Table of Contents
 Editor's Page
 Archaeological Sites Map
 Archaeological Investigations at...
 The Wakulla Springs Lodge Site...
 Chert Distribution and Exploitation...
 A New Look at the Mount Taylor...
 The Over0Exploitation of Hard Clams...
 Archaeological Investigations at...
 Archaeobotany at Bernath Place...
 Prehistoric and Protohistoric Fort...
 Recent Investigations at the O'Connell...
 About the Authors

Group Title: Florida anthropologist
Title: The Florida anthropologist
Full Citation
Permanent Link: http://ufdc.ufl.edu/UF00027829/00190
 Material Information
Title: The Florida anthropologist
Abbreviated Title: Fla. anthropol.
Physical Description: v. : ill. ; 24 cm.
Language: English
Creator: Florida Anthropological Society
Conference: 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: VID00190
Source Institution: University of Florida
Holding Location: Department of Special Collections and Area Studies, George A. Smathers Libraries, 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
    Table of Contents
        Table of Contents 1
        Table of Contents 2
    Editor's Page
        Unnumbered ( 5 )
        Page 80
    Archaeological Sites Map
        Page 81
    Archaeological Investigations at the Colorado Site (8HE241) A Lithic Workshop in Hernando County, Florida
        Page 82
        Page 83
        Page 84
        Page 85
        Page 86
        Page 87
        Page 88
        Page 89
        Page 90
        Page 91
        Page 92
        Page 93
        Page 94
        Page 95
        Page 96
        Page 97
    The Wakulla Springs Lodge Site (8WA329): A Preliminary Report on a Stratified Paleoindian Through Archaic Site, Wakulla County, Florida
        Page 98
        Page 99
        Page 100
        Page 101
        Page 102
        Page 103
        Page 104
        Page 105
        Page 106
        Page 107
        Page 108
        Page 109
        Page 110
        Page 111
        Page 112
        Page 113
        Page 114
        Page 115
    Chert Distribution and Exploitation in Peninsular Florida
        Page 116
        Page 117
        Page 118
        Page 119
        Page 120
        Page 121
        Page 122
        Page 123
        Page 124
        Page 125
        Page 126
        Page 127
        Page 128
        Page 129
        Page 130
        Page 131
    A New Look at the Mount Taylor and Bluffton Sites, Volusia County, with an Outlinne of the Mount Taylor Culture
        Page 132
        Page 133
        Page 134
        Page 135
        Page 136
        Page 137
        Page 138
        Page 139
        Page 140
        Page 141
        Page 142
        Page 143
        Page 144
        Page 145
        Page 146
        Page 147
        Page 148
        Page 149
        Page 150
        Page 151
        Page 152
        Page 153
        Page 154
        Page 155
        Page 156
        Page 157
    The Over0Exploitation of Hard Clams (Mercenaria spp.) from Five Archaeological Sites in the Southeastern United States
        Page 158
        Page 159
        Page 160
        Page 161
        Page 162
        Page 163
        Page 164
        Page 165
        Page 166
        Page 167
    Archaeological Investigations at DeSoto National Memorial: Perspectives on the Site Formation and Cultural History of the Shaw's Point Site (8ma7), Manatee County, Florida
        Page 168
        Page 169
        Page 170
        Page 171
        Page 172
        Page 173
        Page 174
        Page 175
        Page 176
        Page 177
        Page 178
        Page 179
        Page 180
        Page 181
        Page 182
        Page 183
        Page 184
        Page 185
        Page 186
        Page 187
        Page 188
        Page 189
    Archaeobotany at Bernath Place (8SR986) and Other Santa Rosa/Swift Creek-related Sites in Coastal and Non-Coastal Southeastern U.S. Locations
        Page 190
        Page 191
        Page 192
        Page 193
        Page 194
        Page 195
        Page 196
        Page 197
        Page 198
        Page 199
        Page 200
        Page 201
        Page 202
        Page 203
    Prehistoric and Protohistoric Fort Walton at the Thick greenbriar Site (8ja417), Northwest Florida
        Page 204
        Page 205
        Page 206
        Page 207
        Page 208
        Page 209
        Page 210
        Page 211
        Page 212
        Page 213
        Page 214
        Page 215
        Page 216
        Page 217
        Page 218
        Page 219
        Page 220
        Page 221
        Page 222
        Page 223
    Recent Investigations at the O'Connell Mission Site (8le157), Leon County, Florida
        Page 224
        Page 225
        Page 226
        Page 227
        Page 228
        Page 229
        Page 230
        Page 231
        Page 232
        Page 233
        Page 234
        Page 235
        Page 236
        Page 237
        Page 238
        Page 239
        Page 240
        Page 241
        Page 242
        Page 243
        Page 244
        Page 245
        Page 246
        Page 247
        Page 248
        Page 249
    About the Authors
        Page 250
        Page 251
Full Text


2000 Florida Anthropological Society Inc.

The Florida Anthropological Society Inc. holds
source text of the Florida Anthropologist
considered the copyright holder for the text
these publications.

all rights to the
and shall be
and images of

The Florida Anthropological Society has made this publication
available to the University of Florida, for purposes of
digitization and Internet distribution.

The Florida Anthropological Society reserves all rights to this
publication. All uses, excluding those made under "fair use"
provisions of U.S. Code, Title 17, Section 107 are restricted.

Contact the Florida Anthropological Society for additional
information and permissions.

"W I

R" 4 l":

W, 0 IS1111111"'


ii, 9' "il"Y'
i j i'l i 11 1




t I,'
'I . A ll



"I', %




..... ..... .....



4' i- 1 4

4 t



Volume 53 Numbers 2-3
June-September 2000


Editor's Page. Ryan J. Wheeler 78


Foreword. Donna L. Ruhl and Christine L. Newman


Archaeological Investigations at the Colorado Site (8HE241) A Lithic Workshop in
Hernando County, Florida. Elizabeth A. Horvath

The Wakulla Springs Lodge Site (8WA329): A Preliminary Report on a Stratified
Paleoindian Through Archaic Site, Wakulla County, Florida. B. Calvin Jones and
Louis D. Tesar

Chert Distribution and Exploitation in Peninsular Florida. Robert J. Austin and
Richard W. Estabrook

A New Look at the Mount Taylor and Bluffton Sites, Volusia County, with an
Outline of the Mount Taylor Culture. Ryan J. Wheeler, Christine L. Newman, and
Ray M. McGee

The Over-Exploitation of Hard Clams (Mercenaria spp.) from Five Archaeological
Sites in the Southeastern United States. Irvy R. Quitmyer and Douglas S. Jones

Archaeological Investigations at DeSoto National Memorial: Perspectives on
the Site Formation and Cultural History of the Shaw's Point Site (8MA7) Manatee
County, Florida. Margo Schwadron

Archaeobotany of Bernath Place (8SR986) and other Santa Rosa/Swift Creek-Related
Sites in Coastal and Non-Coastal Southeastern U. S. Locations. Donna L. Ruhl

Prehistoric and Protohistoric Fort Walton at the Thick Greenbriar Site (8JA417),
Northwest Florida. Nancy Marie White

Recent Investigations at the O'Connell Mission Site (8LE157), Leon County, Florida.
Rochelle A. Marrinan, James A. Halpern, Gregory M. Heide, and Chelsea Blackmore

About the Authors

Cover: The cover depicts artifacts and ecofacts related to the nine articles presented: 1) Mount Taylor shell
bannerstone (Wheeler, Newman, and McGee); 2) Suwannee/Simpson-like bifacial preform (Jones and Tesar); 3)
Columella hammer (Schwadron); 4) Annual growth cycle of Mercenaria clam (Quitmyer and Jones); 5) Hickory (Ruhl);
6) Silver finger ring (Marrinan, Halpern, Heide, and Blackmore); 7) Diagnostic fossil in chert (Austin and Estabrook);
8) Turtleback scraper (Horvath); and 9) Fort Walton Incised sherd (White). Illustration by Scott E. Mitchell.

Copyright 2000 by the
ISSN 0015-3893


This issue is dedicated to the publication of nine papers
presented in a symposium at the 1997 Southeastern
Archaeological Conference meetings in Baton Rogue,
Louisiana. This "Current Trends" symposium, organized by
Donna Ruhl and Christine Newman, brought together some
of the people conducting the diverse temporal and spatial
research in our state. Donna and Chris subsequently orga-
nized the present volume and serve as its guest editors.
While not a representation of all the research going on in
Florida, these nine papers provide a good snapshot of the
kind of work characteristic of Florida archaeology. Research
conducted in the course of CRM contract archaeology,
conservation and management of protected sites, and
university-based research is represented. The individual
papers are introduced by Ruhl and Newman in their Fore-
On a personal note, the fourth paper-written by Chris
Newman, Ray McGee, and myself-is a continuation of the
research that Ray and I participated in between 1992 and
1994 when we excavated and wrote about the Mount Taylor
period component of the Groves Orange Midden on Lake
Monroe. We wanted to write something synthetic about this
interesting time period back then, but we decided it would
have to wait. The chance to study two important Mount

Taylor period sites, including the type site, seemed like a
good opportunity to pick up where we left off in 1994 (see
the articles in Vol. 47, No. 4, December, 1994). Since our
initial work at Bluffton, Chris Newman returned to the site
in 1999 to help monitor restoration work and the construc-
tion of a dock. Geophysical borings made at that time
revealed another 3 meters of deposits below the ubiquitous
concretion! Yet again we find that the severe alteration
caused by shell mining failed to destroy the entire
site-apparently the concretion served to protect what may be
more significant St. Johns River area wet site deposits.
It is with great pleasure that I present this issue of The
Florida Anthropologist. Donna and Chris deserve a lot of
credit for their hard work bringing these papers together,
and I think you will agree that the authors have crafted some
excellent contributions to our field. Because of the length of
this issue, the publication of abstracts from the 2000 Florida
Anthropological Society annual meeting and the spotlight on
the FAS award recipients will appear in the December issue.
Book reviews also will resume at that time.




VOL. 53(2-3)


"Current Trends and Research in Florida Archaeology" is the
focus of this special issue of The FloridaAnthropologist. What
began as two friends talking about the many exciting archaeo-
logical projects going on in the State soon developed into a
symposium atthe 1997 SEAC Convention and finally this issue
of The Florida Anthropologist. Florida is fortunate to have a
diversity of archaeological projects. This diversity is reflected
in project types, geographical areas, archaeological cultures)
or time periods) researched, and project investigators (see
Figure 1). In this issue we hope to illustrate that Florida
archaeologists conduct research at many scales and on a myriad
of subjects. These range from: 1) chert resource determinations
to mission-related farmsteads, 2) research focusing on the
Paleo-Indian to Post-Contactperiods, and 3) research sponsored
by federal, state, and local agencies to academic institutions and
independent, contract firms.
Nine articles comprise "Current Trends and Research in
Florida Archaeology," and are organized basically by their
temporal componentss. The first five articles are by Horvath;
Jones and Tesar; Austin and Estabrook; Wheeler, Newman, and
McGee; and Quitmyer and Jones. Four additional articles
complete the volume. The later papers are by Schwadron;
Ruhl; Marrinan, Halpern, Heide, and Blackmore; and White.
The first paper in this special issue is by Beth Horvath. Her
paper presents the results of investigations at the Colorado Site,
a Paleo-Indian lithic workshop in Hernando County. She
compares artifacts found at this site with those found at other
sites in the state. The project was conducted by Archaeological
Consultants, Inc. as a result of federally mandated mitigative
work by the Florida Department of Transportation and provides
us with another snapshot of this time period.
Calvin Jones supervised the monitoring of sewer line replace-
ment activities at the Wakulla Springs Lodge site in Wakulla
County. During this project, and during subsequent work
conducted by Louis Tesar and others at the site, Paleo-Indian
and Early Archaic period artifacts were recovered. Louis, who
worked closely with Calvin Jones before his death, has summa-
rized the work and presents us with a detailed description of
some of the recovered artifacts.
Bob Austin and Rich Estabrook present a method for identify-
ing source areas of lithic artifacts. They use features such as
fossil content, secondary inclusions, and rock fabric that are
visible with a low-power binocular microscope. The method,
when mastered, gives Florida archaeologists a useful, accurate
tool to discuss prehistoric lithic variation over time and space.
The Mount Taylor Period is the focus ofRyan Wheeler, Chris
Newman, and Ray McGee's paper. By reexamining the

literature, revisiting the Mount Taylor and Bluffton sites, and
placing the information in the context of today's archaeology
they have refined the definition of the Mount Taylor Period.
Zooarchaeologist Irv Quitmyer and paleontologist Douglas
Jones, team-up in the fifth article to discuss the impact of
prehistoric peoples on animal communities. Their detailed
analysis of the annual growth increments of hard clams
(Mercenaria spp.) fromfive prehistoric sites in the southeastern
United States reveal that the rate of exploitation was great
enough to reduce the average ontogenetic age of their popula-
tion significantly.
In the sixth article of "Current Trends and Research in
Florida Archaeology," Margo Schwadron begins by discussing
the investigations that were carried out by the National Park
Service at the DeSoto National Memorial in Manatee County.
Radiocarbon samples taken from the Shaw's Point site, in the
village area, indicate that the site was occupied for some 1,800
years during the Manasota through Safety Harbor periods. She
discusses numerous tool types, faunal remains, and pottery
types found at the site.
Recently collected archaeobotanical remains from Bernath
Place, a Santa Rosa/Swift Creek site in northwest Florida, are
discussed and compared with other sites dating to this period in
the article by Donna Ruhl. She compares her findings with
sites both in coastal and noncoastal settings and finds some
interesting patterns in the variability of plant husbandry.
Nancy White discusses the results of excavations conducted
at the Thick Greenbriar site, a small, riverbank, fanning village
in the upper Apalachicola Valley. A Fort Walton component
was identified at the site and archaeobotanical remains recov-
ered included maize. A later, Lower Creek/Seminole compo-
nent, also was found and her paper includes a discussion of this
time period and its representation in the archaeological record
in northwest Florida.
Rochelle Marrinan and her students provide an overview of
the Florida State University investigations at the O'Connell
Mission site outside of Tallahassee. Excavations have concen-
trated on structural evidence believed to represent the convent
and church of the mission, and on an isolated knoll that may
contain a mission-related or possibly proto-mission-related
William Marquardt served as discussant at the 1997 SEAC
convention symposium and presented a thoughtful and valuable
commentary-the papers benefitted from his remarks. Unfortu-
nately, time did not permit us to include his comments here.
We hope you enjoy the papers and find them a reflection of
some of the great work that is being conducted in our state. We

Tmu. FWfUDA AN'Fh4UAPAnicn T

.111TNF._q1WPWM1F FR 71

Vni. ;12-11


Figure 1. Archaeological sites discussed in this issue.

wish to thank each of the authors for their patience, time, and
effort in helping to successfully complete this issue of The
Florida Anthropologist. Our great appreciation is extended as
well to William Marquardt for his thoughtful overview and
comments during the 1997 SEAC symposium. We also would
like to thank all the anonymous, and not so anonymous,
reviewers for their work. In addition, our gratitude goes to

Scott Mitchell who designed and drew the creative cover
illustration based on selected artifacts presented in the volume,
and William Stanton who prepared the map showing sites
discussed in the issue. We ask that you, the reader, bear in
mind that many of the articles were written two years ago and
that any new information not included is a reflection on our
lack of "getting things together," not the authors. We hope you
will enjoy.



