

Odell, George H., and Frieda Odell-Vereecken
1980 Verifying the Reliability of Lithic Use-Wear Assess-
ments by "Blind Tests": The Low-Power Approach.
Journal ofFieldArchaeology 7:87-120.

Purdy, Barbara
1975 The Senator Edwards Stone Workshop, Marion County.




The Florida Anthropologist 28:178-189.

1981 Florida's Prehistoric Stone Technology. University of
Florida Press, Gainesville.

Rachels, Thomas, and Robert L. Knight
2004 Dimple Stones A Unique and Early Ground Stone
Artifact Type from the Southeast. The Amateur Archae-
ologist 10:57-75.

Smith, Roger C., James S. Dunbar, and Michael Faught
1997 An Underwater Archaeological Survey in the Santa Fe
River, Florida, July 1997. Florida Archaeological Re-
ports 36. Bureau of Archaeological Research, Tallahas-

Storck, Peter L.
1997 The Fisher Site: Archaeological, Geological and
Paleobotanical Studies at an Early Paleo-Indian Site in
Southern Ontario, Canada. Museum of Anthropology,
University of Michigan No. 30. Ann Arbor.

Tesar, Louis D.
1994 Clubheads, Bola Stones, or What? The Florida Anthro-
pologist 47:295-303.

Tesar, Louis D., and B. Calvin Jones
2004 Wakulla Springs Lodge Site (8WA329) in the Edward
Ball Wakulla Springs State Park, Wakulla County,
Florida: a Summary ofEleven Projects and Management
Recommendations. Bureau of Archaeological Research,
Florida Department of State, Tallahassee.

Tringham, Ruth, Glenn Cooper, George Odell, Barbara Voytek,
and Anne Witman
1974 Experimentation in the Formation of Edge Damage: A
New Approach to Lithic Analysis. Journal ofField Ar-
chaeology 1:171-196.

Wilmsen, Edwin N.
1970 Lithic Analysis and Cultural Inference: a Paleo-Indian
Case. Anthropological Papers of the University of Ari-
zona, No. 16. University of Arizona Press, Tuscon.


2006 VOL. 59(1)



Department ofAnthropology, University of Florida, Mailing address: 6379 Hampshire Court, Lisle, IL 60532
E-mail: gold.melissa@gmail.com


This article presents an osteological analysis of an archaeo-
logical skeletal collection from Charlotte County in west-
central Florida. The collection dates to late in the Manasota
Culture Period, ca. A.D. 410 to 660. The remains were
analyzed with respect to minimum number of individuals
(MNI), age, sex, stature and presence of trauma and/or
pathology. To understand more about the lives of the
Dunwody people, this study examines their overall health and
lifestyle as indicated by skeletal markers. Frequencies of
specific disease states in this population are compared to those
of other skeletal populations along the Florida Gulf Coast in
order to gain a better understanding of how this site fits into
wider patterns in Florida archaeology.

The Dunwody Site

The skeletal collection analyzed here came from the
Dunwody Site (8CH61), a burial area located on Cedar Point
along the shore of Lemon Bay in northwestern Charlotte
County, Florida.' The Dunwody Site burial area was a compo-
nent of the Cedar Point Shell Heap (8CH8), one of the largest
shell middens along Lemon Bay. Florida Park Service
archaeologist Ripley Bullen made a small collection from
8CH8 in 1949 that included artifacts from the post-contact
Safety Harbor Period (Florida Museum of Natural History
[FLMNH] cat. no. 96104). Along with his wife and physical
anthropologist Adelaide Bullen, Ripley Bullen revisited the
site in 1965 in order to salvage artifacts (FLMNH cat. no.
99680) from the recently bulldozed midden and to recover
human burials (FLMNH cat nos. 99660-99677) before they
were covered with dredged spoil. The burials were removed
from the most northwestern shell lens along the shoreline
among mangroves, but the exact location is unclear. In
addition to the 18 discrete burials and "Disturbed Burials"
excavated at this time, potsherds (FLMNH cat. no. 99678) and
artifacts (FLMNH cat. no. 99679) also were collected. The
artifacts found with the burials were not numerous and they
were not diagnostic of a particular culture (Luer 1999a:46-49;
Mitchem 1989:239, 250-251, Table 35).
Due to the hurried, salvage oriented nature of the recovery,
there is no recorded provenience information by the Bullens
for the Dunwody Site. Besides burial numbers, the only
available provenience information consists of three newspaper
photographs showing two in situ primary flexed burials and a
mass of human bones and shells (left-handed whelk, quahog,

and other marine shells) (Cortes 1965). The artifacts associ-
ated with the Dunwody Site specifically, and the Cedar Point
Shell Heap more generally, give only a vague indication of
when the burials might have occurred. In 1993, archaeologist
George Luer returned to Cedar Point and surface collected
artifacts. Interpretation of these artifacts suggests multiple
components of occupation for the Cedar Point Shell Heap
ranging from the Late Archaic Period (ca. 2000-500 B.C.) to
the late Safety Harbor Period (A.D. 1500-1725) (Luer 1999a).
In order to address the issue of time I sent two human bone
samples from Dunwody burials for AMS (accelerator mass
spectrometer) radiocarbon dating.

Radiocarbon Dates and Culture Period

Two samples of bone, one each from Burials 8 and 11
(FLMNH cat. no. 99667 and 99670) were sent for AMS
radiocarbon dating (Table 1). These samples were chosen
from the same element, the right humerus, of two distinct
individuals. Burial 8 is an adolescent male from about 16-19
years old at death, while Burial 11 is a young adult female
aged 18-30 years old at death. By selecting the samples in this
way, dating two different individuals was assured. Dating only
two samples was based on financial considerations and the
knowledge that another author will be publishing at least one
additional radiocarbon date based on Dunwody Site material.
Jennifer Kelly, graduate student at the University of South
Florida, took skeletal samples from twelve Dunwody burials to
perform isotopic analysis in the fall of 2001 (Kelly 2004:73,
Table 11). Kelly performed the analyses of the Dunwody
material, in addition to material from many other sites, and
presented her findings in her Master's thesis. She states that
radiocarbon dating will be obtained in the near future (Kelly
2004) and some dates will be from the Dunwody Site (personal
communication, 2005).
The sample from Burial 8 indicates a measured (uncor-
rected) age of 1150 +/- 40 B.P. (Beta-203115; bone collagen,
"C/12C = -7.2 %o) with a 2-sigma calibrated date of cal A.D.
550 to 660. Burial 11 indicates a measured age of 1290 +/- 40
B.P. (Beta-203116, bone collagen, 'C/2C= -7.7 %o) with a 2-
sigma calibrated date of cal A.D. 410 to 580. These date
ranges, in addition to the location of the skeletal material,
suggest that the Dunwody Site belongs to the late Manasota
Culture. Several other sites around Lemon Bay have produced
similar radiocarbon dates in the Manasota Period (Appendix
The Manasota Culture existed from ca. 500 B.C. to A.D.


VOL. 59(1)


MARCH 2006



0 5 10 15 20 25 30 35 40

l Indeterminate n Right o Left

Figure 1. Relative frequency of skeletal elements recovered (based upon Appendix I).

800 and is characterized by archaeological sites located on or
near the shore with evidence of an economy based on fishing,
hunting and shellfish gathering; undecorated, sand-tempered
pottery; shell tools; primary, flexed burials in or near shell
middens; and relatively few associated grave goods (Luer and
Almy 1982). The diversity of fish remains found at Manasota
Period sites suggests that these people used a variety offishing
techniques including spears, nets, hooks and dugout canoes
(Luer and Almy 1982). In addition to the two radiocarbon
dates, the artifacts and flexed burials in shell from the
Dunwody Site meet the above criteria for the Manasota

Condition ofRemains

Figure 1 shows the relative frequency of the skeletal
elements recovered from the Dunwody Site. The majority of
skeletal elements are in fairly good condition and can be
handled delicately without breakage. Most elements, however,
are incomplete due to taphonomic processes or excavation
damage. Carnivore damage and weathering, as evidenced by
cortical flaking or bleaching, are absent in the majority of the
elements. Typically present are long bones broken into two or
more fragments that can be re-articulated, or complete long
bone diaphyses that are missing epiphyses. Many of the
smaller, dense, or compact bones remain in a relatively
complete state, such as bones of the hands and feet. There also
are a few instances of complete long bones and one complete
cranium (Burial 14, FLMNH cat. no. 99673). Reconstruction

evidence is clear on many of the crania, undertaken by Scott
Mitchell (former FLMNH Collections Manager) and J. Cobb
(a student) in July 2002. Samples for isotopic analysis were
taken in 2001 from Burials 1, 2, 3, 4, 6, 7, 8, 11, 12, 13, 14,
16, and the disturbed burials (FLMNH cat. nos. 99660-99663,
99665-99667, 99670-99673, 99675, and 99677) (Kelly 2004).
This site shows a high degree of commingling between the
burials, and methods of dealing with this problem are dis-
cussed in the following section.

Materials and Methods

The skeletal remains in this study were examined between
September 2004 and March 2005. All accession, catalog, and
burial numbers used are the originals assigned to the material
by Ripley and Adelaide Bullen in 1965. The remains are
permanently curated at FLMNH.
Standard osteological analyses were used to determine the
minimum number of individuals (MNI), age, sex, stature and
presence of trauma and/or pathology for the Dunwody remains
(Bass 1995; Itcan 1989; Moorrees et al. 1963; Ortner 2003;
Reichs 1998; Scheuer and Black 2000; Ubelaker 1989). A full
inventory of the sample was conducted with all elements
counted and sided, when possible (Appendix II). MNI was
calculated by examining all skeletal elements within the
burials for duplication, taphonomic, age or sex differences. As
mentioned previously, due to the nature of the recovery, no
specific provenience information exists for the burials. It is
assumed that the Bullens considered each burial as a distinct


2006 VOL. 59(l)


unit and gave each a unique catalog number. Although the
skeletal material is divided into 18 distinct burials, there is
evidence of commingling in almost every burial. However, it
was obvious in most of the burials that the majority of ele-
ments present made up one primary individual, while addi-
tional elements were "extra." Extra elements were identified
in each burial as either duplicates of primary elements or
elements that do not match the taphonomy, sex, or age of the
primary individual. These extra elements were then examined
and re-associated when possible. When re-association was not
possible it was determined whether or not these elements
represented another unique individual through basic MNI
Age estimates were determined using a combination of
traits present for each burial. Epiphyseal fusion (Iscan 1989),
dental eruption and calcification (Moorrees et al. 1963;
Ubelaker 1989), dental attrition (Ubelaker 1989), cranial
suture fusion (Ubelaker 1989), length measurements (Scheuer
and Black 2000; Ubelaker 1989), pubic symphysis and
auricular surface appearance (Buikstra and Ubelaker 1994), as
well as general condition of the bone were among the traits
considered when estimating age. Dental eruption and calcifi-
cation, pubic symphysis and auricular surface appearance, and
epiphyseal fusion techniques were used when at all possible as
these methods can lead to a narrower age range estimate.
Some burials provided little evidence as to the individual's age
at death due to the fragmentary nature of the sample, so
general age categories were used. Estimation of sex was
determined predominately by morphological analyses of the
crania and morphological and metric analyses of the
postcrania when available.
Standard measurements (Buikstra and Ubelaker 1994) of
the crania and postcrania were taken whenever possible, and
stature was estimated using Genovds (1967) and Trotter and
Gleser (1958). All elements were examined macroscopically
for pathological lesions and evidence of trauma, while all
anterior teeth were examined for linear enamel hypoplasia and
all cranial elements for porotic hyperostosis. Photographs and
radiographs also were taken of pertinent or unusual conditions
for documentation and diagnostic purposes.
In addition to the numbered burials, there is skeletal
material labeled as "Disturbed Burials" (FLMNH cat. no.
99677). The disturbed burials were inventoried and MNI was
calculated, but no further analyses were conducted since the
entire lot was given the same catalog number and no informa-
tion about specific individuals exists.


Minimum Number ofIndividuals

The skeletal material from the Dunwody Site was origi-
nally excavated as 18 burials and additional skeletal material
from disturbed burials. Although only 18 burials were
indicated, analysis of the remains suggests that there are at
least 21 individuals represented. Although many of the extra
elements were re-associated and therefore did not increase the

MNI, there are elements that must be considered additional
unique individuals. Burial 2-2A (FLMNH cat. no. 99661)
represents at least two individuals (Individual A and Individ-
ual B), based on two crania associated with this catalog
number. Two individuals are commingled within Burial 7
(FLMNH cat. no. 99666) but are distinguishable from one
anotherby a large age difference; remains of neither individual
can be re-associated or accounted for within any of the other
burials. Burial 9 (FLMNH cat. no. 99668) represents a child
(Individual A), but additional elements exist that are those of
an infant (Individual B). Lastly, Burial 13 (FLMNH cat. no.
99672) is an old-aged male (Individual A), and infant ele-
ments (Individual B) are also present and cannot be re-
associated with any of the other burials. The additional
material from the disturbed burials (FLMNH cat. no. 99677)
represents a minimum of 10 individuals; however, it is
unknown whether this material could be accounted for within
the 18 previously defined burials, or if it represents additional
unique individuals.

Population Characteristics

Of the 21 individuals present in this sample, all were
placed into at least "general" age categories, with the subadults
given more specific age ranges (Tables 2 and 3). There are 11
adults consisting of three young adults (20-35), six middle-
aged adults (35-50), and two old adults (50+). Ten subadults
are represented by four infants (0-3), three children (3-12), and
three adolescents (12-20). Sex was determined for the 14
adults and adolescents, with at least eight males and six
females present. Sex could not be determined for the children
and infants, since pre-adolescent individuals display no
secondary sexual characteristics that manifest in the skeleton.
Measurements were taken when possible and used to
facilitate age and sex estimation in some cases. Age estimates
using metric analyses were younger than those determined
through analyses of the dentition since long bone growth is
more easily affected by nutrition while dental development is
maintained except in cases of extreme stress (Larsen 1997).
Due to these differences in age estimation, dentition was relied
upon whenever possible. Sex determination through metric
analyses supported morphological analyses of sex for the most
part with the exception of a few adolescent and old-aged adult
individuals. Means for the measurements are shown in Table
4 with "n" equaling the number of individuals measured in
each case. Measurements and non-metric observations of the
available crania revealed tendencies toward brachycranic, or
round-headed, skulls that were relatively short in length and
height as compared to their breadth. Measurements from
reconstructed crania were not used when reconstruction
interfered with proper technique. Stature was estimated using
diaphyseal measurements and formulae for "Mongoloids" from
Trotter and Gleser (1958) and for "Mesoamericans" from
Genov6s (1967) (Table 5). Stature estimates ranged from
approximately 155 167.5 cm (5-feet 1-inch to 5-feet 6-
inches) for two male individuals.