2000 VOL. 53(2-3)



Archaeological Consultants, Inc., P.O. Box 5103, Sarasota, Florida 34277-5103
E-mail: acinorth@compuserve.com

The Colorado site is not located in Colorado, rather it is in
west-central Florida. It is located about 5 km west of
Brooksville in Hernando County (Figure 1). This multi-
component site was discovered in 1989 by Florida Department
of Transportation (FDOT) Archaeologist George Ballo during
a survey for the State Road (SR) 50/50A road-wideningproject
(Ballo 1989). His survey involved surface reconnaissance and
judgmental subsurface testing within the 90-m-wide corridor.
The site extends along the north and south sides of SR 50
some 1.8 km east of Colorado Street, hence the site name. A

Figure 1. Location of the Colorado Site.

variety of environmental features are transected by the high-
way, including sand ridges, ponded depressions, and an
intermittent stream. Portions of the site have been cleared for
pasture. The east end of the site, prior to the recent road
construction, retained its natural vegetation. Numerous large
sinkholes also are present in the area, including one transected
by the highway.
The Colorado site has been divided into five distinct
subareas (Figure 2). Four of these were defined by Ballo based
on physiographic and hydrologic features, as well as artifact
types, density, and distribution (Ballo 1989). The fifth,
.-- referred to as Pond 3, is located within one of the
proposed retention ponds associated with the SR 50
road improvement project. Subarea A is largely
coterminous with the ridge situated at the western end
of the site. It is primarily a Middle Archaic lithic
scatter though one piece of Pasco Plain ceramic was
recovered. Subarea B is a small, temporally indetermi-
nate lithic scatter which contained one small sand-
tempered-plain sherd. Subarea C contains components
from the Early Archaic through the recent Historic
periods. One sand-tempered, perforated ceramic disc,
possibly a spindle whorl, was recovered from Area C.
Subarea D is a relatively ill-defined, temporally indeter-
minate lithic scatter. This subarea is important due to
its chert outcrops. Pond 3 is located to the southwest of
Subarea D, on the south side of SR 50. This locale
contains a possible Paleoindian component of the site,
which is the focus of this paper.
Pond 3 is roughly 90 m east/west by 60 m
north/south. It is located within a tree farm which
previously had been planted in watermelons. As a
result, there had been some site disturbance. At the
time of initial survey, the area was not considered to
have a high probability for archaeological sites because
it is on Micanopy loamy fine sand, which is a some-
S what poorly drained soil. Other areas of the Colorado
site are on better-drained soils. Furthermore, there are
no readily or permanently available sources of potable
water within about 500 m of this locale.


Archaeological investigations within the Pond 3
area began with the excavation of shovel tests at 20-m


VOL. 53(2-3)




intervals along four transects spaced 20-m apart. Based on the
preliminary results, an additional two transects were placed at
10-m intervals. The additional testing took place in the
southern portion of the proposed pond around the shovel tests
that contained the highest density of materials as well as the
largest flake sizes. Though artifact counts were not very high
(the greatest number offlakes in any of the tests was 13), the
debitage was generally much larger than the waste flakes
observed in the other subareas of the site. Thus, a second,
more intensive phase of testing was begun (Figure 3).
This second phase consisted of the excavation of 11 units
(8 1- x-2-m and 2 2-x-2-m) and three trenches. One of the
trenches was 1-m wide by 6-m in length and the other two
were 1-m wide by 10-m long. The trenches were quickly
excavated at the end of the project in the hope of collecting
additional temporally diagnostic tools. The soil was not
screened, but the tool forms noted were piece plotted. The
units were excavated in 20-cm levels and the soil was screened

and additional funding was received from FDOT to conduct
further investigations.
In January and February 1997, ACI conducted mitigative
excavations in the Pond 3 area of the Colorado Site. The
investigations were structured to obtain data that could be
compared to those from the Harney Flats site in Hillsborough
County (Daniel and Wisenbaker 1987). Specifically, excava-
tion of a 220-m2 block in the southeast section of Pond 3 was
conducted. This area was divided into 4-x-4-m units that were
excavated as four 2-x-2-m units. We had anticipated that the
base of the excavation would be roughly 50 to 60 cm below
surface, but the excavations extended to a depth of 120 cm in
the north-central portion of the block (Figure 5). Excavation
was conducted in 10-cm arbitrary levels except for the upper
20 cm of disturbed topsoil, which was removed and discarded
without screening. The tool forms, except for many of the
flake tools, were piece plotted. Many of the flake tools were
not recognized as such until they were washed and analyzed.




\jt...\.\.\ \ x x- x\\x\ x' -- A \\\\.Wz u \ jz -

CE,------ -- - --- AA - -------------




~ C\~,~32 m-- w;

Figure 2. Location of the Five Site Subareas.

through 6.4-mm mesh. Again, the tools were piece plotted
whenever possible. A 19-liter sample from each level of one
2-x-2- m unit (Unit D) was sifted through window screen to
recover a sample of microdebitage.
Only one temporally diagnostic biface fragment was
recovered during this initial phase of testing, and that was the
stem of an Archaic projectile point. However, two lanceolate-
shaped reforms were also found. One appears to be a
Suwannee/Simpson preform and the other is classified as a
lozenge-shaped biface (cf. Daniel et al 1986) (Figure 4). The
relatively high percentage of unifaces, accounting for 10% of
the tools recovered, combined with the recovery of these
reforms, suggested that a Paleoindian component might be
present. Consequently, additional work was recommended


Two slightly different soil profiles were exposed duringthe
investigations, the difference between the two being the depth
of the sand. Most of the site had an upper 20-30 cm layer of
grayish-brown loamy sand underlain by another 20-30 cm of
light yellowish-brown sand. Beneath the sand is a dark
yellowish-brown sandy clay. During the first course of
fieldwork, a depression was discovered during the excavation
of Units D and S. Within those units, the clay subsoil was not
encountered until approximately 120 cmbs. The depression
appeared to be centered on these units. The trenches excavated
proximate to these units revealed a slope in the clay subsoil



\w..WXX\tKX\\ -


-^i?^;~'-~ L --- r:-'1f--------- YKr~T -sC s^t-v.'



;El EN rA A .. \. _

- \\\\\`ZC~

2000 VOL. 53(2-3)

I\ \ \ \\ \ \ \


-\~Y\\\\\\\\\\\\\ ~~~n

\\\\\\~\\\\ r~m \\\\\\\\\\\\\\\\


SR 50 R-O-W



S Trench


Artifact Counts

1 to 3
4 to 6
7 to 9
10 to 12
13 to 15

20 met

Figure 3. Phase I and H Testing within the Pond 3 Area.

leading down to the area of Units D and S. Another depres-
sion was uncovered during the Phase III investigations. That
one appears to have been centered on Unit 19 (Figure 6).
Within that unit, the base of the depression was encountered
at 115 cmbs. Neither of these two depressions was observable
from the surface. Geomorphological analyses by Sylvia
Scudder (1997) suggest that these features are irregularities in
the underlying clay sediments, or possibly erosional features of
the original ground surface.
Scudder (1997) reports that the subsurface depression seen
in the profile of Unit D and adjacent units is most strongly
expressed in the Bt horizons. It presents a picture of a "cup"
formed by the subsurface clay layer which was gradually filled
up with sand. The first layers of sand deposited conformed to
the concave inner surface of the cup but later additions (the

upper AE and Ap horizons)
were flatter. The unique
character ofthe Btl horizon
that outlines or forms the
lining of the depression dif-
X X ferentiates it from other
segments of the Btl in the
Pond 3 area. Its greenish
Dirt Access color is unlike the brown of
Rothe adjacent Btl and indi-
cates that it was constantly
moist. If this depression
X X filled slowly, the early
R stages would have been a
thin sand layer over damp
clay, and would probably
X X not have been inviting for a
habitation site.
It was suggested that
this feature may have been
East utilized as a dump by the
Trench site's occupants (Scudder
MV 1997); however, the
V M debitage assemblage argues
otherwise. A comparison of
the assemblages from
X 4 within and around the de-
pression revealed that they
are virtually the same in
terms offlake size and type.
Table 1 compares the flake
types and flake sizes from
ers the units excavated during
the initial phase of testing.
The microdebitage recov-
ered during the Phase mI
investigations also argues
against the depression be-
ing used as a dump.
Microdebitage accounted
for up to 40% of the total
debitage recoveredfromthe
units within and adjacent to the depression (Table 2). The
microdebitage from the three units south of the depression
(Units 7, 8 and 10) accounted for less than 20% of the total
debitage assemblage from those units. If the depression had
been used as a dump, there would be very little microdebitage
Within the depression uncovered during the Phase III work,
the artifacts appear to have followed the contour of the
underlying clay. In looking at the reversed stratigraphic
depths (i.e. depth relative to the clay surface as opposed to the
original ground surface), it is clear that almost 80% of the
tools are within 30 cm of the clay subsoil (Table 3). This may
suggests that the sand immediately above the clay subsoil
served as a living floor during Paleoindian times, with the site
later being covered with aeolian sands over the years.



completed "bifaces" (Figure 7) making up
the stages of reduction. In this continuum,
5. there is a general decrease in absolute size,
3 .. progressive thinning, increased symmetry,
'.S : shaping of the margins, and refined percus-
sion flaking. Roughouts are bifacially
:. reduced stone tools in the initial stages of
-." ... ,, manufacture. The reduction of this form is
Si : focused mainly on trimming rather than
shaping (Thomas and Bierwirth 1983).
: \ .. ." ... Blanks are modified flakes or bifacially
worked pieces that are larger than finished
S. bifaces and adequate in size and form for
.. *making an implement (Figure 8). How-
.1 ever, the final form of the tool is not evi-
y .1 -' .dent. Preforms are unfinished, unused
N. forms that are most often symmetrically
S shaped. They possess a lower thinning
.. i index and more acute edge angles than
..: blanks. The finished tool form can be
,- .ascertained during this stage.
Projectile points are finely flaked,
i^ 8 symmetrically shaped bifaces which fit into
S*defined type category (cf. Bullen 1975).
They often possess a hafting element and
functioned as the tip of an atlatl, spear, or
A arrow. The Suwannee point is a "large and
fairly heavy, lanceolate shaped, slightly
waisted point with concave base, basal ears,
and basal grinding of bottom and waisted
Figure 4. A) Suwannee/Simpson Preform; B) Lozenge-shaped Biface (actual parts of sides. Basal thinning and sugges-
size). tions of fluting are but rarely present"
(Bullen 1975:55). The base is laterally
thinned, rather than the flake removal
Analysis originating from the base (Goodyear, et al. 1983:46). The
pressure flaking on this point is usually restricted to the biface
The tool assemblage consists ofbifacefragments, reforms, margin for final shaping and for sharpening the tip andblade
blanks, unifaces, flake tools, cores, hammerstones, and tested margins. The Simpson point is smaller and thinner than the
cobbles. Bifaces are bifacially chipped pieces of stone with Suwannee and the basal ears are not as prominent (Figure 9).
flake scars on both the dorsal and ventral surfaces and may Bullen (1975:56) describes a Simpson point as "a wide bladed,
have use damage on the margins. There is a general reduction relatively narrow waisted, fairly thin, concave based, medium
continuum with "roughouts", "blanks", reformsms, and to large sized point with grinding on the bottom and waisted

Table 1. Debitage Distribution Relative to the Depression Uncovered During the Initial Testing.

Flake Size
Small Medium Large X Large XX Large
Inner 297 17.3% 888 51.8% 311 18.2% 125 7.3% 92 5.4% 1713
Outer 231 12.4% 996 53.4% 402 21.6% 122 6.5% 114 6.1% 1865
Flake Type
Primary Secondary Non-decortication Shatter
Inner 68 4.0% 184 10.7% 1379 80.5% 82 4.8% 1713
Outer 114 6.1% 178 9.5% 1498 80.3% 75 4.0% 1865
inner units are within the depression (Units D and S), outer units are well outside the limits of the depression


2000 VOL. 53(2-3)


0 2 4 6 8 10 12 14 16 meters

Figure 5. Base of Excavations (surface of the clay) and Phase m Unit Locations.

edges. Basal ears are present but are not as developed as in the
Suwannee point. Basal thinning is present, but, also, is not
well developed." The tapering or "waisted" base is the
distinguishing feature of this point type (Daniel and
Wisenbaker 1987:53). Lozenge shaped bifaces are small,
sometimes delicately made, lanceolate or leaf-shaped points
(Figure 10). These are usually dated to the Paleo-Indian
period based upon their technological attributes (Daniel and
Wisenbaker 1987; Goodyear, et al. 1983). Lateral thinning is
common though there is no lateral grinding. Some of the
basal flake scars originate from the base. The base of this tool
tends to be more rounded than that seen on other Paleo-Indian
lanceolate points.
Unifaces are pieces of stone possessing flake scars primar-

ily on one face only. Following Daniel and Wisenbaker's
(1987) analysis of the unifaces from Harney Flats, the unifaces
have been divided into two general categories of technological
organization-curated tools/personal gear and expedient
tools/situational gear (cf. Binford 1977, 1979). The curated
tools tend to display considerable effort in manufacture,
evidence oftool maintenance, portability, andflexibilityintool
form (Daniel and Wisenbaker 1987:62). This category
includes end scrapers, discoidal scrapers (Figure 11), and
oblong scrapers. End scrapers are most often triangular or
tear-drop shaped with the broad worked margin opposite the
end of the flake with the bulb of percussion (Figure 12). The
end scrapers from Harney Flats tended to have a flat ventral
surface, the flake blanks were non-decortication, and the tools



0 c 200 400

ZONE I 10YR5/2
20 Grayish brown sand
ZONE I 10YR5S3 10YR6/6
40 Brown sand/Yellowsh brown sand

60 0 rodent burrow
qWrodent burrow
so- Very pale brown sand

-- ZONEV 10YR5/4 Yellowish brown sand
Yellowish brown sandy clay UNEXCAVATED

I I 2
0 20 40

Figure 6. Unit 19 West Wall.

Table 2. Microdebitage and Macrodebitage Samples Recover
the Phase III Testing.

Unit Microdeb. Macrodeb. Total Flakes Microdeb

7SE 57 258 315 18

8SE 61 331 392 15

10SE 58 411 469 12

14SE 113 327 440 25

17SE 125 346 471 26

18SE 708 939 1647 43

20SE 164 483 647 25

21SE 192 505 697 27

22SE 451 1073 1524 29

were thicker at the scraping end (Daniel and Wisenbaker
1987:63). They appear to have been hafted tools. The
discoidal scrapers, on the other hand, were believed to be
hand held. They are similar in manufacture to the end
scrapers except that they have a more circular outline. Oblong
scrapers are oval or ovoid in plan view. Most lack a bulb of
percussion and the morphological characteristics suggest that
the tools were made from thick flakes as opposed to blades.
Marginal retouch tends to occur along both lateral margins
(Figure 13). Daniel and Wisenbaker (1987:72) note that
although these tools are considered to be unifaces, flake
removal on the ventral surface may be conducted to flatten the

ed from surface.
The expedient tools include the thick and thin
unifaces (Figure 14). The difference between these
two types is the thickness of the worked edge. If the
i. % margin was 10 mm or greater, it was classified as a
thick uniface. A flake tool with a worked margin
'. 1% thickness of less than 10 mm was classified as a thin
5.6% uniface. Daniel and Wisenbaker (1987:63) indicate
that both classes were made from flakes of various
.4% sizes though tended to be larger than the formal
unifacial tools described previously. For this analy-
.7% sis, if the flake possessed unifacial marginal retouch,
it was classified as a uniface. If there was simply
S use damage on the margin, it was classified as a
.00/ flake tool.
Cores are pieces of stone that possess flake scars,
.3% indicating that flakes were removed for possible tool
manufacture. They usually do not exhibit usewear
.5% though occasionally, they are re-utilized as
hammerstones. Hammerstones are large pieces of
stone possessing rounded edges and areas of exces-
sive battering due to direct or indirect percussion.
The lithic assemblage recovered during both phases of
testing resulted in the recovery of 546 tools and over 41,000
pieces of debitage. Almost all of the lithic material was the
locally available Brooksville Quarry Cluster chert (Upchurch
et al. 1982). Most of the material was honey to light tan in
color, and it rapidly lightened when removed from the damp
ground. This is indicative of a very porous lithic material.
The quality of the chert is quite variable. Fine-grained and
very coarse-grained chert matrix was occasionally present
within the same tool. The raw material for these tools was
likely obtained from the intermittent stream drainage to the
north. Numerous chert cobbles and evidence of early reduction


2000 VOL. 53(2-3)


Table 3. Phase III Tool Distribution Relative to the Clay Surface.