Table 1. Radiocarbon dates for the Dunwody sample.

FLMNH Beta Measured 13C/12C 15N/14N Conventional 2 Sigma Calibrated
Burial # Analytic # Radiocarbon Age Ratio Ratio Radiocarbon Date
8 203115 1150 40 B.P. -7.2%o +11.9%o 1440 40 B.P. cal A.D. 550 to 660
11 203116 1290 40 B.P. -7.7 %o +10.8 %o 1570 40 B.P. cal A.D. 410 to 580

Table 2. Summary of the sex and age distribution by burial, including the primary sexing and aging criteria.

Burial # Sex Sexing Criteria Age Category Age Estimate Aging Criteria
1 M C Middle Adult 35-50 S/W
2-2A Individual A M C/M Middle Adult 35-50 S/W
Individual B F C/M Middle Adult 35-50 S/W
3 F I/C/M Young Adult 20-35 A/S/W
4 F C/M Middle Adult 35-50 S/W
5 NA NA Infant 1-2 D/M
6 M I/C/M Young Adult 25-40 P/W
7 Individual A F M Adolescent 13-16 F/M
Individual B F C/M Old Adult 45+ D/S
8 M I/C/M Adolescent 16-19 D/F/M
9 Individual A NA NA Child 3-5 D/F/M
Individual B NA NA Infant 0-1 M
10 NA NA Child 3-5 M/Se
11 F C/M Young Adult 18-30 F/W
12 M C/M Middle Adult 35-50 Om
13 Individual A M I/M Old Adult 50+ P/A/Om
Individual B NA NA Infant 0-1 M/Se
14 M C/M Adolescent 13-16 F/D/M
15 M C/M Young Adult 20-35 S/W
16 NA NA Child 5-8 D/F/M
17 NA NA Infant 1-2 D/F/M
C=Cranial Morphology, M=Metric Analysis, I=Innominate Morphology, Mb=Mandible Morphology, S=Suture Closure, W=Tooth Wear, A=Auricular Surface
Morphology, D=Dental Eruption and Formation, P=Pubis Morphology, F=Fusion of Epiphyses, Om=Overall Morphology, Se=Seriation


Table 3. Frequencies of individuals by sex and age categories.





A dult 6 ........................... ............................5 0 .............. ................ ..........11.............
Old 1 1 0 2
Middle 3 2 0 5
Young 2 2 0 4
........................ ................................................................................................. ............................
Subadult ,2 1 7 10
Adolescent 2 1 0 3
Child 0 0 3 3
Infant 0 0 4 4

Totals 8 6 7 21

Paleopathology and Health

Assessment of the overall health of a past population is a
very important step in interpreting their lifestyle through
bioarchaeological analyses. Multiple indicators of physiologi-
cal stress can be used to estimate the health of a past popula-
tion when no written records are available. According to
Larsen (2000:16) dental enamel hypoplasias, porotic
hyperostosis, and infection are three indicators in dental and
skeletal material that are highly informative about health, or
quality of life, and lifestyle. Evidence of each of these condi-
tions, along with evidence of trauma, were documented and
studied in the Dunwody sample. In the following sections
these indicators are discussed for what they reveal generally
about a population, as well as what they reveal specifically
about the health and lifestyle of the Dunwody people.

Dental Enamel Hypoplasia. Enamel hypoplasias consist of a
set of tooth enamel defects that occur when an individual
experiences a"body-wide, metabolic insult sufficient to disrupt
ameloblastic physiology" (Aufderheide and Rodriguez-Martin
1998:405). Thus, these defects provide a record of any life
threatening, yet non-fatal, event that occurs when one's body
is forced to divert energy from non-vital processes, such as
amelogenesis, in order to survive (Aufderheide and Rodriguez-
Martin 1998). According to Larsen (1997:44), hypoplasias
also can result from hereditary anomalies; however, the vast
majority are from stressful events, such as instances of poor
nutrition, disease, or trauma. The permanent crowns of the
incisors, canines, and first molars begin to develop in the first
year after birth and finish development sometime between
three and seven years of age. It is only during this develop-
mental period that hypoplasias caused by stressful events may
develop, because after this time the crown is never remodeled
(Hillson 2000). For this reason, the examination of the dental
enamel for disruptions is informative about stressful events
that occur only during the growth period (Hutchinson and
Mitchem 1996:55).

Defects appear as furrows, steps or pits that surround the
entire circumference of the tooth crown. When defects appear
as "sharply-defined, linear, horizontal grooves of reduced
enamel thickness," they are known as linear enamel
hypoplasia (Aufderheide and Rodriguez-Martin 1998; Hillson
2000). In order to observe hypoplasias, the anterior teeth must
not be worn down more than half their height and they must
be free of calculus that may obstruct viewing defects (Hutchin-
son 2004:67).
Two individuals from the Dunwody Site display evidence
of linear enamel hypoplasia (Tables 6 and 7). Burial 4
(FLMNH cat. no. 99663) is a middle-aged adult female and
Burial 15 (FLMNH cat. no. 99674) is a young adult male.
Burial 4 shows a defect on one canine and Burial 15 has
defects on both central maxillary incisors. However, it is
important to note that many of the burials did not contain a
complete set of teeth, and the anterior teeth were those most
commonly missing. Additionally, even when the anterior teeth
were available for analysis, they usually displayed extensive
wear, as was common with teeth in this sample. These factors
may have masked the quantity of hypoplasias present.

Porotic Hyperostosis. Porotic hyperostosis occurs when the
inner diploe of bone enlarges (marrow hyperplasia) while the
outer table is destroyed, thus giving the bone a porous, coral-
like appearance (El-Najjar et al. 1976; Ortner 2003; Willey
and Smith 1980) (Figure 2). These lesions typically occur on
the cranial vault, within the eye orbits, and occasionally on the
lesser wing of the sphenoid and squamous portion of the
temporal. Typically, porotic hyperostosis results when marrow
hyperplasia occurs while the skull bones are still malleable, for
instance during childhood, and persists long enough to
produce skeletal changes (El-Najar et al. 1976:485). In
adulthood these lesions remain present, although they can
display extensive remodeling. First described by Welcker
(1888) as cribra orbitalia, this condition is also known as
cribra cranii, hyperostosis spongiosa, porotic hyperostosis, or
osteoporosis symmetrica depending on its location (El-Naijar



Table 4. Adult cranial and postcranial measurements by sex.
Male Female
Cranial Measurement i n Mean (mm) n Mean (mm)
Maximum length (g-op) 3 178 3 171
Maximum breadth (eu-eu) X X 2 144.5
Basion-Bregma (ba-b) X X 1 132
Cranial base length (ba-n) X X 1 83
Max. Alveolar breadth (ecm-ecm) X X 1 67
Biauricular breadth (au-au) 3 133.0 2 128.5
Minimum frontal breadth (ft-ft) 1 95.0 1 92
Upper facial breadth (fmt-fmt) 1 98 1 100
Nasal breadth (al-al) X X 1 22
Frontal chord (n-b) 3 112.7 3 111
Parietal chord (b-1) 2 106.5 4 103.5
Occipital chord (1-o) X X 2 98
Mastoid length 4 24 4 21.3
Humerus Max. vertical diam. of head 2 44.0 3 39
Max. diameter at midshaft 1 25.0 X X
Min. diameter at midshaft 1 18.0 X X
Epicondylar breadth 2 66.5 X X
Max length 1 318.0 X X
Femur Max diameter of head 2 42.5 1 39.0
A/P subtroch. diameter 4 27 1 33.0
Trans. subtroch. diameter 4 30.5 1 33.0
Tibia Max. diam at nutrient foramen 4 35.5 1 28.0
Trans. diam. nutrient foramen 4 24.0 1 18.0
Max. distal epiphyseal breadth 1 50.0 X X
Length 1 342.0 X X
Scapula Max breadth 1 102.0 X X
Glenoid Height 1 38.0 1 40.0
Calcaneus Maximum length 1 78.0 X X
Middle breadth 1 42.0 X X
Clavicle Maximum length 2 145.0 X X
Sagittal diam. at midshaft 1 11.0 X X
Vertical diam. at midshaft 1 8.0 X X
Sacrum Sacral base 1 60.0 X X

et al. 1976:477).
Porotic hyperostosis is a non-specific consequence of bone
marrow proliferation caused by such conditions as anemia,
cyanotic congenital heart disease, hereditary spherocytosis, or
pyruvate kinase deficiency among others (El-Najar et al.
1976:477; Mensforth et al. 1978:4). Anemia, the most likely
cause, either can be inherited as with sicklemia or the
thalassemias, or it may be acquired in life (Stuart-Macadam
1985; Willey and Smith 1980). Although "this condition has
been found in skeletal populations of the Old World, in the
Pacific, and in the New World," there is no reason to believe
that the cause must be universally the same (Cybulski
1977:31). Therefore, while porotic hyperostosis occurrences

in areas with malaria may be attributed to an inherited anemia
that may provide protection from the disease, those in the
malaria-free New World are more likely due to an acquired
Originally, acquired anemia was attributed to a predomi-
nantly maize-based diet with a low intake of protein/iron;
however, many more recent reports demonstrate various other
causes of iron-deficiency anemia (Cybulski 1977; El-Najjar et
al. 1976; Hutchinson 2004; Palkovich 1987). Walker (1986)
demonstrates how prolonged breast-feeding, weanling diar-
rhea, and parasitic infections from a seafood-based diet can all
lead to iron-deficiency and possible porotic hyperostosis.
Salvadei and Manzi (2001) divides the causes of acquired iron-


2006 VOL. 59(1)


Figure 2. Cribra orbitalia present on Burial 11
(FLMNH cat. no. 99670).

deficiency anemia into those caused by growth related stress,
dietary stress, and lifestyle and general health conditions.
Therefore, porotic hyperostosis due to acquired iron-deficiency
anemia can be informative in drawing conclusions about a
population's diet, lifestyle and behavior (Salvadei and Manzi
Eight individuals from the Dunwody Site demonstrate
lesions that are characteristic of porotic hyperostosis. Four of
the individuals display both cribra orbitalia and porotic
hyperostosis on the cranial vault, three display porotic
hyperostosis on the vaults only, and one individual displays
only cribra orbitalia. In some of the individuals, the porotic
hyperostosis can be seen radiographically on the cranial vault
as expanded diploe that look like "hairs on end." All of the
individuals with porotic hyperostosis, except Burial 8
(FLMNH cat. no 99667) an adolescent male, are adults at least
over the age of 18. Of these eight individuals with some form
of porotic hyperostosis, five are male and three are female
(Tables 6 and 7). In the Dunwody sample, orbits as well as

Figure 3. Radiograph of undiagnosed pathological lesion within
femur of Burial 9 (FLMNH cat. no. 99668).

vault pieces were often missing due to fragmentation and
many of the crania were reconstructed prior to this study. This
alteration of the crania may have obscured the identification of
porotic lesions on some burials.

Infection. According to Ortner (2003:180), infectious disease
is the single greatest threat to life in human populations of
antiquity. Infectious conditions that leave markers on the
skeleton tend to be subacute, chronic diseases and not neces-
sarily the immediate cause of death. However, these markers
can be used as a nonspecific indicator of stress. Bone inflam-
mation is such an indicator that can be triggered by infection
or other traumatic events (Ortner 2003:181). Periostitis is
inflammation of the periosteum that primarily affects the outer
surfaces of bone. The inner layer of the periosteum retains its
osteoblastic capacity throughout life, so when stimulated by an
outside insult, like infection, it will respond with abnormal
bone growth. Sometimes the morphology of the periostitis
reflects the type of pathological stimulus; however, periosteal
lesions are more commonly undifferentiated, non-specific and
non-diagnostic, although useful as a general indicator of stress
(Ortner 2003).
Three individuals from the Dunwody Site display periosteal
lesions on at least one element of their skeleton. Two of the
individuals are young adult females and one individual is a
young adult male. Burial 3 (FLMNH cat. no. 99662) and
Burial 11 (FLMNH cat. no. 99670), both young adult females,
show diffuse periostitis on multiple long bones including the
femur, tibia, fibula, humerus, radius and ulna. Burial 6
(FLMNH cat. no. 99665), however, a young adult male, only
shows slight periostitis on one tibia.
Osteomyelitis is another type of inflammatory response, but
unlike periostitis, it begins in the marrow and primarily affects
the endosteal (inner) surface (Ortner 2003:181). Most often
osteomyelitis is caused by a bacterial infection, Staphylococ-
cus; however, viruses, fungi and multicelled parasites also can
infect the marrow (Ortner 2003; White 2000). The infection
can enter the bone directly through a wound, spread from an
adjacent soft tissue wound, or spread hematogenously from a
remote septic focus (Ortner 2003:181). Osteomyelitis is best
identified by openings in the surface of the bone,
called cloacae, which serve as drainage canals for
the wound. Bone destruction, reactive new bone
formation, and sequestra, or sections of necrotic
bone, are characteristic of osteomyelitis
(Aufderheide and Rodriguez-Martin 1998; Ortner
2003). One young adult female (Burial 11,
FLMNH cat. no. 99670) has evidence of
osteomyelitis on her right femur.
Additionally, when Burial 9 (FLMNH cat. no.
99668) was radiographed, the right femur dis-
played a radiolucent lesion in the metadiaphyseal
region. The lesion is centrally situated in the
ht medullary cavity and surrounded by a sharply
S defined border that separates this lesion from two
additional smaller lesions located more proximally.



Table 5. Stature estimates for Burial 13 (FLMNH cat. no. 99672) and Burial 8 (FLMNH cat. no. 99667).
Element Burial # Trotter and Gleser 1958 Genov6s 1967
Tibia 13 166.06 +/- 2.812 cm 160.78 +/- 3.27 cm
(64.27 to 66.48") (62.01 to 64.59")

Tibia 8 164.14 +/- 2.812 cm 159.02 +/- 3.27 cm
(63.51 to 65.73") (61.32 to 63.89")

Femur 8 161.58 +/- 3.417 cm 159.94 +/- 3.80 cm
(62.27 to 64.96") (61.47 to 64.46")
**Although Burial 8 is a subadult, this individual's femoral and tibial epiphyses are fused thus allowing
stature estimation.

There also may be a developing lesion more distally in the
metadiaphysis, but none of the lesions seem to be affecting the
cortex of the bone (Figure 3). Other bones from Burial 9 were
radiographed in order to facilitate diagnoses, but all other
radiographs appeared normal.
Based on appearance, the main lesion could be either a
unicameral bone cyst or a symptom of fibrous dysplasia. The
two conditions are indistinguishable in appearance. A
unicameral bone cyst usually occurs within the cancellous
portions of long bone metaphyses, but it may migrate to the
diaphyses through time. Bone cysts are sclerotic and sharply
defined. Fibrous dysplasia is usually unilateral, and affects
only one bone (monostotic) 80% of the time. The lower limb
is most commonly affected (Aufderheide and Rodriguez-
Martin 1998). Due to the similarities between these two
conditions and the often monostotic nature of fibrous
dysplasia, neither condition could be confirmed or ruled out
for Burial 9's lesion.

Trauma. Trauma is often informative about the external forces
acting upon an individual. The lifestyle, environment and/or
occupation of a particular individual may be suggested by
skeletal trauma, and interpretations about the population as a
whole may be inferred. Incidences of trauma and healing in
a population may indicate the prevalence of interpersonal
violence, environmental hazards, or healthcare (Ortner 2003;
Roberts and Manchester 1999; White 2000).
Two of the Dunwody individuals display evidence of
antemortem trauma. Individual A from Burial 2 (FLMNH cat.
no. 99661) has post-traumatic reactive bone present on the
distal end of the left humerus. This bone partially fills both
the olecranon and coronoid fossae and there is macroporosity
and eburnation present on the capitulum (Figure 4). The head
of the left radius also displays signs of trauma, including some
bone resorption and a high degree of eburnation.
All of this trauma could be the result of a Monteggia
fracture of the proximal ulna, which then caused the displace-
ment of the radius. The destruction of the elbow joint capsule,
broken ulna, and displaced radius could have stimulated this
new, abnormal bone growth. Monteggia fractures are classi-
fied into types based on the direction of radial displacement.

This incident was most likely a Type I where the radius is
displaced anteriorly. This can be caused by falling on an
outstretched arm or by force applied to the pronated forearm
(Browner et al. 2003). This individual is an adult male and it
is doubtful that he would have been able to extend his arm
fully, due to the extensive bone growth in the olecranon fossa.
Burial 13 (FLMNH cat. no. 99672), an old adult male, also
shows signs of trauma on the humerus. This individual,
however, shows trauma on the proximal right humerus (Figure
5). There is bony lipping directed inferiorly off the inferior
border of the humeral head and the head appears inferiorly
displaced. There is a possibility that this individual suffered
a surgical neck fracture of the humerus (Browner et al. 2003).
Surgical neck fractures are commonly produced when an
individual falls from a standing height and lands either on
his/her shoulder or outstretched arm. There also tends to be an
increased risk of this type of fracture in the elderly, due to
decreased bone density (Court-Brown et al. 2001). This
trauma may have been the result of an accidental fall, espe-
cially due to the comparatively advanced age (50 or more years
old) of the individual.


Subsistence and Diet

Diet is an additional factor that must be discussed as it
affects a population's health, and thus can be interpreted
through skeletal markers. Suggestions of diet can be inferred
from the presence or absence and frequencies of pathologies;
however, historical, archaeo-botanical, zooarchaeological and
isotopic analyses are perhaps better ways to determine accu-
rately the contents of a population's diet. Although there is
minimal information regarding the diet of individuals at the
Dunwody Site specifically, there is a wealth of information
about subsistence around Lemon Bay and the larger Central
Gulf Coast.
In Luer's (1999b) "Introduction to Maritime Archaeology
of Lemon Bay," he explains that there has been a long history
of cultures dependant on the littoral environment of Lemon
Bay. Archaeo-botanical and zooarchaeological evidence from


2006 VOL. 59(1)


Figure 4. Traumatic remodeling on the left humerus of Burial 2 (FLMNH cat. no. 99661): a) anterior view; b) posterior view.

sites dating from Archaic Period times through the Safety
Harbor Period have provided evidence of strong dependence on
maritime resources. Luer states that Lemon Bay Indians
caught many kinds of edible fish, gathered shellfish and
plants, and hunted small mammals and reptiles. Similar
evidence is presented by Ardren et al. (2003). There is no
evidence that these people practiced maize agriculture,
although they may have kept their own small gardens and
useful trees (Luer 1999b:4).
This assessment is consistent with Hutchinson's (1993,
2004) interpretations of subsistence for the Florida Central
Gulf Coast, as well as definitions of subsistence for the
Manasota Culture (Luer and Almy 1982). Hutchinson states
that most Florida populations did not transition into maize
agriculture until after European contact, unlike their neighbors
to the north. Due to the abundant aquatic resources available
and soil not favorable to maize agriculture, many native
Floridians were able to subsist primarily by foraging into late
prehistory (Hutchinson 2004:5). Isotope analyses of human
bone from Central Gulf Coast sites, such as Palmer Mound
and Tierra Verde Mound, indicate diets heavy in marine fauna
and terrestrial tubers and fruits. Even sites located farther
inland reveal people dependent on freshwater and terrestrial
animals with little indication of maize dependence. Northern
Florida is the only area where any evidence of maize depend-
ence has been found (Hutchinson 1993, 2004).
Limited information about the probable diet of the
Dunwody sample exists in the form of zooarchaeological
information from the Cedar Point Shell Heap. "A wide variety
of 'food shell' was observed in the remnants of Cedar Point
Shell Heap," in addition to remains of deer, sea turtle and bony
fish (Luer 1999a:49).

consistent with the Manasota Culture. Kelly interprets less
negative apatite values (n =12; bone apatite '3C = 5.2%o; n
= 10; enamel apatite '3C = 4.5%o) as indicative of a C4 plant
and of maize specifically; however, this may be based on later
dates (ca. A.D. 800 to 1750) when maize consumption might
have been a possibility at the Dunwody site (2004:73,85).
Kelly (2004) acknowledges that radiocarbon dating is neces-
sary to secure a more specific date for the site, and that "the
isotopic results do not support the ethnohistoric documenta-
tion," which, as mentioned in this paper, provides no evidence
for maize consumption in this geographical area (85). Had the
radiocarbon dates been available to Kelly at the time of her
analyses, perhaps she would have altered her interpretations
accordingly. There are other explanations for apatite values
indicating C4 vegetation, as Kelly posits for the Horr's Island
and Pillsbury sites included in her study (2004:85-86,91).
Kelly states that these values may be attributed to non-maize
plants, such as marine grasses, that were either consumed
directly or indirectly as the result of terrestrial animal and/or
marine fish consumption (2004:85-86,91). It would seem that
in light of the radiocarbon dates reported here, the above
interpretation is a possibility for the Dunwody site. Thus, with
the exception of Kelly's (2004) interpretation of the apatite
values, all other data remains consistent with the Dunwody site
belonging to the Manasota Culture dependent on fishing,
hunting, and gathering. Further analysis is suggested in order
to truly determine the composition of the C4 part of their diet.

Comparisons with other Coastal Florida Populations

In order to understand more about the health and lifestyle
of these individuals from the Dunwody Site, it is necessary to



Table 6. Pathology occurrence by burial.

Burial # Enamel Hypoplasia Porotic Hyperostosis Periostitis Osteomyelitis Undiagnosed Lesion Trauma

1 1
2 1 1
3 1 1
4 1 1
6 1 1
8 1
9 1
11 1 1 1
13 1
15 1 1
Totals 2 8 3 1 1 2

Table 7. Pathology occurrence by sex and age groups.

Sex Age Enamel Hypoplasia Hyperostosis Periostitis Osteomyelitis Trauma
Male ..... 1 51 0 2
Adult Old 0 0 -- --- ------------ -------- ^---- -----------^ ---
Adult Old 0 0 0 0 1
Middle 0 2 0 0 1
Young 1 2 1 0 0

Sub-adult Adolescent 0 1 0 0 0
---------------------------------------------------------------------------------------- ---------------
Female 1 3 2 1 0
...Female ...2----- --------------------0
Adult Middle 1 1 0 0 0
Young 0 2 2 1 0


Figure 5. Possible surgical neck fracture of the humerus
of Burial 13 (FLMNH cat. no. 99672).

samples. The sites used for comparative purposes are
ManasotaKey (Dickel 1991), Palmer (Hutchinson 2004), Aqui
Esta (Hutchinson 2002), Tierra Verde (Hutchinson 1993),
Sarasota Bay Mound (Freas and Warren 2005), and Weeki
Wachee (Hutchinson and Mitchem 1996).
Enamel hypoplasias are indicative of extremely stressful
events, such as malnutrition and disease, in the young life of
an individual. According to Hutchinson (2004), the preva-
lence of enamel hypoplasia is either very high or very low
among groups in Florida. For Dunwody this prevalence is
relatively low with only 25% (n = 2/8) of individuals display-
ing linear enamel hypoplasia. Although this percentage is
slightly higher than that of Aqui Esta (23%) (Hutchinson
2002), Manasota Key (7.7%) (Dickel 1991), and Tierra Verde
(2%) (Hutchinson 1993), it is quite lower than that of the
Palmer Site (77.4%) (Hutchinson 2004), Weeki Wachee (75%)
(Hutchinson and Mitchem 1996), and Sarasota Bay Mound
(87.5%) (Freas and Warren 2005) (Figure 6). This low
percentage of afflicted individuals from the Dunwody Site
could be suggestive of healthy individuals who had access to
stable food resources and low incidences of serious disease.
Increased occurrences of hypoplasia often are linked to the
transition to agriculture or contact with Europeans, which
usually led to more sedentary and concentrated populations.
This type of living situation is more conducive to poor sanita-
tion and the spread of diseases (Hutchinson 2004:108).
Therefore, the low frequency of hypoplasia at Dunwody could
be attributed to the time in which these people lived. The
Manasota Period was well before contact, and the dates for this
site specifically (A.D. 410-660) suggest that these people lived
in smaller settlements before the appearance of chiefdoms
(post A.D. 700) and larger populations (Luer 1999b). How-
ever, this low frequency also could be attributed to sample bias
as only eight out of 21 individuals have the appropriate
dentition to allow analysis of hypoplasia.
Although caries were not quantitatively analyzed at
Dunwody, the dentition present is mostly free from carious

lesions. High percentages of caries are associated with maize
agriculture, which did not occur in Florida until around the
time of contact (Hutchinson 2004). The low occurrence of
caries at Dunwody suggests no maize consumption, which is
consistent with radiocarbon dates in the late Manasota Period.
Approximately 53% (n = 8/15) of Dunwody individuals
show symptoms of porotic hyperostosis. This frequency is
higher than those of five nearby coastal sites that range from
4.3% at Weeki Wachee (Hutchinson and Mitchem 1996) to
30% at Aqui Esta (Hutchinson 2002) (Figure 7). Only
Sarasota Bay Mound (Freas and Warren 2005) has a similarly
high rate of porotic hyperostosis (45%) and cribra orbitalia
(60%). Hutchinson (2002, 2004) indicates that porotic
hyperostosis is common in southeastern coastal populations,
and that incidences of this condition are much higher in
coastal compared to non-coastal regions. Although increased
levels of porotic hyperostosis are often linked with dependence
on maize and contact with Europeans in other parts of the
United States, the early date of this site negates this possible
cause. Additionally, the consistency of porotic hyperostosis
frequencies over time in coastal Florida archaeological
populations suggests another cause. This consistency suggests
that similar conditions must have existed in coastal Florida
from pre-agricultural pre-contact days to post-contact agricul-
tural days, in order to produce similar numbers of affected
individuals. One consistent condition throughout Florida
history is the abundant availability of marine resources for
consumption by indigenous people. Parasites found in these
kinds of foods as well as contaminated water sources are the
most likely causes for porotic hyperostosis in these people
(Hutchinson 2004). As is the case with the Sarasota Bay
Mound sample (Freas and Warren 2005), this higher level of
incidence at Dunwody may be the result of the small sample
size, or alternatively it may relate to increases in parasite load
or water contamination.
The prevalence of periostitis in the Dunwody sample is
around 16% (n = 3/19). This percentage is similar to those
found at Aqui Esta (9%) (Hutchinson 2002), Manasota Key
(17.6%) (Dickel 1991), Tierra Verde (12.5%) (Hutchinson
1993), and Weeki Wachee (13.1%) (Hutchinson and Mitchem
1996) (Figure 8). The large difference in frequency between
Dunwody and Sarasota Bay Mound (69%) (Freas and Warren
2005) is probably explained by the latter's small sample size.
Since periostitis is caused by numerous factors, it is hard to
determine the exact causal mechanism for these reactions in
the Dunwody individuals. Periosteal lesions are very common
in North American skeletal populations (Hutchinson and
Mitchem 1996:54), so this relatively low percentage at
Dunwody suggests a healthy population of individuals.
Hutchinson (2004) states that health problems and thus
incidences of periosteal reactions increase with European
contact and the adoption of agriculture, especially in the
Southeast. Therefore, this low prevalence is consistent with
the radiocarbon dates in the Manasota Period.
Since the Dunwody sample is represented by incomplete
and fragmented burials, it is possible that more individuals
were afflicted with periostitis than were possible to count.