T I__ 0-Depth above clay (cm)
Tool Type 0-10 10-20 20-30 30-40 40-50 50-60 60-70 70-80 80-90 Total
Biface Count 2 4 3 1 0 2 0 0 0 12
% of Row 16.7 33.3 25.0 8.3 0.0 16.7 0.0 0.0 0.0 100.0
Blank Count 17 48 22 13 4 1 0 1 0 106
% of Row 16.0 45.3 20.8 12.3 3.8 0.9 0.0 0.9 0.0 100.0
Core Count 8 16 3 5 0 0 0 0 0 32
% of Row 25.0 50.0 9.4 15.6 0.0 0.0 0.0 0.0 0.0 100.0
Flake tool Count 41 63 41 29 6 6 4 1 1 192
% of Row 21.4 32.8 21.4 15.1 3.1 3.1 2.1 0.5 0.5 100.0
Hammerstone Count 1 0 1 0 0 0 0 0 0 2
% of Row 50.0 0.0 50.0 0.0 0.0 0.0 0.0 0.0 0.0 100.0
Preform Count 4 7 3 1 1 0 0 1 0 17
% of Row 23.5 41.2 17.6 5.9 5.9 0.0 0.0 5.9 0.0 100.0
Uniface Count 10 15 9 5 3 1 1 0 0 44
% of Row 22.7 34.1 20.5 11.4 6.8 2.3 2.3 0.0 0.0 100.0
Total Count _83 153 82 54 14 10 5 3 1 405
Total % 20.5 37.8 20.2 13.3 3.5 2.5 1.2 0.7 0.2 100.0

activities are present in the drainage and in several nearby
sinkholes. Obtaining lithic raw materials for the production
of stone tools was not a problem in this area.
The tool assemblage contains 19 biface fragments, 134
blanks, 39 cores, 270 flake tools, 6 hammerstones, 20 pre-
forms, and 58 unifaces. Several of the broken bifaces, blanks,
and reforms were refitted during the analysis. In refitting
some of the fragments, it became apparent that many of the
tools and tool rejects had multiple functions or at least could
be classified as having different functions. For example, one
of the blanks, which broke during the initial reduction se-
quence, had one end discarded, but the other end had been re-
utilized as a hammerstone.

Figure 7. Biface fragments from the Phase III testing.

In their analysis of Hernando projectile point technology at
the Rock Hammock site in Hillsborough County, Austin and
Ste. Claire (1982:186-187) discuss a phenomenon they term
"emotional displacement." As they put it "It is possible that
during the time immediately following the fracture of a nearly
completed implement, the frustrated manufacturer displayed
his anger by excessively discarding one or more of the remain-
ingfragments" (Austinand Ste. Claire 1982:187). Undoubtab-
ly, this activity occurred at many if not most lithic tool manu-
facturing loci. Many of the refit tools from the Colorado Site
were in the blank and preform stages of the manufacturing
continuum. However, scavenging and recycling of the tool
fragments could also account for the displacement of the
Most of the tool refits occurred
between the fragments recovered
from the block excavation (Figure
15). Each tool is represented by a
unique symbol in the figure. Two of
the tools became fragmentary as a
result of post-depositional breakage
along incipient fracture planes. The
other tools all appeared to have
^ broken during the manufacturing
process. There was a variety of
Manufacturing failures including
lateral snaps, perverse fractures, and
S... reverse fractures in addition to frac-
ftures caused as a result of the incipi-
ent fracture planes present in many
of the chert fragments.
Several of the biface and blank
fragments recovered during the
Phase HI investigation were classi-




Figure 8. Lanceolate biface fragments (actual size).


2000 VOL. 53(2-3)

gure 9. A) Simpson Preform; B) Suwannee/Simpson Preform; Suwannee Preform actual size).
Figure 9. A) Simpson Preform; B) Suwannee/Simpson Preform; C) Suwannee Preform (actual size).


Figure 10. Lozenge-shaped Biface Blanks (actual size).

Phase Il investigation were classified as Florida Archaic
Stemmed points. Interestingly, almost all of these projectile
points had been manufactured from non-local chert, mostly
from the Hillsborough River Quarry Cluster (Upchurch et al.
1982), including one Archaic tool that is made from Bay
Bottom chert from the Tampa Bay coastal area. None of the
debitage recovered during the excavations appeared to be from

Figure 11. Discoidal Scraper.

either of these two source areas.
The debitage assemblage suggests that
middle-to-late stage lithic reduction activities
were being undertaken. The most noticeable
aspect of the assemblage from Pond 3 is the
large flake size. The largest flake recovered was
17 cm long and the site's mean flake weight is
2.5 g. The prevalence of the larger flake sizes
without the higher percentages of primary and
secondary decortication flakes (Figure 16)
suggests that the blanks that were being reduced

were simply larger in size than those recovered from
sites dating from the Archaic and more recent periods.
This may be expected based upon the large bifacial
tool sizes associated with the Paleoindian Period.
Many of the discarded flakes were of sufficient size
and form to be manufactured into bifacial or unifacial
implements, but they showed no modification or use
damage. This suggests that selective utilization was
occurring. In other words, those flakes evidencing the
better flaking qualities and appropriate size for the
anticipated tool form were selected. The abundance of
the chert allowed the manufacturers to pick and
choose the flakes and blanks they wished to reduce.
The quality of the local chert is quite variable and
much of it has poor flaking qualities due to the frac-
ture planes, mineral inclusions, and voids. John
Allen, a local flint knapper, noted that this material
was very difficult to work. Thus, the poor flaking
qualities of the chert may have necessitated the use of
larger flake blanks.
There were 546 tools divided into seven tool type
categories at the Colorado Site. Though there is a
high tool count, almost half of these (some 48%) are expedient
flake tools. The relatively few types of tools may suggest a
short-term occupation. The total tool assemblage, including
the blanks, manufacturing rejects, reforms, and tested
cobbles, make up only 1.4% of the entire lithic assemblage.
The tool assemblage contains a high percentage of tool
manufacturing items including blanks, reforms, cores,
hammerstones, and tested cobbles. That combined with the


Figure 12. A) Ovoid End Scraper; B) Trapezoidal End Scraper (actual

2000 VoL. 53(2-3)


abundant debitage suggests that tool production may have
been one of the primary activities occurring on site. However,
as Binford (1978:335) notes, activities directly related to the
primary function of the site may not leave archaeological
The presence of bifaces, unifaces, and flake tools suggests
that more than stone tool manufacturing was being undertaken.
."" Other activities could have included hunting, butchering, and
meat or hide processing. Almost half of the tools were flake
.:. tools and most of those had unifacial use damage. This
'.' factor,combined with the relatively high number of unifaces
S' recovered, suggests that scraping activities were important on
*this site and were perhaps associated with hide preparation and
working. There was very little evidence of unifacial edge
':'?damage associated with use on hard materials. One would also
expect that cooking and possibly storage activities occurred.
However, the items utilized for these activities such as gourds,
hides, and wooden implements would not have survived in the
acidic soils present on the site.

Figure 13. Oblong/Turtleback Scraper (actual size).

Figure 14. A) Thin Uniface/Side Scraper H; B) Thick Uniface/Side Scraper U (actual size).


COLORADO SrrE Exc~varios


Unit U

Trench 1

Unit D

^---- --

o A

0 kK CD
n"^ ^
1^ --fO

Limits of Block Excavation

2 meters
2 meters

Figure 15. Locations of the Refit Tool Fragments.

Comparisons to Harney Flats

The Harney Flats Site was a large multi-component site
located on a bluff overlooking Harney Flats. A variety of
vegetative regimes were present on the site, including xero-
phytic uplands, mesic vegetation along the slopes with marsh-
like vegetation to the west and a hydric hardwood swamp to
the east (Daniel and Wisenbaker 1987:4). In addition to the
Paleoindian component, evidence for occupation during the
Early and Middle Archaic periods was present. Ephemeral
usage during the later ceramic periods was also indicated.
Evidence of Paleoindian occupation was provided by the
recovery of Suwannee, Simpson, and Bolen projectile
points/knives, as well as numerous unifacial tools. In addition,
a variety of other tool forms and debitage was recovered. The
distribution of the debitage and tools across the site suggested
that there were activity areas within the site though the main
function of the site appeared to be the manufacture of chipped
stone tools. Unlike the Harney Flats site, the boundaries of the
Colorado site were not ascertained outside of the FDOT right-
of-ways and retention ponds. Thus, there is the potential that
activity areas are present outside of the locale tested by ACI.
Ethnographic studies comparing tool assemblages have

shown repeatedly that "specific activities can be performed by
a range of different tool forms within a single cultural system
and that the decision to use one tool form over another is
situationally determined in accordance with the larger settle-
ment organizational strategies" (Cable 1996:122). Conversely,
one tool can be utilized for a variety of tasks.
At the Colorado site cobbles and outcrops are present less
than 250 m away. The importance of easily available lithic
raw material and how that factor effects artifact assemblages
is probably best described by Keeley (1982:803-804):

If a group is occupying a site near a source of lithic raw
material, they may prefer to employ expedient implements
while conserving or even ignoring hated tools. The assem-
blage at such a site would contain large amounts of waste and
a relatively large number of big, minimally retouched tools.
This same group, occupying another location where raw
material was more difficult to obtain may preferentially
employ its hafted implements so that the assemblage depos-
ited would contain little waste and a high proportion of small
intensely worked tools. Although the assemblages from these
two sites would be quite dissimilar, they might nevertheless
be the remains of the same group and could even be of the
same suite of activities.

2000 VOL. 53(2-3)



I i J


The assemblage at the Colorado Site had relatively little
evidence of retooling but it did have a high concentration of
waste and utilized debitage. The lack ofretooling and curated
tools may be the result of chert quality. Many of the almost
complete tools, blanks, and cores broke at incipient fracture
planes, mineral inclusions, and/or voids in the chert. It could
also be, however, that the habitation portion of the site, where
those types of implements would be expected, was outside of
the area subject to the investigations. The high number of
blanks and reforms recovered (29.2%//N=166) and the




35.0% -




10.0% 0

o .^ ?^ .- _,f r.I

assemblage (354 of 1094/32.3%). Unfortunately, the debitage
from Harney Flats was not examined for flake tools as inten-
sively as was done at the Colorado site. Thus the apparent
extensive use of expedient flake tools at the Colorado Site may
simply be the result of the analysis.
The Colorado Site is believed to have been a short-term
occupation site established to utilize the locally available
resources, including the chert outcrops in the area. This is
based upon a number of different factors. For one, there is
relatively little evidence for stone tool maintenance. As Yellen

primary secondary non-decortication shatter
Flake Type Category

Figure 16. Flake Size and Type Distribution.

abundance of debitage suggests that this portion of the site was
abifacialblank or core manufacturing station. Theabundance
of large debitage would have provided ample material for
expediently utilized flake tools. This in turn would have
provided a means for saving the "good" or curated tools for
later use in areas where lithic materials were scarcer.
The use of expedient tools at the Colorado Site was
common. Flake tools accounted for 270 (49.4%) of the tool
forms recovered and the more expedient unifacial tools, i.e. the
thin and thick unifaces (N=36), accounted for another 12.3%
of the tool assemblage. Thick and thin unifaces from the
Harney Flats site made up almost a third of the total tool

(1977) discovered during his studies of the !Kung Bushmen,
there is a greater accumulation of evidence for maintenance
activities the longer a site is occupied. Rehafting and/or
retooling does not appear to have been a prime activity at the
Colorado Site. There does not appear to have been a discrete
living area present at the site and unlike Harney Flats, there
was no apparent "site furniture" such as caches of
hammerstones and abraders. This could, however, be due to
the restricted area investigated. The soil analysis also sug-
gested that the site had not been occupied for an extended
period of time due to the low phosphate content in the soil.
There are some possible activity loci within the area




* Biface
X Blank
- Core

A Hammerstone
* Preform
- Uniface

Figure 17. Distribution of the Tool Forms Across the Block Excavation.

2000 VOL. 53(2-3)



investigated at the Colorado Site. In looking at the piece
plotted tools (Figure 17) some areas of clustering can be
observed. The flake tool category is not included. Unlike the
Harney Flats site, there did not appear to be any caches of tools
or tool form blanks. There are two clusters (defined as three
or more of the same tool type within 2- to 3-m of each other)
ofunifaces. One, consisting of seven unifaces, is located in the
southwest portion of the block excavation and the other,
consisting of five unifaces, is centered on the depression.
Cluster one contains five thin unifaces (Side Scraper II [Coe
1964]) and an oblong scraper and an end scraper. From
cluster two, located in the depression, there is one end scraper
with the others being classified as thin unifaces. That group
contained two from the Side Scraper I category, and one each
from the Side Scraper II and III categories. There is no
clustering of the end scrapers as was seen at Harney Flats
though there were significantly fewer end scrapers recovered
from the Colorado site. This may be related to the lack of
retooling as opposed to lack of use of these types of tools.
The cores were recovered primarily from three cluster
areas. The first area consisted of four cores within 2-m in the
southeastern portion of the block. Another cluster of six cores
was in the western portion of the block. The third cluster,
which was slightly more dispersed, was located south of the
depression between the other two core clusters. There is no
clear clustering of the hammerstones, though there is a line of
four extending through the west central portion of the block
In general, the blanks are fairly evenly distributed.
However, there are some small pockets where several blanks
were recovered within about a meter of each other. The
reforms have a tighter cluster, being located to the west and
north of the depression. Eight of the blanks were recovered
along that western slope. Three others, including the Simpson
projectile point/knife preform, are located slightly northeast of
the depression. The other reforms are widely scattered
through the area south of the depression.


In sum, we believe that the Pond 3 area of the Colorado
Site was occupied primarily during the Paleoindian period
based upon the recovery of Suwannee and Simpson reforms
and blanks as well as the numerous lanceolate bifaces, lozenge
shaped bifaces, and unifaces. The most visible function of the
site appears to have been for the manufacture of blanks from
the locally available chert. The relative lack of completed
bifacial tools and microdebitage suggests that final tool
shaping and sharpening did not occur on site. The high
percentage of flake tools may indicate argues for short-term
use. There were no apparent special-use areas of the site.
This also suggests short-term use as opposed to a long-term
occupation. The chemical and physical analyses conducted on
the site's soil failed to reveal evidence of long-term occupa-
tion. The relatively high percentage ofunifaces and flake tools
with unifacial damage suggests that scraping activities
occurred, perhaps related to meat and hide processing.


Archaeological Consultants, Inc would like to thank
Richard Adair, Environmental Administrator, ProjectDevelop-
ment and Environmental (PD&E) for FDOT District Seven,
George Ballo, Cultural Resources Administrator, Central
Environmental Management Office, FDOT, and Ron Pscion
and Rebecca Spain Schwarz of Post, Buckley, Schuh, and
Jernigan for their assistance and the opportunity to work on
this great site. We would also like to thank Monroe Treman
and his sister Alma Graddy for providing information on the
local history and environment. Chris Jacks, the tree farm
owner, is also thanked for removing the trees prior to our
excavations and providing information on the recent site land-
use activities. Sylvia Scudder of the Florida Museum of
Natural History conducted the geomorphological investiga-
tions. Our most able field crew consisted of Jay Hardman,
Kenny Klopp, Tim Lewis, Tom McIntosh, John Nichols,
Wendy Nettles, Terry Simpson, Vince Warner, Wendy
Weaver, Lorna Weill, and Jerry Westphal. Thanks to all for a
job well done. I would also like to thank Donna Ruhl, Chris
Newman, Bob Austin, Joan Deming, and the anonymous
reviewer who critiqued this paper. Their editorial work has
improved the quality of the paper and any errors in content or
thought processes remain mine.