Manasota Key


Aqui Esta

Tierra Verde

Sarasota Bay Mound

Weeki Wachee

I Mid-Late Woodland

20 40 60 80 100
% Affected Individuals

Mississippian-Influenced Early Contact

Figure 6. Comparison of percentage of individuals affected by enamel hypoplasia in Florida Gulf coast populations. Mid-
Late Woodland = ca. A.D. 0-1000; Mississippian-Influenced = ca. A.D. 1000-1525; Early Contact = 1525-1550. Sources:
Dunwody (this paper), Manasota Key (Dickel 1991), Palmer (Hutchinson 2004), Aqui Esta (Hutchinson 2002), Tierra
Verde (Hutchinson 1993), Sarasota Bay Mound (Freas and Warren 2005), Weeki Wachee (Hutchinson and Mitchem


Manasota Key


Aqui Esta

Tierra Verde

Sarasota Bay

Weeki Wachee

7 53.3

.' 26






0 10 20 30 40
% Affected Individuals

li Mid-Late Woodland

50 60

Mississippian-Influenced O Early Contact

Figure 7. Comparison of percentage of individuals affected by porotic hyperostosis in Florida Gulf coast populations.
Mid-Late Woodland = ca. A.D. 0-1000; Mississippian-Influenced = ca. A.D. 1000-1525; Early Contact = 1525-1550.
Sources: Dunwody (this paper), Manasota Key (Dickel 1991), Palmer (Hutchinson 2004), Aqui Esta (Hutchinson 2002),
Tierra Verde (Hutchinson 1993), Sarasota Bay Mound (Freas and Warren 2005), Weeki Wachee (Hutchinson and
Mitchem 1996).


2006 VOL. 59(1)



Manasota Key


Aqui Esta

Tierra Verde

Sarasota Bay

Weeki Wachee

0 10 20 30 40 50
% Affected Individuals

60 70 80

SMid-Late Woodland 1 Mississippian-Influenced Early Contact

Figure 8. Comparison of percentage of individuals affected by periostitis in Florida Gulf coast populations. Mid-Late
Woodland = ca. A.D. 0-1000; Mississippian-Influenced = ca. A.D. 1000-1525; Early Contact = 1525-1550. Sources: Dunwody
(this paper), Manasota Key (Dickel 1991), Palmer (Hutchinson 2004), Aqui Esta (Hutchinson 2002), Tierra Verde
(Hutchinson 1993), Sarasota Bay Mound (Freas and Warren 2005), Weeki Wachee (Hutchinson and Mitchem 1996).

Additionally, it is important to note that while no diagnosis
can be made regarding the cause for the periosteal lesions, at
the same time a cause such as treponemal infection cannot be
ruled out. Treponemal infections (e.g., syphilis, bejel and
yaws) are caused by organisms entering the host through the
skin or mucous membrane. Both yaws and acquired tertiary
syphilis can cause periostitis. In syphilis, typically more than
one bone is affected bilaterally, with the tibia most commonly
affected (Aufderheide and Rodriguez-Martin 1998; Ortner
2003:274). Two of the Dunwody individuals (Burial 3,
FLMNH cat. no. 99662; Burial 11, FLMNH cat. no 99670) are
affected with diffuse periostitis in this way. Hutchinson (2004)
states that it is likely that proliferative responses found on the
skeletal material of native Floridians are due to treponemal
Osteomyelitis also can be an indication of treponemal
infection, as well as numerous other conditions. Around 5%
(n = 1/19) of Dunwody individuals display evidence of
osteomyelitis as compared to 0.5% at Palmer (Hutchinson
2004), 2.1% at Tierra Verde (Hutchinson 1993), and 2.4% at
Weeki Wachee (Hutchinson and Mitchem 1996) (Figure 9).
Although Dunwody has a slightly higher occurrence than these
nearby sites, this percentage is most likely inflated due to the
small sample size. As Hutchinson (2004) indicates, each
particular area of Florida developed its own specific economic
base, with neighboring regions often exploiting different
resources, which may have led to different qualities of life and
health. This uniqueness of experiences may have allowed the

Dunwody Site people to be exposed to slightly higher inci-
dences of infection due to some difference between their
lifestyle and those of neighboring peoples. This amount of
osteomyelitis, although slightly higher than nearby sites, seems
consistent with that found in healthy Florida coastal popula-
tions. Since high frequencies of infections can be suggestive
of a densely populated area where sanitation is poor, it seems
that the lack if infections seen in these populations is due to
the opposite conditions of more hygienic, dispersed popula-
tions (Larsen 2000:20). These conditions probably would have
been met for the Dunwody Site during the Manasota Period
before the appearance of chiefdoms.
Trauma analysis can be helpful in reconstructing behavior,
lifestyle, or environmental dangers from individual skeletons
in a population. Only 10.5% (n = 2/19) of individuals from
the Dunwody Site exhibit any trauma. This percentage is
consistent with that found at Manasota Key (13%) (Dickel
1991), but higher than Palmer (4%) (Hutchinson 2004) and
Weeki Wachee (1.2%) (Hutchinson and Mitchem 1996)
(Figure 10). Dickel states that the trauma incidence rate at
Manasota Key is low but resembles rates determined from
other relevant samples, such as from the Windover Site, as
well as patterns seen in hunter-gatherer communities
(1991:47). While it is important to note that only 19 Dunwody
individuals are complete enough for trauma examination, the
evidence from the much larger Manasota Key sample suggests
that the similar Dunwody percentage could be comparable to
other samples in the area. As at Manasota Key (Dickel







Tierra Verde

Weeki Wachee

S 5.3




) 1 2 3 4 5
% Affected Individuals

111 Mid-Late Woodland O Mississippian-Influenced L

Early Contact

Figure 9. Comparison of percentage of individuals affected by osteomyelitis in Florida Gulf coast populations. Mid-Late
Woodland = ca. A.D. 0-1000; Mississippian-Influenced = ca. A.D. 1000-1525; Early Contact = 1525-1550. Sources:
Dunwody (this paper), Palmer (Hutchinson 2004), Tierra Verde (Hutchinson 1993), Weeki Wachee (Hutchinson and
Mitchem 1996).


Manasota Key


Weeki Wachee

0 2 4 6 8 10 12 14
% Affected Individuals

Mid-Late Woodland Early Contact

Figure 10. Comparison of percentage of individuals affected by trauma in Florida Gulf coast populations. Mid-Late
Woodland = ca. A.D. 0-1000; Mississippian-Influenced = ca. A.D. 1000-1525; Early Contact = 1525-1550. Sources:
Dunwody (this paper), Manasota Key (Dickel 1991), Palmer (Hutchinson 2004), Weeki Wachee (Hutchinson and Mitchem


2006 VOL. 59(1)


1991:43-47), both of the Dunwody injuries display no evidence
of interpersonal aggression and are most likely the results of

Causes ofDeath

Although many aspects affecting the health of Dunwody
individuals have been considered extensively, the circum-
stances surrounding these individuals' deaths still warrants
discussion. Overall, this sample shows a typical age distribu-
tion with a large percentage (47.6%) of subadult individuals.
This pattern is common in prehistoric populations (Roberts
and Manchester 1999) as the risk of death is highest in infancy
and early childhood. However, most of these subadults show
no skeletal signs of disease or cause of death. It is important
to remember that although these individuals may be free from
those chronic diseases that leave skeletal evidence, they still
could have been afflicted with acute infective diseases that
could have killed them before sufficient time elapsed to leave
skeletal evidence (Roberts and Manchester 1999). Addition-
ally, more evidence as to the cause of death may have existed,
but since pathological bone is often more delicate than healthy
bone, this bone may not have survived. Thus, incidences of
pathology and suggestions as to cause of death may be under-
represented (Roberts and Manchester 1999). Another common
cause of death in prehistoric populations is childbirth, and two
of the infants in this sample are young enough to have died in
this way. Furthermore, a few of the females are of childbear-
ing age and could have died giving birth.


Based on skeletal evidence, the individuals from the
Dunwody Site represent a sample that is relatively free from
severe disease. The Dunwody Site, located along the shore of
Lemon Bay, Florida, dates to late in the Manasota Period, ca.
A.D. 410 to 660. Although there were no indicators to suggest
causes of death, pathological markers suggest afflictions
common to populations along the Central Gulf Coast of
Florida. Low incidences of enamel hypoplasia and
proliferative lesions suggest that these individuals were
relatively well nourished and free from chronic diseases. This
was probably due to living in an environment of low popula-
tion density with a dependable diet concentrated in marine
resources, as is characteristic of the Manasota Culture.
However, the prevalence of subadults in the sample suggests
that these individuals may have suffered from acute infections
or diseases that left no skeletal markers. A high percentage of
porotic hyperostosis in the population could be attributed to a
contaminated water supply or a high parasitic load related to
the large quantities of marine fauna ingested. It is important
to note, though, that small sample size may have altered
interpretations and numbers for each of these conditions.
Nonetheless, the health of the Dunwody individuals is consis-
tent with other Native American skeletal populations from the
Central Gulf Coast of Florida.


' The Dunwody Site was named after the landowners, the Dunwody
family (pronounced "Dunwoody"). They acquired the land in the
1930s and sold it to Charlotte County in 1992 for creation of Cedar
Point Environmental Park, an environmental and educational
preserve operated by the Charlotte Harbor Environmental Center, and
the Charlotte County Parks, Recreation, and Cultural Resources


I would like to thank George Luer for bringing the Dunwody Site
to my attention and providing me with invaluable information and
resources that allowed me to write this paper. Dr. William
Marquardt and the staff at the Florida Museum of Natural History
allowed me access to the collections, space for research, and the
opportunity to conduct special analyses. Donna Ruhl was especially
important in facilitating my research at the museum. I also would
like to thank Dr. Michael Warren for his tireless help and support,
both financially and emotionally, and Megan Hosford for her

References Cited

Ardren, Traci, Bill Burger, and Brian Keith Sullivan
2003 The Mystery River Point Site, Englewood, Sarasota
County, Florida. The Florida Anthropologist 56(1):47-55.

Aufderheide, A. C., and C. Rodriguez-Martin
1998 The Cambridge Encyclopedia ofHuman Paleopathology.
Cambridge University Press, Cambridge.

Bass, William M.
1995 Human Osteology: A Laboratory and FieldManual.
Fourth edition. Missouri Archaeological Society, Colum-

Browner, Bruce D., Jesse B. Jupiter, Alan M. Levine, and Peter G.
2003 Skeletal Trauma: Basic Science, Management, andRecon-
struction. Third edition, Volume II. Elsevier Science,

Buikstra, Jane E., and Douglas H. Ubelaker (editors)
1994 Standards: Proceedings ofa Seminarat TheField Museum
of Natural History. Arkansas Archaeological Survey
Research Series Number 44, Fayetteville.

Bullen, Ripley P.
1971 The Sarasota County Mound, Englewood, Florida. The
Florida Anthropologist 24(1):1-30.

Court-Brown, Charles M., Ashima Garg, and Margaret M. McQueen
2001 The Epidemiology of Proximal Humeral Fractures. Acta
Orthopaedica Scandinavica 72(4):365-371.

Cortes, Josephine
1965 Burial Mound Study Set If Any Bones Are Left: Archae-
ologist Arrives Today. Sarasota Herald-Tribune. Copy on
file, Sarasota County History Center, Sarasota.




Cybulski, J. S.
1977 Cribra Orbitalia, a Possible Sign of Anemia in Early
HistoricNative Populations of the British Columbia Coast.
American Journal ofPhysical Anthropology 47(1):31-39.

Dickel, David N.
1991 Descriptive Analysis of the Skeletal Collection from the
Prehistoric Manasota Key Cemetery, Sarasota County,
Florida (8S01292). Florida Archaeological Reports
Number 22, Tallahassee.

El-Najjar, M. Y., D. J. Ryan, C. G. Turner II, and B. Lozoff
1976 The Etiology and Porotic Hyperostosis among the Prehis-
toric and Historic Anasazi Indians of Southwestern United
States. American Journal of Physical Anthropology

Freas, Laurel, and Michael W. Warren
2005 Osteological Analysis of the Sarasota Bay Mound: A Safety
Harbor Period Site. The Florida Anthropologist 58(1-

Genov6s, Santiago
1967 Proportionality of the Long Bones and their Relation to
Stature among Mesoamericans. American Journal of
Physical Anthropology 26(1 ):67-78.

Hillson, Simon
2000 Dental Pathology. In Biological Anthropology ofthe
Human Skeleton, edited by M. A. Katzenberg and S. R.
Saunders, pp. 249-286. Wiley-Liss, Inc., New York.

Hutchinson, Dale L.
1993 Analysis of Skeletal Remains From the Tierra Verde Site,
Pinellas County, West-Central Florida. The Florida
Anthropologist 46(4):263-276.

2002 Osteological Analysis of the Aqui Esta Mound Population.
InArchaeology ofUpper Charlotte Harbor, Florida, edited
by George M. Luer, pp. 183-193. Florida Anthropological
Society Publications Number 15, Tallahassee.

2004 Bioarchaeology of the Florida Gulf Coast: Adaptation,
Conflict, and Change. University Press of Florida,

Hutchinson, Dale L., and Jeffrey M. Mitchem
1996 The Weeki Wachee Mound, An Early Contact Period
Mortuary Locality in Hernando County, West-Central
Florida. Southeastern Archaeology 15(1):47-64.

1tcan, Mehmet Yaar (editor)
1989 Age Markers in the Human Skeleton. Charles C. Thomas,
Springfield, Illinois.

Kelly, Jennifer Ann
2004 StablelsotopeEvidenceforMaize Consumption and Other
Dietary Practices at Bayshore Homes (8Pi41) and other
Prehistoric Sites in Peninsular Florida. Unpublished
Master's thesis, Department of Anthropology, University
of South Florida, Tampa.

Larsen, Clark Spencer
1997 Bioarchaeology: Interpreting Behavior from the Human
Skeleton. Cambridge University Press, Cambridge.

2000 Skeletons in Our Closet: Revealing Our Past Through
Bioarchaeology. Princeton University Press, Princeton.

Luer, George M.
1999a Cedar Point: A Late Archaic Through Safety Harbor-Period
Occupation on Lemon Bay, Charlotte County, Florida. In
Maritime Archaeology ofLemon Bay, Florida, edited by
George M. Luer, pp. 43-56. Florida Anthropological
Society Publication Number 14, Tampa.

1999b An Introduction to the Maritime Archaeology of Lemon
Bay, Florida. In Maritime Archaeology of Lemon Bay,
Florida, edited by George M. Luer, pp. 1-22. Florida
Anthropological Society Publication Number 14, Tampa.

Luer, George M., and Marion M. Almy
1982 A Definition of the Manasota Culture. TheFlorida
Anthropologist 35:34-58.

Mensforth, R., C. Lovejoy, J. Lallo, and G. Armelagos
1978 The Role of Constitutional Factors, Diet, and Infectious
Disease in the Etiology of Porotic Hyperostosis and
Periosteal Reactions in Prehistoric Infants and Children.
Medical Anthropology 2:1-59.