References Cited

Austin, Robert J., and Dana Ste. Claire
1982 The Deltona Project: Prehistoric Technology in the
Hillsborough River Basin. University of South Florida,
Department of Anthropology, Archaeological Report
Number 12.
Ballo, George R.
1989 Archaeological Resource Assessment Survey SR-50/50A
from Colorado Street, West of the City of Brooksville, East
to US-301. Central Environmental Management Office,
Florida Department of Transportation, Tallahassee, FL.
Binford, Louis R.
1977 Forty-seven Trips: A Case Study in the Character of
Archaeological Formation Processes. In WorkingatArchae-
ology, edited by Lewis R. Binford, pp. 243-268. Academic
Press, New York.
1978 Dimensional Analysis of Behavior and Site Structure:
Learning from an Eskimo Hunting Stand. American
Antiquity 43:330-361.
1979 Organization and Formation Processes: Looking at Curated
Technologies. Journal of Anthropological Research
Bullen, Ripley P.
1975 A Guide to the Identification of Florida Projectile Points.
Kendall Books, Gainesville, FL.
Cable, John S.
1996 Haw River Revisited: Implications for Modeling Terminal
Late Glacial and Early Holocene Hunter-Gatherer Settle-
ment Systems in the Southeast. In The Paleoindian and
Early Archaic Southeast, edited by David G. Anderson and
Kenneth E. Sassaman, pp. 107-148. University ofAlabama
Press, Tuscaloosa.
Coe, Joffre Lanning
1964 The Formative Cultures ofthe Carolina Piedmont. Transac-




tions of the American Philosophical Society, Volume 54,
Part 5. Philadelphia.
Daniel, I. Randolph, Jr. and Michael Wisenbaker
1987 Harney Flats: A Florida Paleo-Indian Site. Baywood
Publishing Company, Inc., Farmington, New York.
Daniel, I. Randolph, Jr., Michael Wisenbaker, and George Ballo
1986 The Organization of a Suwannee Technology: The View
fromHareyFlats. TheFloridaAnthropologist39(1-2):24-
Goodyear, Albert C., Sam B. Upchurch, Mark J. Brooks, and Nancy
N. Goodyear
1983 Paleo-Indian Manifestations in the Tampa Bay Region,
Florida. The Florida Anthropologist 36(1-2):40-66.
Keeley, Lawrence H.
1982 Hafting and Retooling: Effects on the Archaeological
Record. American Antiquity 47(4):798-809.
Scudder, Sylvia
1997 Soil Investigations at the Colorado Site, Hemando County,
Florida. Ms. on file, Archaeological Consultants, Inc.,
Thomas, David Hurst and Susan L. Bierwirth
1983 Material Culture of Gatecliff Shelter: Additional Stone
Tools. In TheArchaeology ofMonitor Valley:2. Gatecliff
Shelter. Edited by David Hurst Thomas. Anthropology
Papers Vol. 59, Pt. 1. American Museum of Natural
History, New York.
Upchurch, Sam B., Richard N. Strom, and Mark G. Nuckels
1982 Methods ofProvenance Determination ofFlorida Cherts.
Report on file, Department of Geology, University of South
Yellen, John
1977 Archaeological Approaches to the Present. Academic
Press, New York.

2000 VOL. 53(2-3)




Bureau ofArchaeological Research, Division of Historical Resources, 500 South Bronough Street, Tallahassee, FL 32399-0250
E-mail: Ltesar@mail.dos.state.fl.us

The Wakulla Springs Lodge sewer line replacement project
(hereinafter referred to as the project) resulted in one of the
last emergency salvage archaeology projects supervised by B.
Calvin Jones.2 Unlike planned research projects, emergency
projects often are not funded beyond the lead archaeologist,
depend heavily on volunteer participation, and require
adjustments in field methods in order to meet construction
deadlines. Unfortunately, because funding and staff assistance
for analysis and report preparation are also lacking or inade-
quate, such project activities generally take longer to complete
than those of fully funded projects.
Public participation in archaeological projects makes a
difference, both financially in reducing project costs and
culturally in improving archaeological resource stewardship.
During most of his 30 years as a state archaeologist, Jones
relied on volunteer assistance to perform emergency archaeo-
logical salvage excavations in advance of imminent construc-
tion activities, sometimes at sites revealed by those very
activities (see Jones et al. 1998). On the sewer line project,
Jones's crew consisted of Florida Park Service staff and area
residents, many of whom had worked as volunteers on his
other projects.
Archaeological excavation in the portion of the project that
affected the Wakulla Springs Lodge site (8WA329) produced
important archaeological data, demonstrating that site to be
among the very few stratified Paleoindian-Early Archaic sites
recorded in Florida. Investigations were limited to the area
affected by the project and were not expanded to include site
loci of particular research interest. However, excavation at
this site provided an opportunity to increase our understanding
of the cultural changes that occurred at the end of the Ice Age.
Because the project was very accessible to the public and
produced important findings widely reported by the media, the
site received hundreds of visitors eager to see artifacts and
observe project excavations. Thus, the project resulted in an
increased appreciation of cultural resource stewardship among
state land managers and the public.


Wakulla Springs is located about 18 km (11 mi.) south of
Tallahassee, Florida (Figure 1). The main spring is famous for
its crystal clear water and for the remains of Pleistocene

megafauna and early Native American artifacts. The artifacts
led to the location being recorded in the Florida Master Site
File as site 8WA24. Subsequently, the underwater component
was designated in the site file as 8WA24A, when the recorded
site area was expanded to include the adjacent upland compo-
nent, designated as 8WA24B.

Figure 1. Map showing location of Edward Ball Wakulla
Springs State Park (from Bryne 1988; reproduced with the
permission of the Florida Division of Historical Resources,
Bureau of Archaeological Research, Tallahassee).


VOL. 53(2-3)




In 1930, while scouting the northern part of the pool
during renovation work, then spring owner George Christie,
"spotted the characteristic brownish-black color of fossil bone
protruding from the bottom sediments 26 feet below him"
(Rupert 1991:11). Using a long-handled rake, Christie
recovered a right lower jaw, a portion of a tusk, and other
bones, before seeking assistance from State Geologist, Herman
Gunter. Gunter organized a team, including Florida Geologi-
cal Survey paleontologists Clarence Simpson and George
Ponton. Gunter's team used a pontoon raft, a suction pump,
and outfitted a diver with a diving helmet to recover the
remainder of the nearly complete American Mastodon
(Mammut americanum), which is now on display in the
Museum of Florida History in Tallahassee (Rupert 1991:11).
From 1955-1957, Wally T. Jenkins and other divers used
SCUBA gear to make some 100 dives into the Wakulla
Springs Cave. Jenkins (n.d.) prepared a type-script narrative
of these efforts entitled, "The Exploration of Wakulla Springs
Cave." Stanley Olsen, a vertebrate paleontologist then with
the Florida Bureau of Geology, identified the animal bones
collected from the spring and briefly described the project
findings (Olsen 1958).
Citing Olsen's (1958) brief Wakulla Springs report to

prove that Pleistocene fauna remains and artifacts found in the
Silver Springs head spring cavern were not an anomaly,
Wilfred T. Neill (1964:21) wrote:

At Wakulla Springs, enormous tusks, bones of
mastodon, ground sloth and deer, charred wood, and
over 600 bone projectile points, were found about
300 feet back in a horizontal cavern from which water
now flows (Olsen, op. cit.). The remains and arti-
facts were concentrated in a small area at depths of
200 to 220 feet.

More recently, Frank Rupert (1991:11) identified "mammoth,
tapir, sloth, giant armadillo, camel, horse and bison" as
coming from the Wakulla Springs fossil mammal bone bed
inside the cave mouth.
Neill (1964:19) wrote that "Suwannee points have ... been
found either in or around Wakulla Springs, Wakulla County."
One such projectile point, a Simpson, used to be on display
(Figure 2) at the Wakulla Springs Lodge. The point was not
included with material conveyed to the State of Florida when
the lodge and surrounding property were acquired for conser-
vation and recreation purposes.3 The distinctive, lanceolate-

Figure 2. Photograph of Simpson point and mastodon jaw on display at Wakulla Springs Lodge in 1970. (Photograph
reproduced with permission of Dan F. Morse).


2000 VOL. 53(2-3)


Figure 3. Map showing site distribution in Edward Ball Wakulla Springs State Park (from Bryne 1988; reproduced with
the permission of the Florida Division of Historical Resources, Bureau of Archaeological Research, Tallahassee).

shaped Suwannee and Simpson projectile points are diagnos-
tic of Florida's Paleoindian culture (see Bullen 1975:55-56;
Daniel and Wisenbaker 1987:44-54, 147-151; Daniel et al.

Probably based on Neill's (1964) report, James S. Dunbar
(1991:208-209 [Table 2]) identified Wakulla Springs as a
Clovis/Suwannee site "located in or near open karst features
(sinks, spring caves, etc.)" and situated in "coastal lowlands
with occasional karst features." Thus, it was surprising that
Dunbar (1991:208-209) described the area around Wakulla
Springs as a "marginal region", rather than as a Tertiary karst
region. The area contains much surface-exposed limerock,
sinks, and fissures, as well as Wakulla Springs and other
springs. Indeed, it is located near the middle of the Woodville
Karst Plain.
While the above finds are of note, little formal professional
archaeological work took place on the property until its
acquisition by the State of Florida in October, 1986, with
Conservation and Recreation Lands (CARL) trust funds.
Thereafter, under the direction of the Florida Division of
Historical Resources, Bureau of Archaeological Research
(BAR) Stephen Bryne assisted by Thomas Kempton con-
ducted limited archaeological field investigations within the
1157-hectare (2860-acre) tract. That survey along fixed

interval transects resulted in the recording of "54 archaeolog-
ical or historical sites and 14 artifact occurrences" (Bryne
1988:1; see Figure 3), a representative sample of what a more
comprehensive investigation could record.
Bryne (1988:50, 89) recommended that the earlier
recorded site number, 8WA24, beginning with his survey, be
restricted to the submerged main spring site while a new site
number, 8WA329, be used for the terrestrial portion of the
site. Site 8WA329 occupies the sandy ridge crest upon which
the Wakulla Springs Lodge is located and extends down slope
to the spring and spring run edge. The Florida Master Site
File made the recommended change in site designations.
Since 1988, the BAR has continued to provide archaeolog-
ical resource management assistance at the Edward Ball
Wakulla Springs State Park to the Florida Department of
Environmental Protection (formerly Department of Natural
Resources), Division of Recreation and Parks (DEP Parks).
During 1990, Jones conducted archaeological monitoring
activities around the Wakulla Springs Lodge for several
small-scale projects (Jones 1990a, 1990b, 1990c). Below
shallower, more recent (post-A.D. 800) material, these
monitoring projects produced Middle and Late Archaic
diagnostic stone artifacts made from grainy, macro-
fossiliferous, Suwannee limestone formation chert. Deeper

0 1 Mile

0 1 Kilometer
Contour Interval 5 Feet
All Site Numbers to be Prefixed with
'8W'" for Complete Number
Numbers Preceded by "A' Indicate Artifact
Finds, Not Site Numbers
--Survey Boundary



ATER: U.6.0.11. tG' Quadronfli of Criforordvlle East, Fla. (1972)


Figure 4. Auger hole excavation at the Wakulla Springs Lodge, Wakulla County, Florida.

exposures produced good quality (almost glass-like), St.
Marks Formation chert debitage (stone tool manufacture
waste material).
Because Paleoindian and Early Archaic artifacts found in
Northwest Florida are frequently made of such good quality
stone and because of previous discoveries in the nearby
spring, the presence of the St. Marks Formation chert artifacts
suggested that Paleoindian to Early Archaic materials dating
some 9500-12,000 years old were likely to be encountered in
the deeper levels of site 8WA329. For that reason, beginning
with the sewer line project in late 1994, the BAR and DEP
Parks agreed to cooperatively investigate the project impact
area in advance of construction activities.
Jones was assigned by the BAR to supervise the archaeo-
logical aspects of the project. The project required that an
approximately 125-meter-long, one-meter-wide trench be
excavated around the lodge, where the new sewer line was to
be laid to a depth of over two meters. The line also connected
to the bath house facilities and extended to the property's
southeast boundary where a drain-field was to be constructed.
The project, a major health and safety concern, had to be
completed in order for the park to continue to function.


Field investigations were carried out in two phases. The
first phase was designed to sample areas of site 8WA329 that
would be affected by project construction to determine which
portions of the project could proceed with no effect on
archaeological resources. During the second phase of the
project it was planned to excavate 1 m2 units to systematically
sample those portions of the sewer line trench in which
concentrations of potentially significant archaeological
remains were identified.
While considered possible, it was not expected that the
Phase II results would lead to an extensive mitigation effort.
However, the discovery of the stratified Paleoindian-Early
Archaic site component, extending continuously along more
than 35 meters of sewer line trench, necessitated an increased
data recovery effort to mitigate project impacts to that signifi-
cant site component.
During the first phase, the sewer line route around the
lodge and to the bathhouse was systematically sampled with
a power auger at fixed intervals (Figure 4). The depth of


2000 VOL. 53(2-3)


V "

Figure 5. View from second floor of Wakulla Springs Lodge site (8WA329) excavation near the northeast corner of the

some units was extended with long-handle post-hole diggers.
Fifty-five auger holes were excavated to depths ranging from
180 cm to 240 cm. These 45 cm diameter tests were dug in
60 cm increments. The auger hole soil was screened through
0.635 cm (1/4 in) hardware cloth to determine the presence of
cultural remains. Few holes generated artifacts below 180 cm.
The auger testing corroborated the findings of earlier
monitoring activities in that area. Middle Archaic and Late
Archaic artifacts, and small quantities of Weeden Island, Fort
Walton, and Seminole Brushed pottery, as well as more recent
twentieth-century historic artifacts, were found in the upper
60 cm. Early Archaic artifacts recovered from the 60-120 cm
level included a Kirk Serrated projectile point/knife (see
Bullen 1975:37), a Bolen projectile point/knife (see Bullen
1975:51-52), an unifacially retouched flake scraper and other
tools, and lithic debitage. The occurrence in the deeper site
levels of non-diagnostic artifacts made of good quality chert
reinforced Jones's belief that identifiable Paleoindian artifacts
also might be found. That information was used to identify
specific portions of the project area warranting more struc-
tured archaeological excavation.
While sewer line construction activities proceeded in an

area south of the lodge where there were no identified
significant archaeological remains, the line route around the
lodge was divided into eleven trench segments, designated
TS01 through TS11. The excavation strategy was to sample
the trench with several 1 m2 units. These units (from five to
eight per trench segment) were to be excavated in arbitrary 15
cm levels (Figure 5), except for features which were to be dug
in natural levels. Excavation of the 1 m2 units within the
planned sewer line trench began on the south, east and west
sides of the lodge to satisfy contractor construction schedul-
Unfortunately, because of time constraints imposed by the
project contract, when project construction activities south of
the lodge neared completion, the field methods had to be
modified along the north side of the lodge-the side facing
the spring-where, as might be expected, the most significant
cultural remains were encountered. Along the north side of
the lodge, the upper 75 cm (Levels 1-5) contain the most
recent (ca. 6,000 B.P. to present) cultural materials. However,
this strata also was the most disturbed soil, particularly in the
upper 40 cm. This disturbance was the result of lodge






Figure 6. Photograph of north profile of TP52 and TP53 at the Wakulla Springs Lodge site (8WA329).

construction and associated drainage and utility installation,
as well as recent (post-A.D. 800) Native American occupation
activities which disturbed underlying cultural components
through the excavation of post holes and pits.
Since delaying the construction project was not an option,
an alternative to changing the field methods would have been
to forego complete formal excavation of the stratified
Paleoindian-Early Archaic component located below the
upper 75 cm. That alternative was rejected. Instead, 36
contiguous text pits were excavated in this area after the upper
75 cm of soil had been removed as a unit without screening.
Moreover, screen size was reduced to 0.318 cm (1/8 in)
hardware cloth, a decision warranted by soil conditions, low
artifact density, and the generally small artifact size in the
deeper site soils.
Three hundred twenty-six 15 cm levels were excavated within
eighty-four 1 m2 units during the project. Less than 20% of
the excavated levels were in the upper 75 cm, while over 80%
were in levels 6-11 (75-165 cm). Of the 1178 artifacts
recovered from Levels 6-11, about 745 (63%) measure less
than 1 cm2. Of particular importance is the occurrence of
calcined bone in TS 11. Bone preservation is rare in Florida's
generally acidic soils, requiring chemical alteration through
burning, similar to the process of turning wood to charcoal.
Its recovery from 8WA329 provides an opportunity for
increasing our understanding of subsistence and other poorly
understood aspects of Paleoindian and Early Archaic cultures

in Florida.