Mitchem, Jeffrey M.
1989 Redefining Safety Harbor: Late Prehistoric/Protohistoric
Archaeology in West PeninsularFlorida. Ph.D. Disserta-
tion, Department of Anthropology, University of Florida,

Moorrees, Coenraad F. A., Elizabeth A. Fanning, and Edward E.
Hunt, Jr.
1963 Age Variation of Formation Stages for Ten Permanent
Teeth. Journal ofDental Research 42(6):1490-1502.

Ortner, Donald J.
2003 Identification of Pathological Conditions in Human
Skeletal Remains. Academic Press, San Diego.

Palkovich, A. M.
1987 Endemic disease patterns in paleopathology: porotic
hyperostosis. America Journal ofPhysicalAnthropology

Reichs, Kathleen J. (editor)
1998 Forensic Osteology: Advances in the Identification of
Human Remains. Second Edition. Charles C. Thomas,
Springfield, Illinois.

Roberts, Charlotte, and Keith Manchester
1999 The Archaeology of Disease. Second Edition. Cornell
University Press, Ithaca.

Salvadei, L., F. Ricci, and G. Manzi
2001 Porotic Hyperostosis as a Marker of Health and Nutritional
Conditions during Childhood: studies at the transition
between Imperial Rome and the Early Middle Ages.
American Journal ofHuman Biology 13(6):709-717.


2006 VOL. 59(1)


Scheuer, Louise, and Sue Black
2000 Developmental Juvenile Osteology. Academic Press, San

Stuart-Macadam, P.
1985 Porotic hyperostosis: Representative of a Childhood
Condition. American Journal of Physical Anthropology

Trotter, Mildred, and Goldine C. Gleser
1958 A Re-evaluation of Estimation of Stature Based on Mea-
surements of Stature Taken During Life and of Long Bones
After Death. American Journal ofPhysical Anthropology

Ubelaker, Douglas H.
1989 Human Skeletal Remains: Excavation, Analysis, Interpre-
tation. Second Edition. Taraxacum, Washington D.C.

Walker, P. L.
1986 Porotic Hyperostosis in a Marine-Dependent California
Indian Population. American Journal ofPhysical Anthro-
pology 69(3):345-354.

Welcker, H.
1888 Cribra Orbitalia, Ein ethnologischdiagnostisches Merkmal
am Shadel mehrerer Menschenrassen. Archiv Fur
Anthropologie 17(1).

White, Tim D.
2000 Human Osteology. Second Edition. Academic Press, San

Willey, Patrick S., and Fred H. Smith (editors)
1980 Skeletal Biology ofAboriginal Populations in the South-
eastern United States. Tribute Press and Print Co.,



Appendix I. Radiocarbon dates from sites around Lemon Bay, Florida. The measured and conventional ages are in
radiocarbon years before present (B.P.; present = A.D. 1950). All ages and 13C/12C year corrections are rounded to the
nearest ten. All 13C/12C ratios are rounded to the nearest tenth. Asterisks indicate estimated ratios and ages. The values
for stable isotopes, the 13C/12C ratios, are typical of the kinds of materials dated. Equivalent values (in years) for 13C/12C
ratios (at 1 o/oo = 16.4 years) are stated, and these are reflected in the corrected ages. All calibrated dates in this table have
been derived by Beta Analytic, Inc., using the Intcal98 Radiocarbon Age Calibration. Dates for the Dunwody Site are based
on Gold (this article). The uncorrected date from Paulsen Point Midden is based on Bullen (1971:13), and those for Manasota
Key Cemetery are in Dickel (1991:2). Uncorrected dates and 13C/12C ratios for Mystery River Point are based on Ardren
et al. (2003:Table 1). One sigma age ranges have 68% probability, and two sigma age ranges have 95% probability. This
legend and table were provided by George Luer.

Provenience, Lab ID# or Measured, 13C/12C Ratio, Conventional, Cor- Calibrated, Calendrical
Submitter's ID#, Material Uncorrected Age and Equivalent reacted Age B.P., 1 Range, 2 Sigma
B.P., Value in Years Sigma
1 Sigma (to nearest 10) (* indicates

Dunwody Site (8CH61)

1. Burial 8, Beta-203115, 1150+/-40 -7.2 (17.8 x 16.4 1440 +/-40 cal A.D. 550-660
bone = 290)

2. Burial 11, Beta-203116, 1290+/-40 -7.7, (17.3 x 16.4 1570+/-40 cal A.D. 410-580
bone 280)

Paulsen Point Midden

1. Pit 3, Level 5, 1650 +/- 50 -25* (0 x 16.4 = 1650 +/- 50* cal A.D. 260-530
FSU-332, "charcoal" 0)

Manasota Key Cemetery
(8S01292) _

1. Upper midden, FS#613, 1660 +/- 80 0* (25 x 16.4 = 2070 +/- 80* cal A.D. 110-470
fighting conch 410)

2. Upper midden, FS#981, 1630 +/- 80 0* (25 x 16.4 = 2040 +/- 80* cal A.D. 140-540
fighting conch 410)

3. Lower midden, FS#763, 1830 +/- 90 0* (25 x 16.4 = 2240 +/- 90* cal 100 B.C. to
unspecified clam shell 410) cal A.D. 330

4. Burial #25, 1800 +/- 80 -7.5 (17.5 x 16.4 2090 +/- 80 cal 370 B.C. to
FS#591, bone =290) cal A.D. 70

Mystery River Point

1. Test Unit B, 1610+/-50 -27.0 (2 x 16.4 = 1580+/-50 cal A.D. 390-600
UGa-5662, wood 30)

2. Test Unit B, 1610+/-60 -28.2 (3.2 x 16.4 1560+/- 60 cal A.D. 390-630
UGa-5663, wood = 50)

3. Test Unit A, UGa- 1510 +/- 60 -27.8 (2.8 x 16.4 1460 +/- 60 cal A.D. 450-670
5664,"charcoal" = 50)

4. Test Unit B, UGa- 1410 +/- 50 -27.8 (2.8 x 16.4 1360 +- 50 cal A.D. 620-770
5665, "charcoal" = 50)___

2006 VOL. 59(l)


Appendix IL Fragment count inventory by element, including disturbed burial material.

Element Total Right Left Midline or Complete Indeterminate
relatively intact crania 12 12
cranial indeterminate 160 160
frontal 37 9 6 12 10
temporal 46 13 7 12 14
parietal 44 13 12 2 17
occipital 43 5 5 19 14
zygomatic 18 7 11
nasal 5 4 1
sphenoid 41 8 9 24
maxilla 49 16 19 1 13
palatine 4 3 1
mandible 40 15 19 3 3
malleus 1 1
teeth 243 43 70 130
vertebrae 265 17 25 188 35
rib 482 151 151 180
sternum 13 13
clavicle 36 21 14 1
scapula 52 22 18 12
humerus 90 43 32 15
ulna 63 29 27 7
radius 46 23 21 2
hamate 5 4 1
triquetral 2 1 1
trapezoid 1 1
scaphoid 7 2 5
capitate 5 4 1
lunate 3 2 1
trapezium 2 2
metacarpal 1 6 3 3
metacarpal 2 2 1 1
metacarpal 3 4 2 2
metacarpal 4 6 6
metacarpal 5 3 1 2
mc indeterminate 79 34 32 13
manual phalange 110 10 10 90
os coxa 82 25 35 22
sacrum 13 1 10 2
femur 101 39 34 28
patella 11 5 5 1
tibia 95 38 32 25
fibula 51 15 15 21
calcaneus 20 10 10
talus 21 10 10 1
navicular 9 4 5
cuboid 9 5 4
1st cuneiform 7 4 3
2nd cuneiform 2 1 1
3rd cuneiform 7 3 3 1
metatarsal 1 6 3 3
metatarsal 2 5 3 2




2006 VOL. 59(1)

Element Total Right Left Midline or Complete Indeterminate
metatarsal 3 8 2 5 1
metatarsal 4 7 !4 2 1
metatarsal 5 6 '2 4
mt indeterminate 44 23 19 2
pedal phalange 26 1 1 24
shaft frags 66 66
indeterminate 65 65
minute fragments 144 144
TOTAL 2830 706 701 272 1151



Department ofAnthropology, University of Florida, 1112 Turlington Hall, P.O. Box 117305, Gainesville, FL 32611
Email: njwallis@ufl.edu

Thirty years ago, Frankie Snow (1975:57) suggested that
an important future avenue in the study of complicated
stamped pottery would be in considering the practice of
totemism among Swift Creek groups. As a prolific Woodland
period decorative style that includes abstract representations of
animals and plants, Swift Creek designs seem to be likely
candidates for totemic representation (Broyles 1968; Snow
1998). In addition, Rebecca Saunders's (1986, 1998) recogni-
tion ofintra-site circumscribed distribution of designs at Kings
Bay, Georgia (9CM171a) lends credence to the idea that Swift
Creek designs might have been crests for some form of kinship
group, perhaps clans or lineages. These qualities of represen-
tation and spatial distribution alone, however, may only hint
at the social implications of Swift Creek iconography. A
thorough understanding of the social function and meaning of
Swift Creek iconography can only be achieved through an
inductive interpretive framework grounded in context. Since
the metaphoric and metonymic meanings attached to animal
representations vary widely across ethnographic groups (Tilley
1999:49-57), their presence in Swift Creek iconography does
not necessarily denote totemism. This paper uses anthropolog-
ical theory and ethnographic analogy to assess the possibility
that Swift Creek designs represent totems and closes with
some hypotheses for representative meaning based on the
archaeological occurrence of Swift Creek iconography.

Swift Creek Designs and Distributions

Swift Creek denotes a distinctive Woodland period pottery
tradition of complicated stamped decoration. Complicated
stamping was accomplished by applying intricately carved
wood or, in rare cases, earthenware paddles to the surface of
wet clay vessels prior to drying and firing, leaving an impres-
sion of the design. The wide distribution of Swift Creek
pottery, covering all of moder-day Georgia and major
portions of all adjacent states, has been attributed to migration,
diffusion, and exchange (Goggin 1952:49; Wilson 1965), but
recent research suggests that Swift Creek pottery was produced
locally even on the utmost peripheries of its distribution
(Ashley 1992, 1998; Ashley and Wallis 2006; Wallis 2004).
The most spectacular aspect of Swift Creek pottery is its
ability to show direct connections across the landscape. Using
signature flaws of stamp designs, from cracks in the paddles
or idiosyncratic unconformity in the design, vessels found
considerable distances apart have been confirmed to have been
stamped with the same paddle (Ashley 1995; Kirkland 2003;

Snow 1975, 1998). Using petrographic or neutron activation
analyses it has been possible to determine whether the paddle
used for stamping or the vessel itself was moved across the
landscape (Smith 1998; Stoltman and Snow 1998). In the
most comprehensive petrographic study, which focused on the
Hartford site in central Georgia, Stoltman and Snow (1998)
concluded that paddle matches across sites predominantly
reflected the movement of paddles, not pots. Besides this
study, there have been only cursory and isolated attempts to
distinguish between paddle and pot movement. Both paddles
and pots were carried between sites in the lower Southeast, and
more study of the distribution of paddle matches is likely to
yield insights toward social explanations for Swift Creek
design distributions.
Bounded distribution of paddle designs within sites has
long been noted. At Kolomoki, Betty Broyles (1968:50)
observed that many designs found in Mound E were not found
anywhere else on the site. This observation does not preclude
the possibility that these designs could be found at distant
midden and village sites. For example, paddle designs from
the Dent (8DU68), Mayport (8DU96), and Browne (8DU62)
burial mounds in northeastern Florida match paddle impres-
sions found at several midden sites in southeastern Georgia
and northeastern Florida (Figure 1) (Ashley 1995; Ashley and
Wallis 2006; Wallis 2004). Likewise, Snow and Stephenson
(1998) have confirmed dozens of paddle matches between a
Hartford sub-mound context and quotidian areas across
Georgia. Thus, paddle designs may have functioned according
to a sacred and secular dichotomy in some cases (Broyles
1968), but the multiple contexts for many designs indicate
more contingent and fluid distinctions.
At the Kings Bay (9CM171a) and Mallard Creek
(9CM185) sites in Georgia, Saunders (1986, 1998) noted that
particular designs showed significant circumscribed distribu-
tions within village sites and between sites and surmised that
this distribution corresponded to a "social hierarchy" of
representation. Largely drawing on Wobst's (1977) informa-
tion theory, Saunders (1986, 1998) suggested that designs
unique to each site represented lineages, those more common
between sites stood for political affiliation or "polity," and the
most ubiquitous designs were cosmological symbols. While
Saunders (1998) is correct to highlight the importance of an
inductive approach which searches for associations of elements
within designs and for respective spatial distributions at sites,
the assignment of representative categories may be premature


VOL. 59(1)


MARCH 2006



& Florida

i River

Atlantic Ocean

ys River

rins River

0510 20 30 40
H H i---i Kilometers

Figure 1. Locations of Swift Creek sites with confirmed paddle matches mentioned in the text.

and assumes modes of social structure and symbolic represent
station that have not been demonstrated.
As pottery is always the most prevalent artifact on Swift
Creek sites, archaeologists have tended to assign undue
symbolic significance to its decoration. Williams and Elliot
(1998:10-11) and Snow (1998) argue appropriatelythatpottery
decoration in many cases may have been somewhat of an
afterthought, merely a consequence of binding clay coils in the
manufacturing process. Indeed, frequent over-stamping on
many vessels suggests that careful design execution was often
not important (Snow 1998). In other cases, precisely placed
zoned stamping indicates careful execution (Broyles 1968).
Importantly, however, it may have been the perishable wooden
paddles that carried the most symbolic meaning.

Swift Creek designs depict both seemingly abstract and
unrecognizable motifs deemed "cosmological" designs, as well
as images suggestive of plants, animals, and humans (Broyles
1968; Saunders 1998; Snow 1975, 1977, 1998). In her
analysis of designs recovered from Kolomoki and the Swift
Creek type site, Broyles (1968:52) suggested representations
of birds, animals, serpents, and flowers. In the most compre-
hensive published analysis to date, Snow (1998) speculated
that designs might represent specific animals such as rabbits,
serpents, mosquitoes, buzzards, woodpeckers, homed owls,
and roseate spoonbills, as well as mythological beings that
have characteristics of multiple animals. In addition, Snow
(1998) identifies "mask" designs that depict zoomorphic and
anthropomorphic faces that might represent a wolf, owl, bear,





2006 VOL. 59(1)


Figure 2. Possible "mask" design reconstruction from
a Mayport Mound vessel. This design has also been
identified at sites in McIntosh and Glynn Counties, GA.

and long-nosed god, among others (Figure 2). While Snow
(1998) cautiously emphasizes that these representations are
speculative, it is clear that many designs have identifiable
elements from the natural world.

Totemism: Social Reality or Theoretical Chimera?

What anthropological understandings of totemism can be
brought to bear in considerations of Swift Creek totemism? In
truth, the concept of totemism in anthropology is quite
antiquated and has been in virtual disuse for nearly fifty years,
followingthe publication ofLevi-Strauss's (1963) dismantling
treatment. Prior to this publication, however, totemism had
enjoyed considerable popularity beginning with nineteenth
century anthropologists' study of "primitive man," in whom
failure to ontologically separate himselffrom the natural world
was deemed psychologically and culturally symptomatic of this
early evolutionary stage (e.g. Frazer 1910; Maine 2002 [1866];
Morgan 1871; Tylor 1958 [1871], etc). As increasingly wider
degrees of variation in the manifestation of 'totemism' were
recorded ethnographically, Boas (1916) led the "American
tradition of skepticism" in the conclusion that totemism was
an anthropological construction, not a cultural institution
(Goldenweiser 1910; Lowie 1920; Murdock 1949). While
Durkheim (1915) and Malinowski (1948) continued to focus
on totemism as a meaningful category through functionalist

explanations, Levi-Strauss (1963) completed the critique
levied by the Boasian school (Shapiro 1991).
Through an overview of ethnology, Levi-Strauss (1963:37-
54) demonstrated that "totems," or classes taken from the
natural world, were applied to a myriad of social categories
depending on cultural context. Among Australian aboriginal
groups, for example, there are moiety, section, subsection,
clan, lineage, personal, and generational totems, as well as
sexual emblems. Moreover, the logic through which totemic
symbols are chosen is variable across cultures. While
Malinowski (1948) asserted that the prominence of animals
and plants in totemic religion was due to their importance as
food and indigenous attempts to achieve magical control over
their production and acquisition, Levi-Strauss (1963:73)
observed enough heterogeneity to confound this principle.
Among the Nuer of southern Sudan, for example, totems are
drawn from a diverse list that includes lions, trees, rivers, hide,
rope, specific parts of animals, and diseases (Levi-Strauss
1963:78). Thus, Levi-Strauss (1963) invalidated the religious
importance inherent in the concept of totemism through his
elaboration of contextual variability. For Levi-Strauss (1963),
it was the structural relationships between categories, rather
than their content, that yielded insight about cultures.
Conceptions of human-animal relationships are indeed
diverse, and cannot be explained in a concept as trite as
totemism. The various qualities of similarity and difference
between humans and animals allow possibilities for both the
contiguities of metonymy and the analogies of metaphor
(Tilley 1999:49). In other words, persons or groups can easily
be conceptualized as both embodying particular animals and
demonstrating qualities similar to animals. The process of
symbolic construction is complex and constituted through an
"interplay of tropes" that is historically and contextually
situated (Turner 1991:123). While animals can be used to
designate social groups like clans, the meanings that animals
invoke are often more complicated. For example, among
Dinka and Nuer pastoralists in East Africa, colors, personal
names, and body postures are conceptualized according to the
qualities of oxen, while in Melanesia, pigs often represent
prestige and rank and are understood as the carrier of a man's
spiritual essence (Tilley 1999:51-57). Moreover, particular
tropic elements may function on multiple symbolic levels at
the same time, and meaning may change over time since its
production is dependent on continuous social action (Turner
1991:151). The recognition of extraordinary heterogeneity
and dynamism in the meaning of animals cross-culturally
necessitates closer consideration of animal representations in
archaeological contexts.

Ethnographic Parallels: Systems of Reciprocity

Granted that the possibilities for representational meanings
with animals are various, so-called 'totemism,' in which
groups name clans and lineages according to animal, plant,
and other natural categories, does occur ethnographically.
Among historic and modern American Indians of the eastern
United States, typically matrilineal clans were delineated by




animal and plant names, some of the most common of which
were Bear, Wolf, Deer, and Beaver (Swanton 1946:655).
These clans were divided into moieties or phratries, which
were designated as "white and red" or "chiefs and warriors'"
rather than as animal names. Although represented by animal
names, historic period clans were not represented in art,
following the same principles by which a deceased person's
name was not mentioned, as a form of honor (John H. Moore,
personal communication, March 2005). This acknowledge-
ment does not invalidate the possibility that Swift Creek
paddles were totems, since representational transformations
may have occurred in the intervening millennium and because
historic American Indians' ancestral relationship to Woodland
period groups is conjectural and not well understood.
Accordingly, we must draw on other ethnographies. The
nineteenth century Tlingit of the Pacific Northwest provide an
example of an ethnographic group that depicted clan totems in
wood. Here the production of totem art was a specialized and
gender specific task, with men specializing in woodcarving
(Brotherton 2000). In addition, individual artists specialized
in particular kinds ofwoodcarvings like shamanic masks and
were likely contracted to produce them for multiple villages
(Brotherton 2000:363). The production of the more stylized
images characteristic of crests, in turn, was open to multiple
specialists within a lineage (Brotherton 2000).
Crests among the Tlingit were sacred possessions embod-
ied in multiple carvings (Kan 1989:194). Individual represen-
tations of clan totems were viewed as heirlooms, and they were
conspicuously presented in ritual events, such as funerals and
potlatches. Although designating a particular clan, however,
totems were indicative of a wider network of social relation-
ships since they could only be confirmed and legitimate
through recognition of opposing clans. An important feature
of the Potlatch, in fact, was the announcement and subsequent
repetition of each of the hosting clan names by the guests (Kan
1989:197). It was only during these ritual events in which
guests were present that new clan names and corresponding
totems could be introduced because their confirmation rested
on the outsiders' recognition.
This interdependence was emphasized in funeral rituals
where totems reached the pinnacle of their importance. In
Tlingit funerary ritual, members of the moiety opposite that of
the deceased were exclusively responsible for preparations of
the corpse and all ritual performances (Kan 1989:40-42).
These mutual obligations therefore tied clans together with
their "opposites" in a perpetually reaffirmed relationship. The
Potlatch, as the concluding mortuary rite, was conceived as
repayment to the opposites for their handling of the previous
funerary events, thus continuing the relationship of reciprocity.
Eventually, the cremated remains of the deceased individual
were housed in an elevated crypt that was designated for a
particular matriline through display of conspicuous crests on
its surface. It was the opposites' central role in the ritual
preparations associated with each death that cemented rela-
tionships between clans.
Ethnographic examples of social interdependence among
matrilineal clans are common. While Mvskokes' and other

historic southeastern American Indians left burial to nuclear
families and close kin, among northeastern groups such as the
Iroquois, mortuary practice was the responsibility of the
deceased's opposite moiety (Snow 1996:106). The Kula Ring
participants in New Guinea provide a more cogent archaeolog-
ical analogy, where abundant burial accoutrements are
garnered from opposite clans. These exchanges are primarily
a way for lineages to show regional influence and status,
thereby encouraging the endless cycle of retribution that can
only be repaid in the style of Pacific Northwest Potlatches after
a death in an affinal village (Lepowski 1989:207-208; Macin-
tyre 1989:147). Thus, among Kula partners, personhood is
constructed through the social relationships that are embodied
partly in "inalienable" possessions, a model which may have
relevance to Woodland burial contexts with exotic grave goods
(Macintyre 1989, Strathern 1988; Wallis 2005; Weiner 1992).

Conclusions: Suggestions for Future Study

Ethnographic research thus shows that many kin-based
societies with clans function according to systems of reciproc-
ity, especially in the context of mortuary rites. Since neither
clan reciprocity in burial nor artistic representation of clan
crests were practiced by historic southeastern American
Indians, Swift Creek cultures, if characterized by totemic
representation, may be more analogous to reciprocal clans in
other parts of the world. An understanding of this form of
social structure is important in suggesting models for both
paddle design meaning and the movement of paddles and
earthenware pots. A few suggestions are offered below.
First, Swift Creek paddles may be attributable to
woodworking specialists, as Saunders (1986, 1998) has
suggested at Kings Bay (9CM171a). With large enough
databases it may be possible, in some cases, to isolate the work
of individual artists or at least "analytical individuals"
(Saunders 1998). The social context of paddle production,
however, has received little attention. As with the
woodcarving traditions of the Tlingit, Swift Creek
woodcarvings may have had various social functions and
referential meanings, and may have been produced by special-
ists for particular purposes like ceremony and by non-special-
ists at other times.
Second, the distribution of paddle matches may denote
systems of reciprocity. The movement of pots and paddles has
been tentatively attributed to systems of exchange and post-
marital residence, respectively (Stoltman and Snow 1998).
While I agree that exotic vessels may represent exchange, I
argue that the movement of paddles need not imply the
movement of potters. Rather, the movement of paddles might
have been a much more complex affair that also included
mortuary exchanges. At sites deemed civic-ceremonial centers
(Smith 1992:209-13; Stephenson et al. 2002), for instance, the
distribution of paddle matches may not represent residential
mobility but instead the periodic confluence of various groups
of people carrying ritual paraphernalia and gifts, including
carved paddles. It is possible that during periodic gatherings
at ceremonial centers, paddles that were produced by


2006 VOL. 59(l)


woodcarving specialists for various lineages or clans would
have been gifted in the style of other ethnographic mortuary
exchanges (e.g. Brotherton 2000; Kan 1989). As in
ethnographic cultures with totems, new emblems likely could
not be introduced except during ceremonial gatherings, where
opposite lineages and clans could grant their legitimacy (Kan
1989:197-198). Thus, we would expect the impressions of
new paddles with new designs to be introduced in the context
of ceremonial centers, where the paddles would be distributed
and subsequently taken to various home locations across the
landscape. This trend has been observed at some mortuary
centers, as in paddle matches between the Mayport Mound
(8DU96) and Lewis Creek site, and the Browne Mound
(8DU62) and Schmidt site (8SJ52), where the sherds from
mounds have much crisper impressions from newly carved
paddles (Ashley and Wallis 2006)2. Moreover, Snow and
Stephenson's (1998) recognition of the highest sherd counts
bearing particular designs at peripheral areas rather than at
"ceremonial centers," like the Hartford site, may corroborate
this model as well.
Finally, several of the animals tentatively identified by
Snow (1998) are analogous to historic clan names, particularly
bear, rabbit, buzzard, wolf, and snake. It is possible that still
others would have been represented more abstractly on
paddles, such as earth, wind, water, or night clans. Even if not
ancestors to historic southeastern American Indians, Swift
Creek groups are likely to have shared some clan names since
they were surrounded with a largely identical natural world
from which meaning could be derived. Since the metaphorical
possibilities for animals are virtually endless, however,
distinguishing totemic symbols from other artistic renderings
will only come through continued attention to archaeological
context. I have suggested in this paper, in particular, that the
distribution of Swift Creek designs may indicate systems of
mortuary exchange and clan reciprocity. It is my hope that
future research on a regional scale will begin to test these


1 The term "Mvskoke" is a "native" spelling that is preferred in the
literature over the older "Muskogee" version (e.g. Moore 1994).

2 The Swift Creek pottery of the Browne Mound might have been
included in fill by later Colorinda phase construction, and thus may
not provide the best example. See Sears (1959) for details.


This paper was presented at the 57' annual meeting of the
Florida Anthropological Society, Gainesville, FL, and was awarded
the 2005 Student Paper Competition Prize. Thanks go to the FAS
organizing board, especially Donna Ruhl, and the competition judges,
Jeffrey Mitchem, Christine Newman, and Michael Russo. Louis
Tesar and Ryan Wheeler made helpful editorial suggestions. I also
appreciate Bill Keegan's thoughts on an earlier version of the paper.
Lastly, I thank Frankie Snow, who graciously shared the design
match information that I used in this paper. Any shortcomings of
interpretation, of course, are my own.

References Cited

Ashley, Keith H.
1992 Swift Creek Manifestations along the Lower St. Johns
River. The Florida Anthropologist 45:127-138.

1995 The Dent Mound: A Coastal Woodland Period Burial
Mound Near the Mouth of the St. Johns River, Florida.
The Florida Anthropologist 48:13-34.

1998 Swift Creek Traits in Northeastern Florida: Ceramics,
Mounds, and Middens. In A World Engraved: Archaeol-
ogy of the Swift Creek Culture, edited by Mark Williams
and Daniel Elliot, pp. 197-221. University of Alabama
Press, Tuscaloosa.

Ashley, Keith H., and Neill J. Wallis
2006 Northeastern Florida Swift Creek: Overview and Future
Research Directions. The Florida Anthropologist 59(1):5-

Boas, Franz
1916 The Origin of Totemism. American Anthropologist, New
Series 18:319-326.

Brotherton, Barbara
2000 Tlingit Human Masks as Documents of Culture Change
and Continuity. In Interpretations of Native North Ameri-
can Life: Material Contributions toEthnohistory, pp. 358-
397. University Press of Florida, Gainesville.

Broyles, Betty J.
1968 Reconstructed Designs from Swift Creek Complicated
Stamped Sherds. SoutheasternArchaeological Conference
Bulletin 8:49-73. Morgantown, West Virginia.

Durkheim, Emile
1915 TheElementaryForms oftheReligious Life. Translated by
J.W. Swain. Alien and Unwin, London.

Frazer, James George
1910 Totemism and Exogamy: a Treatise on Certain Early
Forms ofReligion and Society. Macmillan, London.

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

Goldenweiser, A.
1910 Totemism, an Analytical Study. Journal ofAmerican
Folklore 23:179-293.

Kan, Sergei
1989 Symbolic Immortality: The Tlingit Potlatch of the Nine-
teenth Century. Smithsonian Institution Press, Washington,

Kirkland, S. Dwight
2003 Human Prehistory at the Sadlers Landing Site, Camden
County, Georgia. Early Georgia 31:107-192.

Lepowski, Maria




1989 Death and Exchange: Mortuary Ritual on Vanatinai (Sudest
Island). In Death Rituals and Life in the Societies of the
Kula Ring, edited by F. Damon and R. Wagner, pp. 199-
129. Northern Illinois University Press, DeKalb.