The soil zones were marked and then photographed
(Figure 6) and sketched when the soil was freshly exposed. In
a typical stratigraphic profile around the lodge, Munsell Soil
Color Charts (1975 edition) soil colors from top to bottom are
as follows:
dark gray (7.5YR 4/0) sandy humus surface layer
brownish-yellow (10YR 6/6) sand
light brownish-gray (10YR 6/2) sand
light gray (10YR 7/2) sand
very pale brown (10YR 8/3) sand, and finally
whitish (10YR 8/1-8/2) sand overlaying a dark
reddish-brown (2.5YR 3/6) clay laminae over limestone.

A couple of thin clay laminae bands, representing former
water table stands, are also seen in the profile, but occurred
subsequent to the cultural deposits. The middle of the
transitional Paleoindian to Early Archaic horizon occurs in
the light gray sand zone. The entire soil column above the
clay-limestone zone is a very fine, powdery, sand with little
organic content and is of a type generally associated with


2000 VOL. 53(2-3)




'(6ZCVAM) .sS alpoI sguudS lInnAtM aq; ae papa axa uuoja.d la!J!iq '3)n!-uosdm!S/aauuAins avils lruy alj.l jo Sm.A aSJ3AO.J pue ap!s 'asIjAqo /L .na.l


wind deposited dune formations.
In the ridge area around the lodge these soils reflect a
transition from a well-drained, xeric habitat (similar to
Florida's coastal dunes today), represented by the older wind-
blown fine-grained sands with little organic content, to one of
increasing moisture and vegetative cover as modern times are
approached, including responses to the sea level and associated
climatic fluctuations noted from the Late Archaic to present.
In contrast to the well-drained higher ridge crest, it is expected
that analysis of the soil profile in the lower-lying area near and
within the spring run would reveal a wetter, mesic to hydric
setting owing to the moisture holding capacity of the shallower
marly clays found in that area.


For this preliminary report only the artifacts from the
formally excavated portion of the sewer line trench along the
eastern half of the north face of the lodge have been analyzed,
since the most significant cultural remains came from that
area. That portion of the trench was divided into two trench
segments: TS10 on the east and TS11 on the west. Twelve 1
m2 test pits (TP49-60) were excavated in TS10, while 24 test
pits (TP61-84) were excavated in TS11. The upper five levels
(0-75 cm below surface) were removed without screening from
both TS10 and TS11. Lithic artifacts from Levels 6-12 are all
of moderately to heavily patinated, generally cream-colored,

Trench Segment 10 (TS10)

In TS10, Levels 6 and 7 were individually excavated in four
units (TP49, TP50, TP53 and TP54), yielding a total of 37 and
63 artifacts, respectively. However, because of disturbance
caused by water from a drainage pipe that broke during the
project, the remaining material from Levels 6 and 7 were
combined in the other eight units, yielding 118 artifacts. The
218 artifacts excavated from Levels 6 and 7 of TS10 represent
61% of all excavated artifacts (n=355) from Levels 6-10 of
TS10. Level 10 was not excavated in TP51, TP58, TP59 and
TP60 and yielded no artifacts in TP49 and TP53. In TP60, pit
Feature 6 extending into Level 8 from Level 7 yielded 10 chert
wasteflakes. Of the remaining TS10 artifacts 59 (16.6%) were
excavated from Level 8,49 (13.8%) from Level 9, and 19 (5%)
from Level 10. All but nine of the 355 artifacts excavated from
Levels 6-10 of TS10 are identified as unutilized, chert waste
flakes. The nine utilized artifacts are described below.
Near the northeast corner of the lodge, Jones made a
discovery that changed the course of the project, bringing in
more volunteers, hundreds ofvisitors, and more than justifying
his earlier assertions about the site's suspected Paleoindian
component. The discovery occurred when he was cleaning the
exposed face of a backhoe cut dug to install a sewer line man-
hole system. The find is identified as a bifacial core shaped
into a final-stage Suwannee/Simpson-like preform (BAR
accession number 95.1.36). It was located immediately below
one of the concrete walkway curbs, which served as a conve-

nient line level from which to measure the artifact's depth
below surface. It was fortuitous that it was excavated in place
and not found in backhoe spoil. The artifact was recovered in
TS 10, TP49 at the interface of Levels 7 and 8 (ca. 103-105 cm
below surface).
The Suwannee/Simpson-like, bifacial preform from
8WA329 is made from a locally available, somewhat coarse
grained, Suwannee limestone chert, rather than the expected
more glass-like, St. Marks Formation chert. It measures 186
mm long and has a maximum thickness of 16 mm at its
midsection, 9 mm at its base, and 12 mm at its tip (Figure 7).
It has a maximum width of 112 mm at a distance of 72 mm
from its tip. It was formed by percussion flaking (with
detached large flakes likely used as expedient tools). Some
small secondary flake scars are observable along the blade
edges. These likely are the result of working the preform's
edge to reestablish a platform from which to detach additional
flakes in a controlled manner. There also are four locations
showing signs of edge abrasion or dulling (probably with a
grooved abrader) to prepare a striking platform for further
flaking. If these two actions are not taken, then when the
hammerstone or billet strikes the artifact edge, uneven
protrusions could misdirect the impact force and the crushing
of a thin flake edge would diminish the force of the impact
wave energy, resulting in loss of control of the intended flake
shape and direction of detachment. There is no apparent use
wear on this large bifacial artifact, although if it was used
expediently to separate a hide from a bison or other large
animal, such soft tissue would leave little evidence of use
wear. No hafting polish was detected. Thus, if the preform
was used on soft tissue it likely was hand-held, not hafted.
Bifacial cores are generally considered to have been
included in portable Paleoindian and Early Archaic tool kits
(Gramly 1990:7), having been trimmed of excess waste
material and shaped to facilitate the manufacture of tool
blanks and expedient flake tools.4 Flakes are detached from
the bifacial core to satisfy an expedient need, following which
they are discarded, unless they are deemed suitable for
modification to meet other tool needs-in which case they are
also considered to be tool blanks.
Most lithic tools found at hunting camps, where the butcher-
ing of animals takes place, or even those found at primary or
base camps, where a much wider range of activities take
place, are expedient flake tools that were discarded following
their use as sharp-edged cutting or shaving tools. However,
accumulated evidence suggests that flint knappers often
planned for multiple tool needs, modifying an artifact to meet
a sequence of uses (Bradley 1989).
As Bradley (1989) demonstrates, the knapper would begin
by detaching a series of flakes (from the core) to be used as
raw-edged knives to cut through an animal's hide and muscle.
Those not suitable for further modification were discarded
once they became dull or their need was fulfilled. Others
would be reworked for further use, including adding a serrated
edge to facilitate butchering needs or to strip meat, exposing
a greater surface area, to facilitate air drying and smoking for
preservation-like beef jerky. The next modification might


2000 VOL. 53(2-3)


include unifacial retouching (flaking along one side of a tool
edge) to create a beveled edge for use, such as, for instance, in
scraping fat from the inside surface of a hide. After serving
that purpose, the beveled edge could serve as an exaggerated
platform from which the knapper will detach flakes to begin
the process of converting the artifact into a bifacial projectile
point-perhaps to replace one broken during the hunt.
Ultimately, through the process of controlled reduction, the
knapper may shape the core itself into a needed bifacial tool,
such as a chopper or the large Suwannee/Simpson-like preform
excavated at 8WA329.


0 2cm 4cm 6cm 7cm

Florida Bureau of Archaeological Research

Figure 8. Obverse and reverse views of Wakulla Springs
Lodge site (8WA329) Paleoindian artifact: bifacial core
flake retouched as unifacial scraper.

Two more key artifacts, an expedient flake tool and a
lanceolate-shaped, bifacial projectile point, were excavated at
the Paleoindian level in TS10, TP57 and TP52, three to four
meters west of where the large bifacial preform was found.
The expedient flake tool (BAR accession number;
Figure 8) was recovered in TP57 at the interface of Levels 7
and 8, essentially at the same depth (105 cm below surface) as
the large bifacial preform. It measures 73 mm long, 41 mm
wide, and 13 mm thick. It was prepared by unifacially retouch-
ing one edge of a blade detached from a bifacial core, similar
to the one that we believe was reduced to prepare the large
biface. However, it was made from a better quality (less
grainy) local chert than that of the large biface.
The lanceolate-shaped, bifacial, projectile point (BAR
accession number; Figure 9) was excavated from

0 2cm 4cm 6cm 7cm

Plorida Bureae of Archaeological Research

Figure 9. Obverse and reverse views of Wakulla Springs
Lodge site (8WA329) Paleoindian artifact: Folsom-
like/Clovis-like, bifacial projectile point.

TP52, Level 8 from about 115 cm below ground surface.
Unfortunately, it was recovered in the screen during thin-
scraping of the unit. Thus, only the approximate coordinate
location and depth are recorded. It is made from a locally
available, good quality, St. Marks Formation chert and
measures 52 mm long, 20 mm wide, and 5.5 mm thick. It has
slight basal edge dulling in the hafting area, a slightly nippled
area on its base, and a basal thinning flake on one side.
(Basal nippled areas on Paleoindian lanceolate points gener-
ally served as platforms for the detachment of channel flute
flakes.) The knapper took advantage of a lateral flake scar on
one face to create a flute-like surface. This type of pseudo-
fluting occurs frequently (e.g., 18% at the Lindemeier site) on
Folsom points (Wilmsen and Roberts 1978:111-112) and
occasionally on Clovis points (Calvin Jones, personal observa-
tion). Overall, this specimen appears to be more Folsom-like
or Clovis-like, than Simpson-like. That is to say it lacks the
narrow, waisted base with slightly flared basal ears character-
istic of Simpson projectile points, and while its form is more
like that of Clovis points, its size and pseudo-fluting are more
like those of Folsom points. Regardless of its ultimate
identification, there is no doubt that it is a Paleoindian
lanceolate-shaped, bifacial projectile point.
Another early artifact was excavated in TP59 at the
interface of Levels 8 and 9 (about 120 cm below surface). It
is identified as an expedient, unifacially retouched scraper
(BAR accession number; Figure 10), which may
also represent the distal end of a failed early stage preform
with an exaggerated striking platform. It is made from a good
quality, St. Marks Formation chert and measures 23 mm long,
50 mm wide, and 3.5 mm thick.





N % 4

0 2cm 4cm 6cm 7cm

Plorida Bureau of Archaeological Research

Figure 10. Obverse and reverse views of Wakulla
Springs Lodge site (8WA329) Paleoindian artifact:
unifacial scraper reworked from core platform
rejuvenation flake.

A late stage Bolen preform (BAR accession number; Figure 11, bottom right) was excavated from
TP54, Level 6. It was encountered at 84 cm below surface
during the cleaning of the south profile after partial wall
collapse, resulting from soil drying. It is made from good
quality, St. Marks Formation chert, measures 53 mm long, 35
mm wide, and 10 mm thick and apparently was discarded
when one basal corer broke, perhaps when the knapper
attempted to begin corer notching.
A Bolen Beveled corner-notched projectile point/knife
(ppk) (BAR accession number; Figure 11, bottom
left) was excavated from TP55, Level 7 at a depth of 104 cm
below surface. It too was excavated during cleaning of the
south profile. It is made from St. Marks Formation chert, and
measures 37 mm long, 20 mm wide, and 5 mm thick.
Another Bolen Beveled corner-notched ppk (BAR accession
number; Figure 11, center left) was excavated
from a dark, charcoal stained lens (the bottom of a fire pit or
burned tree root) that intruded into TP52, Level 10. It is made
from good quality, St. Marks Formation chert and is fire
cracked, although no other pieces were recovered. It measures
25 mm wide across its base and is 6 mm thick; other measure-
ments were not taken because of its fragmented condition. It
is an example of how bioturbation and/or cultural disturbance
can move a younger artifact to deeper layers of a site.

2cm 4cm 6cm 7cm

i i

Figure 11. Wakulla Springs Lodge site (8WA329) Early
Archaic artifacts: (top left to bottom right) unifacial
"thumbnail" end scraper; exhausted Bolen Beveled corner-
notched projectile point/knife (ppk) recycled as drill; expedient
unifacially retouched flake knife; fire cracked Bolen Beveled
corner-notched ppk; bifacial final stage preform tip fragment;
expedient flake knife/shave; Bolen Beveled corner-notched ppk;
Bolen Beveled corner-notched ppk; and, Bolen bifacial preform.

The last TS 10 artifact mentioned here is a bifacial preform
tip fragment (BAR accession number; Figure 12,
bottom left) excavated from TP50, Level 10. It is made of a
moderate quality chert and has been roughly flaked by
percussion. Its thinness and provenience suggest it is a
Paleoindian artifact, rather than an Early Archaic artifact.



2000 VOL. 53(2-3)


Trench Segment 11 (TS11)

In addition to having twice as many excavation units as
TS10, TS11 contrasts with TS10 by having calcined bone
fragments scattered across much of its area. Furthermore,
except for TP78-84, the top six levels (0-90 cm below surface)
were removed without screening, and formal unit excavation
began with Level 7. Finally, only TP61, TP63 and TP64 were
terminated at the bottom of Level 9, while TP67-70 and TP82-
84 were excavated to Level 12, with pit Feature 7 in TP68-69
being dug to Level 13.
Excavation of TS11 yielded the following artifact occur-
rences. Fifty-five artifacts (6.7%) came from Level 6, 227
(27.6%) from Level 7, 226 (27.5%) from Level 8, 171 (20.8%)
from Level 9, 105 (12.8%) from Level 10, 35 (4.3%) from
Level 11 and 4 (0.5%) from Level 12. Of the 823 chert
artifacts, all but 18 (roughly 2%) are identified as unutilized,
tool manufacture and maintenance waste material, including
four exhausted cores.
The eighteen TS11 artifacts of note are all made from good
quality, St. Marks Formation chert. They are identified as
* Bolen Beveled corner-notched ppk (BAR accession number; Figure 11, bottom center) excavated at 73 cm below
surface during cleaning of the south profile of TP71, Level 6.
It measures 44.5 mm long, 20.55 mm wide across the base,
27.5 mm wide across the blade shoulder barbs, and 7 mm
* Bolen Beveled corner-notched ppk (BAR accession number; Figure 11, top center) excavated at 56 cm below
surface above a pit feature in TP78, Level 4. It measures 44.5
mm long, 26 mm wide across its base, and 7.5 mm thick; its
blade width was not measured as one of the blade shoulder
barbs is missing. This artifact was recycled as a drill, probably
by Middle or Late Archaic site occupants.
* Unifacial triangular-shaped "thumb-nail" scraper (BAR
accession number; Figure 11, top left) excavated
from TP79, Level 7. It measures 18 mm long, 14 mm wide,
and 4 mm thick. It likely was maintained as part of a personal
tool kit, rather than expediently made and discarded following
temporary use.
* Unifacial oval-shaped scraper (BAR accession number; Figure 12, bottom center) excavated from TP81,
Level 10. It measures 29 mm long, 21 mm wide, and 3 mm
thick. It probably was maintained as part of a personal tool kit,
rather than expediently made and discarded following its
temporary use.
* Bifacial preform tip fragment (BAR accession number; Figure 11, center) excavated in TP83, Level 8. It
measures 28 mm long, 24 mm wide, and 5 mm thick.
* Final stage bifacial preform (BAR accession number; Figure 12, center right) excavated from TP66,
Level 10. It measures 80 mm long, 57 mm wide, and 10 mm
thick. It also is broken across the base and the remaining
unbroken basal corner has unifacial retouching that appears to
be the remains of an exaggerated knapping platform prepared


j l






0 2cm 4cm 6cm 7cm

Figure 12. Wakulla Springs Lodge site (8WA329) Late
Paleoindian artifacts: (top left to bottom right) unifacially
retouched (serrated) expedient flake knife; expedient
serrated bladelette knife; broken hafted (reworked nippled
base) blade tool; hematite (red ochre)/hematitic sandstone
nodule; core platform rejuvenation flake reworked as
unifacial scraper; broken final stage,bifacial Clovis/Simpson
preform; bifacial preform tip fragment; unifacial flake
scraper; and, expedient flake shave.

to permit basal thinning.
Unifacially retouched expedient scraper/knife (BAR
accession number; Figure 13, bottom right)
excavated from TP72, Level 10. It may also be identified as
a possible early stage preform with an exaggerated striking
platform-the unifacially retouched edge. It measures 70 mm
long, 48 mm wide, and 7 mm thick.