Levi-Strauss, Claude
1963 Totemism. Translated by R. Needham. Beacon Press,

Lowie, Robert H.
1920 Primitive Society. Liveright, New York.

Maine, Henry S.
2002 [1866] AncientLaw. Transaction Publishers, New Bruns-

Malinowski, Bronislaw
1948 Magic, Science, and Religion. Beacon Press, Boston.

Macintyre, Martha
1988 The Triumph of the Susu: Mortuary Exchanges on
Tubetube. In Death Rituals and Life in the Societies of the
Kula Ring, edited by F. Damon and R. Wagner, pp. 133-
152. Northern Illinois University Press, DeKalb.

Moore, John H.
1994 Ethnoarchaeology ofthe LamarPeoples. In Perspectives on
the Southeast: Linguistics, Archaeology, andEthnohistory,
edited by Patricia Kwachka, pp. 126-142. University of
Georgia Press, Athens.

Morgan, Lewis H.
1871 Ancient Society. World Publishing, New York.

Murdock, George P.
1949 Social Structure. Macmillan, New York.

Saunders, Rebecca
1986 Attribute Variability in Late Swift Creek Phase Ceramics
from King's Bay, Georgia. Unpublished Master's thesis,
Department of Anthropology, University of Florida,

1998 Swift Creek Phase Design Assemblages from Two Sites on
the Georgia Coast. In A World Engraved: Archaeology of
the Swift Creek Culture, edited by M. Williams and D. T.
Elliott, pp. 154-189. University of Alabama Press,

Sears, William H.
1959 Two Weeden Island Period Burial Mounds, Florida.
Contributions of the Florida State Museum 5, Gainesville.

Shapiro, Warren
1991 Claude Levi-Strauss Meets Alexander Goldenweiser:
Boasian Anthropology and the Study of Totemism. Ameri-
can Anthropologist 93:599-610.

Smith, Betty A.
1998 Neutron Activation Analysis of Ceramics from Mandeville
and Swift Creek. In A World Engraved: Archaeology of
the Swift Creek Culture, edited by M. Williams and D.T.
Elliot, pp. 112-129. University of Alabama Press,


Smith, Bruce D.
1992 Hopewellian Farmers of Eastern North America. InRivers
of Change: Essays on Early Agriculture in Eastern North
America, by Bruce D. Smith., pp. 201-48. Smithsonian
Institution Press, Washington, D.C.

Snow, Frankie
1975 Swift Creek Designs and Distributions: A South Georgia
Study. Early Georgia 3:38-59.

1977 An Archaeological Survey of the Ocmulgee Big Bend
Region: A Preliminary Report. Occasional Papers from
South Georgia, Number 3. South Georgia College,

1998 Swift Creek Design Investigations: The Hartford Case. In
A World Engraved: Archaeology of the Swift Creek
Culture, edited by M. Williams and D. T. Elliott, pp. 61-
98. University of Alabama Press, Tuscaloosa.

Snow, Frankie, and Keith Stephenson
1998 Swift Creek Designs: A Tool for Monitoring Interaction. In
A World Engraved: Archaeology of the Swift Creek
Culture, edited by M. Williams and D. T. Elliott, pp. 99-
111. University of Alabama Press, Tuscaloosa.

Stephenson, Keith, Judith A. Bense, and Frankie Snow
2002 Aspects ofDeptford and Swift Creek on the South Atlantic
and Gulf Coastal Plains. In The Woodland Southeast,
edited by D. G. Anderson and R. C. Mainfort, Jr., pp. 318-
351. University of Alabama Press, Tuscaloosa.

Stoltman, James B., and Frankie Snow
1998 Cultural Interaction within Swift Creek Society: People,
Pots, and Paddles. In A World Engraved: Archaeology of
the Swift Creek Culture, edited by M. Williams and D. T.
Elliott, pp. 130-153. University of Alabama Press,

Strathern, Marilyn
1988 The Gender of the Gift: Problems with Women and Prob-
lems with Society in Melanesia. University of California
Press, Berkeley.

Swanton, John R.
1946 The Indians of the Southeastern United States. Bureau of
American Ethnology, Bulletin 137. Washington, D.C.

Tilley, Christopher
1999 Metaphor andMaterial Culture. Blackwell, Oxford.

Turner, Terrance
1991 "We are Parrots, Twins are Birds": Play of Tropes as
Operational Structure. In Beyond Metaphor, edited by J.
Fernandez, pp. 121-158. Stanford University Press,

Tylor, Edward Burnett
1958 [ 1871 ] Primitive Culture: Researches into the Development
of Mythology, Philosophy, Religion, Language, Art and
Custom. Harper, New York.


2006 VOL. 59(l)


Wallis, Neill J.
2004 Perpetuating Tradition on the Lower St. Johns River:
Pottery Technology and Function at the Mayport Mound,
8DU96. The Florida Anthropologist 57:271-298.

Weiner, Annette
1992 Inalienable Possessions. University of California Press,

Williams, J. Mark and Daniel T. Elliott
1998 Swift Creek Research: History and Observations. InA
World Engraved: Archaeology of the Swift Creek Culture,
edited by J. M. Williams andD.T. Elliot, pp. 1-11. Univer-
sity of Alabama Press, Tuscaloosa.

Wilson, Rex L.
1965 Excavations attheMayportMound, Florida. Contributions
of the Florida State Museum 13, Gainesville.

Wobst, H. Martin
1977 Stylistic Behavior and Information Exchange. In For the
Director: Research Essays in Honor of James B. Griffin,
edited by C.E. Cleland, pp. 317-342. Anthropological
Papers No. 61, Museum of Anthropology, University of
Michigan, Ann Arbor.



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.


The Great Journey: The Peopling ofAncientAmerica (Revised
edition). Brian M. Fagan. 2004. University Press of Florida,
Gainesville. 288 pages, illustrated, annotatedbibliography and
indexed. $24.95 (paperback). First published in 1987 and first
revised in 1989.

Florida Division of Historical Resources, 500 S. Bronough
Street, Tallahassee, FL 32399-0250

Fifteen years have passed since the latest revision of The
Great Journey. One would have thought we might have gained
much knowledge of early prehistory during that time span.
Mostly, though, we still hear the same arguments that we
heard 20 to 30 years ago. On the other hand, Fagan has
benefitted from having many more sites and studies to exam-
ine for his latest version of this book. For, as he aptly puts it,
"archaeology has been compared to an unperformed play."
In any event, this work entailed combing through countless
tomes to learn where, when and how the original inhabitants
of the Americas arrived. It also aimed to discover just who
these people were. Realistically, short of someone entering a
wormhole or a parallel universe and then coming back and
reporting on his findings, we most likely will never learn the
answers to all the questions that have stoked our passions for
knowing more about the peopling of the Americas.
Fagan begins this epic journey by examining many theories
of the origins of people into the Americas. These hypotheses
vary from absurd to insightful and from fanciful to factual.
Nowadays, few believe that Phoenicians, Tartars,
Carthaginians, the Lost Tribe of Israel, mound builders or the
Minoans from Atlantis first populated the Americas. Fagan
offers that people came to North America at least 15,000 years
ago. Others argue for much earlier (as much as 200,000 years
ago), to moderately earlier (40,000 years ago), dates when
people set foot in the New World. Most probably, humans
came from Siberia to pursue game. They likely crossed a land
bridge that connected Asia with North America near the end
of the Pleistocene epoch. His synthesis indicates that these
people hunted and gathered rodents, birds, fish, mollusks and
plant foods, as well as larger Ice Aged mammals. By the time
Europeans "discovered" the Americas, thousands, if not
millions, of inhabitants made up 2,000 or more distinctive
societies that made use of every corner of the Western Hemi-
Fagan reminds us, though, that the great antiquity of
humanity did not gain wide acceptance until sometime in the
1840s. This was when Boucher de Perthes discovered extinct
animals in France in association with stone tools made by
humans. In the New World, Folsom, and later Clovis, points

found with extinct bison and mammoth remains, revealed to
us that people had been in North America much longer than
conventional wisdom had led us to believe.
He goes on to examine the origin and evolution of primates,
which began some 80 million years ago. Fagan informs us
when scientists speculate that higher primates branched into
hominids. He then revisits when and how these gifted mam-
mals advanced and dispersed throughout the world. The
development of tools and culture soon allowed them to
dominate the planet. It took fully modern humans, though, to
conquer earth's last, vast habitable landmasses of Australia
and the Americas. These early voyagers probably brought
fairly sophisticated Upper Paleolithic technologies with them
from the Old World. Fagan uses not only archaeological data,
but also findings from linguistics and physical anthropology
dentitionn and blood proteins) to support his case. Most
evidence suggests several waves of new immigrants. The first
being the Paleo-Indians, followed by the Athapascans (or Na
Dene) and finally the Aleuts and Eskimos. These people
walked, or some believe paddled along the northwest coast of
North America, across the Bering Strait. Fagan believes this
to be the only logical route for early humans to have reached
the Americas. While this would have been possible between
35,000 to 15,000 years ago, he interprets the best available
evidence to indicate that his occurred sometime after 15,000
B.P. He bases much of this on the Bluefish Cave site in
eastern Beringia and the opening of the Ice Free corridor in
western North America at that time.
In a chapter titled "On the Track of Earliest Settlers," he
reviews and delves into many of the presumed earliest archae-
ological sites, such as Meadowcroft Rock shelter (Pennsylva-
nia), Pikimachay Cave (Peru) and Pedra Furada (Brazil),
Monte Verde (Chile) and many others. He suggests that Monte
Verde in Chile represents the early site probably best able to
withstand the rigors of meticulous scientific scrutiny. One
most intriguing omission in this section, though, is of
Kennewick man from the Pacific Northwest. Some believe this
early archaeological find to be the most significant ever in
North America. If nothing else, Kennewick certainly has
managed to mesmerize the media. Whatever, this find lends
support that some relatively early peoples in North America
may have arrived via a different route than Siberia and from
some place other than eastern Asia. Fagan also skirts mention
of the controversial Cactus Hill site in Virginia. Cactus Hill
allegedly bridges the gap between the Solutrean peoples of
France and the Clovis makers of North America.
While his book makes scant mention of Florida, a segment
touches on Warm Mineral Springs and Little Salt Spring, as
"sinkholes in limestone country of southeast (sic) Florida."
Fagan makes a highly salient point: "There are hundreds of


VOL. 59(1)

MARCH 2006


Florida sinkholes that still await archaeological exploration,
many of them known to contain artifacts and animal bones.
Such sites offer some of the best opportunities we may ever
have for recovering evidence ofpre-Clovis occupation in North
America at the very end of the Ice Age." For this to occur,
though, underwater archaeologists in Florida must quell their
fears of probing drowned caves. Fagan concludes this chapter
by surmising that humans most likely occupied the extreme
northern portions of North America by 15,000 years ago and
had spread to near the tip of South America by 14,000 to
12,000 years ago.
His review of Clovis people focuses on the Great Plains,
where Paleo-Indians hunted large megafauna such as mam-
moths, mastodons and extinct bison beginning about 11,500
years ago. They also hunted tapir, camel, bear and rabbit. In
this section, he correctly deducts that larger animals were the
ones most often preserved. Most everyone now agrees, though,
that early peoples made good use of small animals as well as
collected plants for food. Kills of large animals, though, such
as mammoth provided them with meat, hooves, tusks, bones,
pelts and firs, which they used for making clothing, tools,
shelter as well as food. Although some experts seem to believe
the Clovis makers were the first New World inhabitants,
Fagan perceptively notes that Clovis points have not been
found in Alaska or the Yukon. He suggests, and is probably
correct, that Clovis culture most likely arose in the Great
Plains. Conversely, he also dismisses the notion of any pre-
projectile people, now in vogue again by some, being Amer-
ica's earliest arrivals. Unfortunately, he covered few Paleo-
Indian sites in eastern North America that also held extinct
faunal remains. In any event, by 10,500 years ago, well-
defined regional point/blade forms became evident throughout
North America. This indicates that distinctive cultural
adaptations occurred much earlier than many investigators had
previously thought.
His chapter on "The Bison Hunters" also takes a
Plains/Southwestern slant. He reveals that bison flourished
while other large animals came under mounting stress. So, in
that region of North America, bison became the dominant
species in archaeological sites such as Folsom (New Mexico),
Scottsbluff (Nebraska) and Hell Gap (Wyoming). Again,
Fagan notes many uses, beyond food, of bison such as for
clothing, containers and lodge coverings. Also the natives
made pemmican (a mixture of dried meat and fat) from buffalo
and other mammals. This portable food allowed these people
to become much more mobile in their search for resources.
Unfortunately, after Europeans introduced guns and horses to
the Plains Indians, bison nearly became extinct.
Fagan goes on to discuss aboriginal peoples in the far north
of North America-in the harshest of ecological niches. He
concludes the book with how native peoples became more and
more diverse and developed increasingly sophisticated
technologies and life ways through time.
While Fagan's Great Journey suffers a few minor flaws, it
provides us with a readable yet comprehensive account of the
peopling of the Americas. The tale of how eastern Asians
crossed the Bering land bridge began with the dispersal of

modern humans from tropical Africa some 150,000 years ago.
The book constructs several scenarios for settlement of the
New World and selects from the most plausible ones-based
on scientific evidence. Fagan examines the contents of many
early sites and the ways they were excavated. Some of the most
heated debates in archaeology today still rage from the
controversial interpretations of many of these sites. The book
also touches upon what role far-ranging humans may have
played in the extinction of many large animals in the Americas
near the end of the Ice Age.
All in all, Brian Fagan should be commended for revising
this magnum opus. Readers from fledgling adolescents to
seasoned scholars will benefit from studying about the peo-
pling of the Americas. Much still remains to be learned,
nonetheless, about how and when these people arrived and
how they slowly transformed the American landscape over the
course of thousands of years.

A Concise Natural History of East and West Florida, by
Bernard Romans. Edited and with an introduction by Kathryn
E. Holland Braund. 1999. 442 pages, maps, illustrations.
University of Alabama Press, Tuscaloosa. $47.50 (cloth).