* Expedient flake tool, a combination scraper and knife
(BAR accession number; Figure 13, bottom left),
excavated from TP83, Level 10. It measures 66 mm long and
45 mm wide.
* Three artifacts excavated from TP70, Level 11 (150-165
cm below surface). The first is a thick flake measuring 63 mm
long, 50 mm wide, and 18 mm thick, that may have served as
an unmodified expedient scraper and could have been kept as
atoolblank (BAR accession number; Figure 13, top
right). It was found during cleaning of the north wall profile.
The second is an exhausted bladelette core, measuring 35 mm
high from striking platform to base and 43 mm across (BAR
accession number; Figure 14, left). Indeed, it is
possible that a number of the small flakes or bladelettes now
identified as unutilized tool manufacture and maintenance
waste flakes may have functioned as unmodified, expedient
flake knives-the exacto blades of their day-to cut soft tissue,
which would have produced little evidence of use wear. The
third artifact is a unifacial end scraper (BAR accession number; Figure 14, right), measuring 40 mm long, 26 mm
wide and 10 mm thick.
The remaining five TS11 artifacts with evidence of
reworking and use wear are all nondescript expedient flake
tools. They are not further described in this report. Interest-
ingly, none of the TS11 tools were found in association with
any of the calcined bone.

Calcined Bone Area

No calcined bone was recovered in TS10, where the
diagnostic Paleoindian artifacts were found. Fifteen meters
west of TS10, however, in an area of TS11 centered on TP68
and TP69, small fragments of calcined bone, charred soil
lenses, and 5-20 cm diameter post mold-like features (Figure
15) were encountered. Several samples of soil with charcoal
and bone were wrapped in foil for possible radiocarbon dating.
When the calcined bone fragments and associated post
mold features were first located it appeared that they repre-
sented food remains and the remains of a possible cooking
rack; however, this interpretation of the post molds and bone
became equivocal, when Jones and Tommy Abood excavated
levels 10, 11, 12, and the bottom of what became Feature 7 in
Test Pits 68 and 69. At the interface of levels 10 and 11, about
150 cm below surface, the calcined bird, deer, and other
mammal bone fragments became concentrated in an obvious
pit feature. Foil-wrapped samples of soil with charcoal and
bone were collected for flotation and for radiocarbon dating,
then the remaining pit contents was bagged and water screened
for later analysis. Analysis of the charcoal and bone is
planned for early 1999 and the results will be included in the
final project technical report.

Project Findings and Conclusions

Excavations along the sewer line demonstrated that most
Northwest Florida Native American cultural components are
represented at the Wakulla Springs Lodge site (8WA329). In

0 2cm 4cm 6cm 7cm

iK Il I I I _I I

Figure 13. Wakulla Springs Lodge site (8WA329) Late
Paleoindian artifacts: (top left to bottom right) expedient,
unifacially retouched flake scraper/shave from 120 cm below
surface; core platform rejuvenation flake/rejected tool blank
from 160 cm below surface; expedient, unifacially retouched
flake knife/shave from 135-150 cm below surface; and,
unifacial flake scraper/bifacial point blank from 135-150 cm
below surface.

surface-to-bottom cultural sequence, Levels 1 through 4, the
first 60 cm of soil, contain twentieth-century historic material
associated with the lodge construction and operation, and
Seminole, Fort Walton, and late Weeden Island period
artifacts. Overlapping these is a Late Archaic Norwood fiber-
tempered component in Levels 3 through 5, from 30-75 cm,
with a few deeper pit features. The Middle Archaic compo-
nent, with its characteristic thermally-altered chert, stemmed
projectile point/knives, expedient tools, and debitage, occurs
primarily in Levels 4 through 6, from 45-90 cm. The transi-
tional Paleoindian-Early Archaic component, represented by
Bolen projectile point/knives and projectile point blanks, and


2000 VOL. 53(2-3)

: ~Y
Vk;i~. F'r
~c'lr~ ~y


various unifacial artifacts, occurs primarily in Levels 6 through
8,75-120 cm. The Paleoindian component, with its lanceolate-
shaped Clovis-like/Folsom-like projectile point, large
Suwannee/Simpson-like final stage, bifacial preform, and
various flake tools, occurs in Levels 8 through 12 plus, from
105-200 cm.

-. .



6cm 7cm


Figure 14. Wakulla Springs Lodge site (8WA329) Late
Paleoindian artifacts: (left) single platform microblade core
and (right) exhausted hafted unifacial end scraper or adz.
Both from 150-165 cm below surface.

As noted above for the TP78, Level 4, Bolen projectile
point/knife recycled as a drill, the excavation of post holes,
hearth pits, and storage pits provided later Middle and Late
Archaic site occupants with an opportunity to find and collect
Early Archaic and possibly Paleoindian artifacts (Daniel and
Wisenbaker 1987:37-40). These were probably collected as
curios and as a source of good quality chert for use in making
expedient tools. This resulted, in part, because access to good
quality chert may have become increasingly difficult as sources
were mined out or otherwise made inaccessible by rising sea
levels and rising interior water tables. Indeed, scavenging of
good quality stone from older quarry-related base camps
became increasingly common among later, post-Archaic
Florida cultures (Daniel and Wisenbaker 1987:39; Tesar
1994a:7, 1994b:292, 1996:277). Thus, it is not surprising that
earlier cultural materials are found intermixed with those of
later site occupations. Likewise, site reoccupation and
bioturbation activities have resulted in some younger artifacts
being moved to deeper, older levels. Overall, however, there
appears to be minimal artifact mixing in Levels 7-12.5 An
analysis of pollen in the soil samples, charred wood remains,
and calcined bone remains collected during the project will
help further our understanding of climatic conditions and
changes through time.
Although the sampling of this over 1.5 hectare site was
limited, with approximately 49 cubic meters formally excavated

during this project, it is clear that significant Paleoindian and
Early Archaic cultural remains are located within an intact
stratified context. The earliest or deepest artifacts in our
sample appear to date from late Paleoindian times. Owing to
the limited sample area, excavation was restricted to the
project pipeline trench location and not expanded to discover
more diagnostic artifacts and features, few diagnostic artifacts
were found in the 120-160 cm below surface levels. The
Paleoindian component is definitely represented in the 105-
120 cm below surface level, particularly in the area near the
northeast corner of the lodge, and, extends some 36 meters
west of where the large Suwannee/Simpson-like final stage,
bifacial preform was found.
Most of the lithic artifacts are small flakes, and surpris-
ingly there was no evidence of quarry-related lithic reduction
activities within the sewer line sample transect area, although
such activities may have occurred elsewhere within the site
area. The soils in the area of the site tested generally lack
organic content beyond charcoal lenses and calcined bone.
This, and the absence of well defined spodic horizons,
suggests that the ridge crest area around the present-day lodge
was well-drained and relatively dry, much like a dune ridge,
during Paleoindian through Early Archaic times.
Our excavation data, admittedly from a restricted portion
of the site area, suggest only short term, perhaps seasonal,
occupation and reoccupation of the site during much of the
Paleoindian and Early Archaic, as well as a relatively dry
surrounding upland ridge area. Camping on the then drier
relic dune ridge would not have interfered with (frightened
off) animals seeking access to the water hole in the cavern
mouth. Prehistoric hunters could have laid in wait for the
animals on the crest of the overhanging ledge or along the
trail coming up the then relic (now again active) stream
drainage. In the latter instance, animals could have been
herded into the spring basin pond, making them an easier
An interpretive problem results from the artifacts, burnt
wood concentrations, and animal remains reported by Jenkins
(n.d.) and others from within the spring cave. Their observa-
tions suggest that people entered a then open cave mouth,
perhaps to obtain water and hunt trapped animals. However,
Michael J. Wisenbaker (personal communication, 1998)
points out that the cave entrance proper is about 40 m (130 ft)
deep and most likely was never dry during human history. He
suggests that the artifacts and bones found deeper in the cave
by Jenkins and others more likely migrated, down slope into
the cave when they were lost or discarded in the spring's sink
hole pond.
We agree with Wisenbaker and suggest that at the end of
the last glacial interval, sea levels rose and rainfall increased
with the result that the cave was gradually inundated forming
a small pond basin at its mouth. It provided an oasis water
hole for use by animals and the first people to arrive in the
area. The river channel and its borders likely became more
wooded, providing increased habitat for white-tailed deer,
raccoons, opossums, squirrels, and other small mammals. By
this time most of the large Pleistocene megafauna were





77.1.l S

Figure 15. Calcined bone concentration area of the Wakul
(8WA329). Twigs mark locations where bone found. Dark area
post-mold stain.

extinct. Migratory birds and reptiles also were attracted to the
spring pond, which probably overflowed during periods of
heavy rain in the spring recharge areas and receded during
times of extended drought. It was a time when mixed pines
and hardwoods spread across the landscape; a time when
lightning-producing spring and summer thunderstorms and
lightning strike-fires shaped vegetative succession in drier
uplands (Fredlund and Johnson 1992:120).

The habitat shift to a more
S-- wooded setting required a change
in subsistence patterns, including
the manner in which game was
hunted. This would have been
particularly true with respect to
hunting deer, which prefer a more
wooded setting than the earlier
grazing buffalo and slower, larger
animals. The technological
changes to accomplish this in-
cluded the development of the
atlatl, or spear-thrower. Replica-
tion studies have demonstrated
that the spear-thrower served as
an extension of the arm, allowing
more force and distance to be
achieved, than possible with hand-
thrown spears.
The use of bone points, ap-
pears to have become increasingly
popular, based on the numbers
that have been found. These gen-
erally were manufactured from
deer cannon bones. It is generally
suggested that the longer, bi-
pointed bone points are spear tips,
while the shorter ones with bev-
eled ends are generally identified
as foreshafts. There are techno-
logical reasons why the foreshaft
interpretation for these bone arti-
facts, in contrast to the earlier
thicker, similar-shaped ones of
S ivory, appears to us to be flawed.
A foreshaft with a tapered, point-
ed basal end would have become
wedged in the spear or dart shaft.
7 ..While this would have been a
useful attribute in the hand-thrust
spear shaft, it would be a design
S. flaw in dart shafts leading to fre-
; "' quent breakage of the foreshaft
Sand possibly the dart shaft as the
lla Springs Lodge site animal brushed past obstructions
at bottom of photo is a or rolled on the shaft. The wedge-
shaped foreshaft in the dart shaft
would have made replacement of
a broken one difficult. Thus, we
believe that the shift from spears to darts resulted in modifica-
tion of the basal end of the foreshaft from the earlier wedge-
shape to a tenoned one.
While many of the bi-pointed bone points probably were
used as individual spear tips, we suggest that most of the bone
points found at Wakulla Springs are the remains ofleisters-a
group of two to four short bone points with beveled bases
affixed to a long central point to make a compound, fish

2000 VOL. 53(2-3)


spear. Indeed, it is likely that the majority of the bone points
date to Late Archaic and early Woodland times, when Wakulla
River was flowing and fish would have been plentiful.
While we note that north Florida did not experience the
same temperature extremes as in the Arctic, freezing tempera-
tures are an annual occurrence. Lewis Binford's (1979:262-
263) discussion of the seasonal circumstances pertaining to the
use of bone versus stone points among the Eskimo, suggests to
us that similar reasons should be considered for their occur-
rence and use among Paleoindian and Early Archaic peoples,
including those who occupied the Wakulla Springs area.
The earlier thinner, lanceolate-shaped, bifacial projectile
points were gradually replaced by side-notched and corner-
notched points, such as the Early Archaic Bolen points in
Florida. The Bolen points likely were hafted to a foreshaft that
socketed at a tenon-shaped end into the spear or dart shaft
(somewhat like a harpoon point), so that the dart shaft was
thrown free when the target was hit. This prevented the dart
shaft from breaking, as can happen when a large animal falls
over on a spear shaft or, when deeply wounded, fractures the
shaft against a tree while trying to flee. It also allowed for the
reuse of the shaft with the rapid replacement of used points by
extra ones carried for that purpose. Moreover, the foreshaft
also served as a handle when the point was used as a knife
(Goodyear 1974:33).
Another technological innovation developed during the
transitional Paleoindian-Early Archaic was the unifacial
resharpening of opposite edges of hafted bifaces, such as
Daltons and Bolens, which created an increasingly beveled,
serrated cutting edge, while at the same time conserving the
amount of material removed in the resharpening process
(Goodyear 1974:25-33; Morse 1973:25-26). Unifacial
resharpening removes only about half the amount of material
as bifacial resharpening, although the latter technique contin-
ued in use. A serrated knife edge also could have been used to
strip meat fillets, creating a greater exposed surface area, thus,
allowing them to dry more rapidly. The dry meat jerky was
preserved and weighed much less then fresh meat, making it
easier to transport (Goodyear 1982:391).
One of the anonymous reviewers notes, "I believe (that)
many Bolens were used as knives. The side notching made for
a firm attachment that could withstand torque and pressure
when cutting, twisting, etc. Bone points could have been used
as projectiles. The resharpening patterns in Bolens are not
those that would be associated with a piercing implement, but
more likely to have been used on cutting/scraping tools." We
concur with this observation, but continue to believe that Bolen
points and many later Middle Archaic points served as both
dart points and knives.
Bolen points and associated artifacts generally were found
from 53 cm to 105 cm below ground surface, as well as in
deeper pit features, within the sewer line trench around the
lodge, in subsequent testing across the ridge, and around the
bathhouse to the East. These help demonstrate the extent of
the site area.
Overall, the Paleoindian and Early Archaic stone artifacts
recovered at site 8WA329 indicate activities associated with

hunting and butchering of animals, and working animal
hides, wood, and bone. The many bone points recovered in
the nearby spring waters confirm the latter activity. However,
the lack of grooved abraders, shaves, and burins in our
excavation sample suggests that the manufacture of bone
points occurred in another area of the site, perhaps down slope
closer to the spring run in the area of the present day boat
dock. The low artifact density and the small amount of
accumulated organic material in associated soils suggest
limited periodic, perhaps seasonal, occupation of the site area
around the lodge during Paleoindian and subsequent Early
Archaic times. It is speculated that the main habitation
portion of the site was oriented along the higher ridge crest
between the lodge and bathhouse down slope to where the
boat docks are now located, as that is the most likely route
from the upland ridge to the then spring pond at the cave
mouth. Likewise, it is in the present-day shallow waters in
that area, in addition to those from the spring cave, that many
of the Pleistocene faunal remains and bone points recovered
from the spring waters are reported to have been found.
The archaeological salvage excavation project demon-
strated that an intact Paleoindian through Early Archaic
cultural zone is present at site 8WA329, and that further
research at the submerged site 8WA24A likely would yield
additional significant cultural material. The two site areas
complement each other. Further research at and interpreta-
tion of sites 8WA24A and 8WA329 will add another dimen-
sion to the public interpretation of the property surrounding
Wakulla Springs.


' December 28, 1998 version.
2 B. Calvin Jones died on February 15, 1998. However, he was the
senior author in the preparation of the draft version of this article
prior to his death and appropriately retains that designation.
3 We are fortunate that Dan Morse's father took a photograph of the
original display with the point and that 25 years later Dan let us
make a copy of that slide.
4 A stone "tool blank" generally is a flake that has a size and shape
suitable for reduction to a desired form through the process of
unifacial or bifacial detachment of flakes. An "expedient" tool is
one made for use in a particular situation, after which it is generally
discarded. In contrast, a "curated" tool is one made specifically for
inclusion in personal gear to meet routine recurring needs. It
generally is resharpened and repaired to meet that purpose until it
becomes "exhausted," or reduced to a size too small or with edge
angles too steep to meet its intended function.
5 As a word of caution, Michael Wisenbaker (personal communica-
tion, 1998) notes that when Randy Daniel and he attempted to
separate Bolens from true Paleoindian points at the Harney Flats
site, they found no statistical difference in their depths below surface
and concluded that apparently there was little deposition of sedi-
ments occurring at the site during that transitional time. In general,
Paleoindian lanceolate-shaped projectile points tend to have thin,
tapered basal hafting areas, while Bolens generally have thicker,
wedge-shaped basal hafting areas. This is a technological adjust-
ment, which required a thickening of points to better withstand the
increased impact stress resulting from use of the atlatl or spear-
thrower or use as a knife. This trait was continued in later Archaic




projectile point/knives.


We wish to thank Christine Newman and Donna Ruhl for
encouraging us to write this article, expanding upon an earlier
presentation that we had given at the 1997 Southeastern Archaeo-
logical Conference. We also wish to thank Michael Wisenbaker,
Mary Glowacki, Dana Ste. Claire, and Michael Faught for review-
ing an earlier draft of this article. Likewise, we extend out thanks
to Donna Ruhl, Christine Newman, and the anonymous reviewers
who critiqued our final draft. While we have benefited from their
comments, any problems and errors contained in the document are
our own, as are the opinions that we have expressed. With the
exception of Figure 2, all photographs are reproduced courtesy of
the Florida Department of State, Division of Historical Resources,
Bureau of Archaeological Research, Tallahassee. All artifact
photographs were taken by Roy Lett. Finally, we wish to thank the
Florida Park Service staff and many volunteers who assisted in the
project excavation and, thereby, made the writing of this article

References Cited

Binford, Lewis R.
1979 Organization and Formation Processes: Looking at
Curated Technologies. Journal ofAnthropological Re-
search 35(3):255-273.
Bradley, Bruce
1989 Flintknapping. VHS Videotape produced by INTERpark
and Primitive Tech. Enterprises, Cortez, Colorado.
Bryne, Stephen C.
1988 Archaeological Survey at the Edward Ball Wakulla
Springs State Park. Florida Archaeological Reports 6,
Florida Department of State, Division of Historical Re-
sources, Bureau of Archaeological Research, Tallahassee.
Bullen, Ripley P.
1975 A Guide to the Identification ofFlorida Projectile Points
(Revised Edition). Kendall Books, Gainesville, Florida.
Daniel, I. Randolph, Jr. and Michael Wisenbaker
1987 Harney Flats: A Florida Paleo-Indian Site. Baywood
Publishing Company, Inc. Farmingdale, New York.
Daniel, I. Randolph, Jr., Michael Wisenbaker and George Ballo
1986 The Organization of a Suwannee Technology: The View
from Hamey Flats. The Florida Anthropologist 39(1-
Dunbar, James S.
1991 Resource Orientation of Clovis and Suwannee Age
Paleoindian Sites in Florida. In Clovis: Origins and
Adaptations, edited by R. Bonichsen and K. Turnmier,
pp. 185-213. Corvallis: Center for the First Americans,
Oregon State University.
Fredlund, Glen and William C. Johnson
1992 Paleoenvironmental Reconstruction. Draft Eglin Air
Force Base Historic Preservation Plan, Technical Syn-
thesis of Cultural Resources Investigations at Eglin,
Santa Rosa, Okaloosa and Walton Counties, edited by
Prentice M. Thomas, Jr. And L. Janice Campbell, pp. 87-
122. Report of Investigations 192, New World Research,
Inc., Fort Walton Beach, Florida.
Goodyear, Albert C.
1974 The Brand Site: A Techno-Functional Study ofa Dalton
Site in Northeast Arkansas. Research Series No. 7, Ar-

kansas Archeological Survey, University of Arkansas
Museum, Fayetteville.
1982 The Chronological Position of the Dalton Horizon in the
Southeastern United States. American Antiquity
Gramly, Richard Michael
1990 Guide to the Paleo-Indian Artifacts of North America.
Persimmon Press Monographs in Archaeology, Buffalo,
New York.
Jenkins, Wally T.
n.d. The Exploration of Wakulla Springs Cave. Typescript
manuscript report on file, Florida Division of Historical
Resources, Bureau of Archaeological Research, Talla-
Jones, B. Calvin
1990a Archaeological Resource Management Activities in
Vicinity of Lodge at Edward Ball Wakulla Springs State
Park, Wakulla County, Florida. Report on file, Florida
Division of Historical Resources, Bureau of Archaeolog-
ical Research, Tallahassee.
1990b Post Hole Testing within Area of Proposed Greenhouse
at Wakulla Springs State Park. Report on file, Bureau of
Archaeological Research, Tallahassee, Florida.
1990c Monitoring of Facility Improvements in the Vicinity of
the Lodge at Edward Ball Wakulla Springs State Park,
Wakulla County, Florida. Report on file, Florida Divi-
sion of Historical Resources, Bureau of Archaeological
Research, Tallahassee.
Jones, B. Calvin, Louis D. Tesar and Jonathan Lammers
1998 B. Calvin Jones: Comments and Commentary, Video-
taped Interview Excerpts. The Florida Anthropologist
Morse, Dan F.
1973 Dalton Culture in Northeastern Arkansas. The Florida
Anthropologist 26(1):23-38.
Munsell Color
1973 Munsell Soil Color Charts. Macbeth Division of
Kollmorgen Corporation, Baltimore, Maryland.
Neill, Wilfred T.
1964 The Association of Suwannee Points and Extinct Ani-
mals in Florida. The Florida Anthropologist 17(1):17-
Olsen, Stanley J.
1958 The Wakulla Cave: Amateur Paleontologists Investigate
an Underwater Site in Florida. Natural History
Rupert, Frank R.
1991 The Wakulla Springs Mastodon. Florida
Paleontological Society Newsletter 8(2):10-12.
Tesar, Louis D.
1994a Johnson Sand Pit (8LE73): An Analysis and Compara-
tive Review of a Paleoindian through Early Deptford
Base Camp in Leon County, Florida. Florida Archaeo-
logical Reports 32, Florida Department of State, Divi-
sion of Historical Resources, Bureau of Archaeological
1994b What Do You Call This Point? The Florida Anthropolo-
gist 47(3):287-294.
1996 Block-Sters Site Middle School Project Artifacts. In
Emergency Archaeological Salvage Excavation within
the Swift Creek Subarea of the Block-Sterns Site
(8LE148), Leon County, Florida: A Public Archaeology
Project by B. Calvin Jones and Louis D. Tesar, pp. 239-


2000 VOL. 53(2-3)


538. Florida Department of State, Division of Historical
Resources, Bureau of Archaeological Research.
Wilmsen, Edward N., and Frank H. H. Roberts, Jr.
1978 Lindenmeier, 1934-1974: Concluding Report on Investi-
gations. Smithsonian Contributions to Anthropology
Number 24, Washington, D.C.

A new video on florida's native peoples

"Shadows and Reflections:
Florida's Lost People"
Produced by the Florida
SFunded by the
S I lorida Department
r of State

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 $23.45 ( $18.81, plus $1.14 tax and $3.50 S&H) to:
Terry Simpson, 7751 Avocet Drive, Wesley Chapel, FL 33544

o- 0 ^ ",
L/ '** < -r.^
*" *L,
ijfci~a g. .



1Southeastern Archaeological Research, Inc., P.O. Box 2818, Riverview, Florida 33568-2818
E-mail: bob@searchinc.com
2 Panamerican Consultants, Inc.-Tampa, 1207 North Himes Avenue, Suite 5, Tampa, Florida 33607
E-mail: estabrr@worldnet.att.net

The "quarry-cluster" method for assigning chert artifacts
to general source areas has been known but underutilized by
archaeologists in Florida for nearly two decades. In this paper
we describe the method and present analysis results that
illustrate it's usefulness in an archaeological context. Our data
set consists of lithic artifacts from 33 excavated or tested sites
in central and south Florida. While many of these sites lie
outside the state's major chert-bearing regions, we have
included in our sample some sites from chert-rich areas for
comparison. The site data have been partitioned according to
two broad temporal periods-the Archaic and post-
Archaic-resulting in a total of 34 assemblages.' This has
allowed us to examine temporal as well as spatial variation in
the use of different raw materials. We are particularly
interested in the patterns of chertprocurement and distribution
that characterize these chert-poor areas, and whether these
changed through time (e.g., Austin 1996, 1997; Estabrook
1992). The results indicate clear differences between the
patterns of chert procurement during the Archaic and post-
Archaic periods.

Chert Quarry Clusters
and Identification of Source Locations

The identification of lithic raw-material provenance was
achieved using the "quarry-cluster" method developed by
geologist Sam Upchurch and his colleagues at the University
of South Florida (Upchurch et al. 1981). "Quarry cluster"
refers to a geographic region where chert outcrops originate
from a single geological formation and share similar diagnos-
tic features. Because most Florida cherts were formed by the
replacement of limestone with dissolved silicates, these
replacement rocks often retain the content of the parent
limestone in much the same way that wood grain is preserved
in petrified wood. Consequently, the assignment of individual
artifacts to a particular quarry cluster can be accomplished
with a high degree of accuracy using physical properties
familiar to field geologists. These include rock fabric, fossil
content, and the presence or absence of secondary inclusions
such as quartz sand, phosphate pellets, and calcite crystals. In
some cases, identification can be made with the unaided eye.
In most cases, however, the diagnostic features must be viewed
and identified with the aid of a binocular microscope. The
availability of a large and reasonably representative type
collection of geologically derived chert samples from more
than 200 source locations representing all the identified quarry

clusters in the state increases the confidence of our assign-
Quarry-cluster locations are shown in Figure 1. Note that
all of the known exposures of chert are restricted to central
Florida and the panhandle. Although Upchurch et al. identi-
fied 19 quarry clusters in Florida, Austin (1997) recently

S- Wrghtl Creek
2 Maianna
3 Wacis.
4 Upper Suwannee
5 Alapwn River
6 Swift C-r k Sweap
7 Whit Spring.
8 Lower Suwnne /Lake Pansoffkee
9- Stntl Fe
10- Ocall
11 Brockville
12- Upper WthlNccoche
13 Caladu
14- Hillbborough River
I Turntlecwl Point
16- P .ce River

0 50 milomes
0 50 kilometers

Figure 1. Map of chert quarry clusters in Florida (from Austin
1997:Figure 19).

combined several ofthese into mega-clusters reducing the total
number to 16. This was necessary because of the lack of
unambiguous criteria for assigning cherts derived from silica
replacement of the Ocala Limestone to specific source areas.
In this paper, we focus our attention on cherts derived from the
three quarry clusters that occur with the greatest frequencies
in lithic assemblages in central and south Florida: the
Hillsborough River, Upper Withlacoochee, and Peace River.3
Table 1 provides summary descriptions of the cherts from
these quarry clusters.
The Hillsborough River Quarry Cluster extends along the
Hillsborough River and its many tributaries west to Tampa
Bay. It contains many well-known quarry locations (e.g.
Chance 1982; Chance and Misner 1984; Palmer et al. 1982;


VOL. 53(2-3)




Table 1. Diagnostic criteria of raw-material types discussed in the text.

Geological Chert Types anduy
Formation Diagnostic CriteriaQuary users

Peace River

Peace River

Hillsborough River

Quartz replacement of opaline sediments. Dark gray,
brown, or yellowish-brown in color with abundant
phosphate pellets and quartz sand inclusions. Very
fine-grained and brittle. Opal content gives this chert
a natural luster. Corticates easily.

Translucent chalcedonic chert Gray, brown, or tan-
colored with light-colored pelletal material in a
wackestone fabric." Quartz sand and occasional fossil
fragments are sometimes present

Fossiliferous packstone or grainstone. Fossil
Peneroplids are abundant, quartz sand may or may not
be present Opaque, relatively coarse-textured
material. Gray, brown, or tan in color.

Bay Bottom chert. A pelletal packstone, occasionally
brecciated. Opal clasts and large areas of chal-
cedenous void filling are common giving the chert a
clotted or splotchy appearance. Peneroplids,
particularly Sorites, are common fossil inclusions.
Quartz sand is present in varying amounts ranging
from rare to abundant. Color is variable and can be
white, brown, light gray to dark gray.

Cow House Creek chert. Silicified mudstone.
Identified by the presence of charophyte oogonia.

"Generic" quartz replacement of mudstones and
wackestones. Color and fossil content are variable.
Peneroplids, pelecypods, occasional small gastropods.
Voids and burrows filled with chalcedony and quartz

Dark gray to black chert; small fossil fragments,
common sand.

Fossiliferous grainstone. Color is variable, but tends
toward browns and grays. Quartz sand may be pre-
sent, but is not common. Miliolids are diagnostic.
Echinoid molds and Dictyconus molds also are com-

Upper Withlacoochee

NOTE: Silicified coral is common throughout the Tampa and Suwannee Limestones, but it has not been included in this
analysis because of the difficulty in assigning individual specimens to a specific quarry cluster.

'Terms and definitions ofrockfabric follow Dunham (1962). Generally, this classification is one of increasing content of the constituent sedimentary components, such
as fossils, that compose the host limestone. Mudstones contain less than 10% of these components suspended in clay or silty mud followed by, in order of increasing
component proportion, wackestones, packstones, and grainstones. A fifth category, boundstones, refers to rocks where the original components have been completely
bound together during deposition. The most common boundstone in Florida is coral.

Simpson 1939, 1941; Upchurch 1980). Cherts from this cluster
originate from the early Miocene-age Tampa Limestone
Member of the Arcadia Formation. They vary widely in color,
fabric, and fossil content. Peneroplid foraminifera are diag-
nostic, particularly the genera Peneroplis, Archaias,
Stensionia, and Sorites (Figure 2a-d). Some cherts also
contain quartz sand inclusions, unidentified pelletal material,
quartz-lined or chalcedony-filled voids, and charophyte
oogonia. The last is the fossilized reproductive apparatus of a

silica-secreting plant (Figure 2e).
The Upper Withlacoochee Quarry Cluster is located in the
Green Swamp region of central Florida. Cherts that originate
from this cluster formed as silica replacements of Oligocene-
age Suwannee Limestone. They vary in color from brown to
dark gray and become pinkish-red when thermally altered.
Diagnostic fossils include Miliolid foraminifera (Figure 2f),
often seen as small, round discolorations in a grainstone
fabric. Dictyoconous (Figure 2g) is another common fossil

Hillsborough River. Very common
around Tampa Bay and along the
lower Hillsborough and Alafia ri-

Hillsborough River, Tampa Bay

Hillsborough River. Restricted to
Cow House Creek, Flint Creek, and
Harey Flats region of upper Hills-
borough River.