2111 Delancey St., Philadelphia, PA 19103

"Bernard Romans'sA Concise Natural History ofEast and
West Florida, William Bartram's Travels and James Adair's
History of the American Indians comprise the three most
significant accounts of the Deep South published during the
later eighteenth century," writes Braund, who has edited or
written about all three. And if you haven't yet explored
Romans's Natural History, Braund's new edition is afine way
to enter his world of the 1760s and 1770s.
Romans was blustery, opinionated, sometimes unfair and
insulting, sometimes wrong-but never in doubt. His great
saving grace was that he was curious about the details and
texture of his world and the people who lived in it. Here is a
typical example.
One day Romans attended an afternoon dance at a Creek
Indian village and saw that nine women were wearing on their
legs dozens of "bells" made of deer hooves. The bells made a
sound like castanets. Intrigued, Romans examined one
costume and found 493 of these bells. He calculated that 1,110
deer must have been killed to outfit the nine Creek la-
dies-"An instance of luxury in dress scarcely to be paralleled
by our European ladies," wrote Romans. Of course, the hooves
may have been a by-product of the flourishing Creek business
of hunting deer for their hides to sell to traders. But neverthe-
less, 493 bells on a pair of leggings sounds like sounds like a
lot of work, which could be one definition of luxury.
And if some Creek women were fashion plates, some Creek
men were dandies. "The men are also very fond of dress; my
guide across the Peninsula, employed about two hours at his
toilet, at Mr. Moultrie's house, four miles from St. Augustine,
before he would venture to shew himself in town."
Romans Natural History is a sprawling grab bag of a book.


2005 VOL. 59(1)


It is mainly a promotional tract to encourage investment in the
region, and a manual for would-be settlers, filled with advice
and cost estimates and much information about soil, crops,
weather, cuisine, diseases, the region's aboriginal and Euro-
pean peoples, its flora and fauna, recent history, and aids to
navigation, not to mention advice on camping out when
traveling through the Florida woods. Here are samples of what
Romans noticed:

* An Indian man "discharges his urine in a sitting
posture," Indian women perform the same chore
standing up.

White Floridians typically wore in summer "a slight
waistcoat of striped cotton, and a pair of trowsers of
the same, and often no coat; if any, it is a short one
made of light stuff."

"Blacks from Guinea" introduced the peanut to
America, and also sesame.

St. Augustine is "very ill built, the streets being all
except one crooked and narrow. The date on one of
the houses I remember to be 1571; these are of stone,
mostly flat roofed, heavy, and look badly," he wrote
as part of an overview of Florida's coastal areas.

Many people attracted Romans ire: among them, the
naturalist John Bartram; Dr. Andrew Turbull, proprietor of
the troubled settlement at New Smyrna; Romans's boss, the
surveyor William De Brahm.
Romans was a transplanted Dutchman. The details about the
first part and last part of his life are sketchy. He was born,
perhaps in Amsterdam, in 1720, and lived in England as a
young man. He had training in surveying and navigation, and
immigrated to the Americas around 1757 during the Seven
Years War. He seems to have worked as a merchant seaman
and as a privateer on voyages that took him from Panama to
Labrador, writes Braund.
September 1766 found Romans hauling mahogany from the
Isle of Pines south of Cuba to Georgia, where he found a
government job as deputy surveyor of Georgia, which quickly
led to a bigger surveyingjob as "principal deputy surveyor" for
the Southern District. The new post meant travels around
modern Florida and portions of modern Alabama, Louisiana,
andMississippi, and quarrels with De Brahm, another difficult
Dutchman. Thejob also provided Romans ample opportunities
to collect the information he would use later to write his
Natural History, including sailing directions and navigation
charts of the coast of East and West Florida.
Romans headed north in 1773, ending up in New York
looking for investors in his Natural History, and working on
other projects to bring in cash. Along the way his book, as
Romans noted, "swelled imperceptibly to about 800 pages."
Ten years in the making, his Natural History finally emerged
in just about the worst possible market for a book about
investing in an American colony-April, 1775. That was the

same month as the battles of Lexington and Concord that
marked the beginning of the War of the American Revolution.
A patriot for his adopted country, Romans joined the rebellion.
Romans, no stranger to quarrels, embroiled himself in contro-
versy over his wrongheaded plans for fortifications to control
river traffic on the Hudson. Captured by the British, he died on
his way back to America after the war. He may have been
The Natural History is also a bully pulpit for Romans to
share his views on topics as varied as the origins of the first
peoples of America and the pros and cons of trade monopolies.
Romans was a man of his times and his views mirror much of
the conventional wisdom of his era about the rightness of
slavery, the intractable nature of the Indians (a.k.a. savages)
and so on and so on.
"If the modern reader understands the biases and cultural
ignorance under which Romans and other European men
labored, their facts-if not their interpretations-can prove
helpful in understanding eighteenth-century Indian societies,"
writes Braund.
Romans traveled among the Creek, Chickasaw, Choctaw and
other native peoples. And here as elsewhere Romans's eye for
detail comes through for his twenty-first century readers. His
attention was caught by Chickasaw hand-held deer decoys,
Choctaw child-rearing practices (their children "are never
beaten or otherwise rudely chastised"), and Creek culinary
preferences (beef, rice, and, in season, "melons, peaches,
plumbs, grapes, or some other wild fruit"). Indian warfare,
sexual practices, housing styles, drinking habits are all grist
for Romans's mill.
But Romans also got things wrong. For example, writes
Braund, Romans did not understand matrilineal kinship and
gender-related work roles in the relations between aboriginal
men and women. Fortunately Braund is there to help when
Romans misses the point of what he is seeing. Her crisply
written essays and notes are there to guide the general reader
and reassure the specialist.
This handsomely produced volume includes a biography of
Romans; a critique of his book as a source of information on
the eighteenth century south; the text of his Natural History;
and satisfyingly comprehensive notes to annotate his text, and
illustrations. The original 1775 edition has been re-set,
eliminating pointless distractions like upside down letters and
other typos, as well as the eighteenth century long S which
turns words like "west" into "weft."
Braund's edition does not include Romans's map sheets that
are available in out- of-print facsimile editions-if you can
find them in a complete condition in the second hand market.
On the other hand, the principal value for us of the Natural
History is its written information, not its charts. And her
publisher did keep the cost to under $50.


Anna Madgigine Jai Kingsley: African Princess, Florida
Slave, Plantation Slaveowner. Daniel L. Schafer. 2003.
University Press of Florida, Gainesville. 178 pages, illustra-
tions, maps. $24.95 (hardcover).

2111 Delancey St., Philadelphia, PA 19103

The bare outlines of her story would probably be dismissed
as too unlikely to even merit a Hollywood studio pitch: a black
13-year old girl is kidnapped, shipped across the Atlantic as a
slave, then sold to a middle-aged white Florida businessman
30 years her senior. He marries her. Five years later frees her
and their three children. She continues to help him run his
businesses, buys her own slaves and gets rich, becomes poor,
and lives to a ripe old age.
If it were not for Daniel L. Schafer the odyssey that was the
life of Anna Madgigine Jai Kingsley would only be known in
its barest outline and that outline would be encrusted with
make-believe: Anna as daughter of an African king given as
a gift to a slave trader and so on. Schafer has done something
marvelous: recovered an almost lost life and illuminated a lost
It was the make-believe stuff about Anna Madgigine that
Schafer heard back in 1975 from an enthusiastic but somewhat
overwrought volunteer guides at Kingsley Plantation near
Jacksonville that started Schafer on his quest to disinter the
facts about Anna, who left no diary, letters, or even a picture.
The quest over the years has taken Schafer from his base at
the University of North Florida to Europe, Africa, and the
Caribbean, not to mention through old tax rolls, probate
inventories, pension applications and the like. Meanwhile
Schafer continued with his researches into eighteenth and
nineteenth century Florida.
Schafer previously has written short accounts of Anna's life.
The latest iteration of her story is a full work-up by Schafer of
her life, with a useful and welcome emphasis on explaining the
broader social context of her time and place, whether it was
Africa in the 1790s or Florida in the 1840s.
Anna Kingsley was a Wolof. While Europe was convulsed
first with the wars of the French Revolution and then by the
wars of Napoleon, the four Wolof-speaking states of the
Senegambia region of West Africa were also going through
their own wars. The principal Walof state was Kajoor, and its
ruler, Amari Ngoone Ndella, relied on professional cavalry,
known as tyeddo, to extend his authority, to repress dissent
fuelled by religious piety, and to raid for slaves he could sell to
pay for his wars. The tyeddo, in an echo of the Ottoman
Turkish empire'sjanissaries, were themselves slaves.
The future Anna Kingsley was born into a family that
counted Wolof rulers in the family tree. It is likely she was
taken prisoner during a tyeddo raid in 1806 into the Wolof
state of Jolof. It is also possible that Anna was sold into
slavery by her family. Schafer has interviewed oral historians
in West Africa, and he navigates carefully and sympathetically
through their evidence
Based on the records he has found, Schafer believes the

barely teenage Anna was probably transported to Havana on
the Sally, a vessel registered in Danish St. Thomas. She was
certainly bought by Zephaniah Kingsley in late September or
early October 1806. She most likely arrived in Orange Park, a
suburb of Jacksonville, on Kingsley's vessel the Esther, along
with four hogsheads of molasses, 28 half-pipes and 12 whole
pipes of rum, and two other enslaved African women. We
know because all were listed on his customs declaration form.
Today part ofKingsley's Laurel Grove plantation in Orange
Park is covered by the Continental Club on Astor Street, while
the rest is now under nearby subdivisions. It is hard to imagine
what the plantation was like in 1806, but Schafer expertly
outlines how Laurel Grove functioned as a self-sufficient
community of mostly Ibu, Susu, and Calabari peoples, with
names like Bonafy, Qualla, Abdalla, Comba, Coonta, and
McGuindo. They spoke Kamba, Calabari, Susu, Baga, Ibo--
and Wolof. They worked in the cotton fields, and as carpenters
and blacksmiths, and in other crafts. Kingsley also crewed his
trading vessels with enslaved sailors.
Kingsley was a singular and contradictory character who
could hijack any story, although Schafer does not let him.
Kingsley had views on how to improve slavery as an institu-
tion and advocated both abolition and polygamy
After being manumitted by Kingsley when she was 18 Anna
set up her own establishment across the St. Johns River in
modern suburban Mandarin, possibly as part of Kingsley's
mercantile operations. Years later Kingsley was asked what
attracted him to her. "She was a fine tall figure, black as jet
but very handsome. She was very capable, and could carry on
the affairs of the plantation in my absence as well as I could
myself. She was affectionate and faithful, and I could trust
her," said Kingsley.
One day, two years after she was freed, she demonstrated
how capable she could be. She burned down both Laurel
Grove's main house and her own new home. The first was
destroyed to drive out marauders who were a threat to her. The
second to make sure other raiders could not make use of it.
The twin events happened during the waning days of the
Patriot Rebellion of 1812, an attempt to annex Spanish Florida
to the United States. The venture also attracted looters and
slave hunters from across the border. Anna and her three
children, and 12 slaves --seven of them children- were in
danger from the slave hunters. Kingsley was away.
One day in November 1813 two Spanish gunboats arrived
off Laurel Grove to pummel 70 looters who were using the
place as a base for raids. Anna made contact with one gunboat
captain, and made plans to fire the two homes to get rid of the
plantation's unwelcome guests. By the end of the episode the
initially suspicious Spanish captain was calling her a "hero-
ine" in his official report.
Kingsley moved to Fort George Island in March 1814 and
Anna continued as his principal wife there for 24 years, until
1838. The residence she built near the main house is still
there, as are the tabby walls of the cabins the Kingsleys built
for their slaves.
The transfer of Florida to the United States in 1821 brought
increasingly harsh legal conditions for free and enslaved


2005 VOL. 59(1)


blacks, to the point that re-enslavement of Kingsley's several
families was a real risk. In the 1830s Zephaniah Kingsley
decided to move those dear to him to what is today part of the
Dominican Republic but in those days was part of Haiti.
Schafer shows us the 70-year old riding across Haiti to create
a 36,000-acre refuge for his families and children, some of
whom by now had married white spouses.
Conditions in Haiti deteriorated and Anna Kingsley returned
to Florida after Kingsley's death in 1843 in time to go to legal
war to successfully protect her rights in her husband's will.
Schafer briskly navigates us through the provisions of the
treaty which transferred Florida to the U.S. and protected
black people who had been free in Spanish Florida. Anna later
returned to the courthouse with further petitions to protect her
substantial assets, which included bond people. The Civil War
was a financial disaster for Anna, who probably spent the war
in New York and then returned to Jacksonville. She died in
This remarkable human being is buried in an unmarked
grave in Clifton Cemetery, which Schafer visits.

About the Authors:

Keith H. Ashley is an Assistant Professor of Liberal Arts (Anthropology) at Savannah College of Art and Design. His current
research centers on the native history of northeastern Florida, with an emphasis on political economy and historical process.

Simon Barker-Benfield is a writer with an interest in Florida history.

Melissa Gold received her Master's degree in Biological Anthropology from the University of Florida in April 2005. She
is now residing in her hometown of Lisle, Illinois while working as an adjunct instructor and tutor.

David Thulman is a doctoral candidate in Anthropology at Florida State University, where he is working on Paleoindian
social structures in Florida. He is also an attorney for the Florida Department of Environmental Protection.

Neill J. Wallis is pursuing a Ph.D. at the University of Florida. Using the Woodland period Atlantic coast as a case study,
his research centers on the social role of exchange in small-scale societies in the American Southeast. He also is interested
in the social implications of pottery production and exchange, particularly in the Woodland period Southeast.

Mike Wisenbaker is employed by the Florida Division of Historical Resource. He has been involved in cultural resource
management, public archaeology and historic preservation for about 28 years.


2006 VOL. 59(1)



Editor's Page

Northeastern Florida Swift Creek:
Overview and Future Research Directions. Keith H. Ashley and Neill J. Wallis

A Suwannee/Bolen Artifact Assemblage from the Santa Fe River. David K. Thulman

Osteological Analysis of the Manasota Period Dunwody Site. Melissa L. Gold

The Case for Swift Creek Paddles as Totemic Symbols:
Some Anthropological Considerations. Neill J. Wallis


Fagan: The Great Journey: The Peopling ofAncient America. Michael Wisenbaker

Romans: A Concise Natural History of East and West Florida. Simon Barker-Benfield

Schafer: Anna Madgigine Jai Kingsley:
African Princess, Florida Slave, Plantation Slaveowner. Simon Barker-Benfield

Copyright 2006 by the
ISSN 0015-3893

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