Hillsborough River

Peace River. Appears limited to
area between Ft. Meade and

Tampa Limestone
Arcadia Formation

Suwannee Limestone


2000 VOL. 53(2-3)


that typically appears as a V-shaped or U-shaped void. Other
diagnostic features include the presence of Echinoid casts,
calcite crystals, and occasional quartz sand inclusions. The
Upper Withlacoochee Quarry Cluster contains many aboriginal
quarry sources (e.g., Johnson 1986; Piper and Piper 1980;
Wharton 1984), many of which contain silicified coral. A
portion of this cluster near Wesley Chapel is perhaps the most
significant source of silicified coral in peninsular Florida.
Recorded chert outcrops associated with the Peace River

~ I,,, I

Figure 2. Diagnostic fossils used in the identification of che
Peneroplis sp., Family Peneroplidae; b) Archaias sp., Family
Stensionia sp., Family Peneroplidae; d) Sorites sp, Family Pene
cross-section views; e) charophyte oogonia, plan and cross
Nummoloculina sp., Family Miliolidae; g) Dictyconus sp., Fa
Redrawn from the following sources: Cole (1945); Cushman (194
Upchurch et al. (1981).

Quarry Cluster are limited primarily to the channel of the
Peace River in south-central Florida, although small outcrops
are known or reported in Horse Creek. This chert, which is
derived from the late Miocene-age Peace River Formation, is
distinctive in that it contains phosphate pellets and is opal-
rich, which gives the stone a waxy luster not unlike that
produced by thermal alteration. Figure 3 shows a chert sample
containing large phosphate pellets. Typically these pellets are
much smaller and can best be seen under magnification.
Quartz sand also is very common in
these cherts.
We acknowledge that the visual
identification of lithic raw materials is
controversial, and we are aware that a
battery of instrumental techniques is
available for identifying the sources of
siliceous lithic materials (see, for exam-
ple, the reviews in Luedtke 1992 and
Rapp 1985). Nevertheless, an extensive
C review of the available literature (Aus-
tin 1997) indicates that the success rates
of these studies are highly dependent on
the geographic and depositional unique-
ness of the source materials. When the
depositional environments of the source
materials are significantly different,
geochemical signatures often provide
highly satisfactory results. However,
for cherts that are located within a re-
stricted geographic range, such as the
Coastal Plain of the lower Southeast,
including Florida, differences in the
origins of deposition of the original host
I limestones are not great enough to
produce unambiguous geochemical
signatures (Upchurch et al. 1981:162-
e 163). Under these conditions, visual
identification provides results with
success rates as high or higher than the
use of instrumental techniques (Goad
1979, 1984; Luedtke 1979; Upchurch et
al. 1981).
Another important consideration in
the choice of methods for identifying
chert provenance is the time and cost
associated with the use of instrumental
techniques, particularly when used with
more than just a few artifacts. To un-
derstand patterns of chert procurement,
rt provenance: a) it is necessary to examine and identify
Peneroplidae; c) the source locations of a large and rep-
roplidae, plan and resentative sample of artifacts, includ-
-section views; ) ing debitage. Consequently, we have
mily Valvulinidae. concluded that the use of visually ob-
18); Taylor (1981); servable physical properties is the most
efficient and productive method for
determining raw-material provenance.





Ila rr~

r. 14 .r;

r.r a

I) '

Figure 3. Opal-rich chert from the Peace River Quarry Cluster. Note the abundant inclusions. Arrow points to a large
phosphate pellet. Photograph by Richard Estabrook.

Archaeological Examples

To illustrate the usefulness of the quarry-cluster method,
raw-material data from 34 lithic assemblages (18 Archaic, 16
post-Archaic) recovered from 33 archaeological sites were
examined for spatial and temporal trends. Table 2 provides
these data for each of the 34 assemblages used in this study
and Figure 4 shows the geographic locations of the 33 sites
from which these assemblages were derived. Note that many
of these sites are to the east and south of the major chert
exposures (compare with Figure 1). Thus, the total number of
lithic artifacts from sites in Osceola, Highlands, parts of
Hardee, Glades, and Lee counties is often small in comparison
to sites located in chert-rich areas, such as Hillsborough and
Pasco counties.

Spatial Distribution

Figure 5 shows the spatial distribution of artifacts made
from cherts derived from the Hillsborough River Quarry
Cluster expressed as a percentage of each assemblage. The
contour intervals are 5%. The 0% line also is included in
order to show the maximum geographic range of these chert
Both the Archaic and post-Archaic data exhibit fairly wide
geographic distributions for these cherts, with the major trend
to the south and east of source locations in the Hillsborough

River Basin. Figures 5a and 5b both show decreases in the
abundance of these cherts in the immediate vicinity of the
Peace River where the principal exposures of Peace River
Quarry Cluster cherts are located. However, at present we
have no data from Archaic sites in the area between Horse
Creek and the Lake Wales Ridge. Thus, the large trough
shown for Archaic components in the vicinity of the Peace
River may under-represent the amount of Hillsborough River
cherts at Archaic sites there, particularly if the three Archaic
sites near Horse Creek are any indication of the use of
Hillsborough River cherts in this part of the state (see Table 2).
Similarly, the gradual drop-off of Hillsborough River
cherts at Archaic-period components to the south of the
Hillsborough River Quarry Cluster (Figure 5a) is somewhat
misleading because there are no Archaic sites in our data set
for south Florida. Since the Hillsborough River Quarry
Cluster would have represented the closest source of high-
quality cherts for people living along the southwest coast, we
expect that Archaic sites in this area would also contain a
reasonably high proportion of these cherts; however, this is
unproven at present.
Both Archaic and post-Archaic assemblages from sites in
the Kissimmee River region display abundance peaks for
Hillsborough River cherts. This is interesting since one would
expect that the closer, Peace River cherts would be more
dominant at these sites. The post-Archaic occupations at Fort
Center and Pineland also have high amounts of Hillsborough


2000 VOL. 53(2-3)


Table 2. Proportional representation of raw-material types (quarry clusters) in Archaic and post-Archaic components.

Hillsborough River ith hee Peace River Other
Site No. W acoohTotals
N Pet. N Pet. N Pet. N Pet.

Archaic Assemblages

8HG18 20 50.00

8HG20 21 72.41

8HG34 25 89.29

8HG35 13 23.21

8HG51 63 10.66

8HG678 25 44.64

8HG767 5 10.64

8HI2187 803 90.33

8H12188 604 94.82

8H13281 2291 97.04

8HI5429 292 98.65

8HR68 205 74.01

8HR71 143 54.37

8HR92 61 11.96

80S123 4 0.79

8PA667 279 53.86

8P04760 12 37.50

8P04761 33 67.35

Post-Archaic Assemblages

8GL13 247 27.66

8HG18 58 60.42

8HG20 79 66.39

8HG27 28 35.44

8HG678 41 45.05

8H1515 8328 97.67

8H1981 442 87.70

8HR44 11 6.15

8HR48 59 29.65

8LL33,36,37 78 59.54

8PA53 3 5.66

8PI894B 58 10.18

8P11753 215 47.78


















































38 4.26

3 3.13

1 0.84

0 0.00

29 31.87

0 0.00

0 0.00

161 89.94

81 40.70

0 0.00

0 0.00

0 0.00

0 0.00

15 37.50

7 24.14

3 10.71

16 28.57

98 16.58

21 37.50

15 31.91

62 6.97

14 2.20

59 2.50

4 1.35

57 20.58

56 21.29

130 25.49

68 13.36

161 31.08

20 62.50

9 18.37

574 64.28

24 25.00

27 22.69

45 56.96

18 19.78

198 2.32

62 12.30

6 3.35

57 28.64

46 35.11

25 47.17

510 89.47

228 50.67



































Table 2. (Continued).

Hillsborough River Upper Peace River Other
Site No. Withlacoochee Totals
N Pct. N Pct. N Pct. N Pct.

8P04737 39 42.86 16 17.58 3 3.30 33 36.26 91
8P04777 132 30.41 42 9.68 175 40.32 85 19.59 434
8P04778 288 27.40 64 6.09 431 41.01 268 25.50 1051

0 5 10

Figure 4. Locations of archaeological sites used in the study.

River cherts. To the north, the proportion of Hillsborough
River cherts in both Archaic and post-Archaic assemblages
drops off rather quickly as they are replaced by local cherts

from the Upper Withlacoochee and the nearby
Brooksville quarry clusters. (The latter is located to the
northwest of the Upper Withlacoochee Quarry Cluster
and extends to the Gulf Coast; see Figure 1).
Figure 6 shows the geographic distribution of
artifacts made from Suwannee Limestone cherts from
the Upper Withlacoochee Quarry Cluster. Cherts from
the Upper Withlacoochee Quarry Cluster exhibit a
slightly larger distribution geographically in post-
Archaic assemblages, although this may be due to the
absence in our data set of Archaic sites south of the
S Peace River region. However, use intensity is higher
during the Archaic, as indicated by high abundance
values (i.e., > 20 %) extending all the way to the Peace
River and Kissimmee River basins (Figure 6a). The
high spot for the Archaic data is actually to the east of
the main source area for these cherts. This is due to the
presence of one site (80S123) with over 80% represen-
tation for Upper Withlacoochee cherts (see Table 2).
This site is interesting because it is located nearly 40 km
(25 mi) from the nearest known source location for these
cherts. They are well represented at this site because,
despite its distance, the Upper Withlacoochee Quarry
Cluster represented the nearest source of lithic raw
materials for the site's occupants.
To the southwest, the representation of Upper
Withlacoochee cherts drops to well below 10% in both
Archaic and post-Archaic assemblages. The exceptions
are two of the Archaic sites in the Peace River Valley
(8HR71 and 8HR92) where Upper Withlacoochee cherts
represent 22% and 58% of the respective lithic assem-
blages. In general, however, the distributional trend for
these cherts during both periods appears to be to the
The geographic distribution of artifacts made from
Peace River cherts is shown in Figure 7. Interestingly,
among the Archaic assemblages the highest values for
these cherts occur at sites to the east of the Peace River;
that is, along the Lake Wales Ridge and in the
Kissimmee River Valley. Archaic sites in the Peace River
Valley generally have low abundance values for these cherts,
usually less than 10% (Table 2). Peace River chert artifacts


2000 VOL. 53(2-3)

0 5 10

Figure 5. Spatial distribution of Hillsborough River Quarry Cluster cherts in archaeological assemblages: left) Archaic Period; right) post-Archaic Period.
Contour intervals are 5%.


Gulf of Mexico Gulf of Mexico

K Lake Okeechobee Lake Okeechobee

0 5 10

Figure 6. Spatial distribution of Upper Withlacoochee Quarry Cluster cherts in archaeological assemblages: left) Archaic Period; right) post-Archaic Period. w
Contour intervals are 5%. ?

0 5 10

Figure 7. Spatial distribution of Peace River Quarry Cluster cherts in archaeological assemblages: left) Archaic Period; right) post-Archaic Period. Contour
intervals are 5%.


Table 3. Comparison of local (Peace River Quarry Cluster) versus non-local raw-material use between Archaic
and post-Archaic assemblages in the Peace River Valley.
LocaP Non-Localb
Site No. Totals
N Pct. N Pct.

Archaic Assemblages

8HR68 12 4.33 265 95.67 277

8HR71 5 1.90 258 98.10 263

8HR92 25 4.90 485 95.10 510

Post-Archaic Assemblages

8HR44 161 89.94 18 10.06 179

8HR48 81 40.70 118 59.30 199

8P04777 175 40.32 259 59.68 434

8P04778 431 41.01 620 58.99 1051

a Peace River Quarry Cluster chert.
b All other siliceous stone.

Table 4. Comparison of local (Peace River Quarry Cluster) versus non-local raw-material use between
occupation zones at two post-Archaic sites in the Peace River Valley.
LocaP Non-Localb
Site Components Totals
N Pet. N Pct.


Late (Levels 1-5) 150 57.69 110 42.31 260

Early (Levels 6-10) 25 14.37 149 85.63 174


Late (Levels 1-4)C 233 57.39 173 42.61 406

Middle (Levels 5-7)d 168 45.65 200 54.35 368

Early (Levels 8-12) 32 11.51 246 88.49 278

" Peace River Quarry Cluster chert.
bAll other siliceous stone.
ccal A.D. 1285-1305, carbon on exterior of ceramic sherd, Level 3.
d cal A.D. 90-225, carbon on exterior of ceramic sherd, Level 6.


2000 VOL. 53(2-3)


exhibit a wider geographic distribution during the post-
Archaic Period, with the center of distribution in the heart of
the source area. Four sites located within 10 km (6.25 mi) of
the main Peace River outcrops have assemblages that contain
40% or more of these cherts (Table 2). Two post-Archaic sites
on the Lake Wales Ridge also contain high abundance values
(> 30 %). Although the contour lines in Figure 7 make it
appear that the abundance values for these cherts drop off
gradually to the west, in reality we know of no sites in the
Hillsborough River region, even on the periphery, that contain
Peace River cherts.4 Moreover, no Peace River cherts appear
at the Pineland site on the southwest coast (Austin 1995a).
This is surprising given that the Peace River flows southwest
and empties into Charlotte Harbor, just north of Pine Island.
Instead this site is dominated by cherts from the Hillsborough
River Quarry Cluster. On the other hand, Fort Center, which
is located on Lake Okeechobee, does contain small quantities
of Peace River cherts (ca. 4 %).
Not surprisingly, the geographic distribution of the three
quarry-cluster cherts examined here suggests least-cost, least-
risk strategies of lithic procurement in that, in most cases,
cherts from the nearest source areas were used most inten-
sively by the inhabitants of those areas, as well as by people
living in nearby, chert-poor areas.5 Peace River cherts may be
an exception to this generalization. These cherts exhibit a
patchy distribution among sites in south Florida, including
those within the Peace River Basin, and they are virtually
nonexistent in assemblages to the north and west, suggesting
that it was not considered a high-utility resource. The brittle
nature of these opal-rich cherts and the presence of many
earthy inclusions may have made them less desirable than the
better-silicified cherts from other quarry clusters. Another
possibility for the limited use of Peace River cherts is their
restricted distribution, which may have been even more
restricted in the past. Many of the known sources of Peace
River cherts have been encountered in phosphate mines or in
areas along the Peace River where increased river flow
resulting from upstream mining activities may have exposed
chert sources that were previously unavailable.
Although the data set admittedly is incomplete, the spatial
patterns of distribution do suggest possible routes of movement
for chert or chert artifacts. For example, Peace River cherts
(and perhaps other cherts as well) appear to have moved down
the Kissimmee River or Lake Wales Ridge to Lake
Okeechobee, while Hillsborough River cherts may have moved
along the west coast, perhaps in canoes, to Pineland and other
sites along the coast. At present we do not know why the
cherts moved along the routes that they did. It is possible that
late prehistoric territorality and political circumscription were
factors that influenced the routes that people used to move
these cherts into south Florida, as local groups controlled
access to some quarries and directed the flow of raw materials
along trade routes.

Temporal Distribution

Turning now to a closer consideration of temporal varia-
tion, we focus on the Peace River region where temporal
differences in the uses of local and non-local cherts are best
exemplified in individual assemblages. The data used for this
comparison come from 7 sites: 3 Archaic and 4 post-Archaic
(Table 3). Figure 8 shows the proportional representation of
local (i.e., Peace River) versus non-local (i.e., non-Peace
River) cherts for both Archaic and post-Archaic assemblages.
The differences between the two are dramatic. Archaic sites
contain much higher proportions of non-local cherts, primarily
from the Hillsborough River Quarry Cluster, while post-
Archaic sites exhibit a much greater reliance on the local
Peace River cherts. The representation of local cherts at post-
Archaic sites ranges from 40% to more than 80%.
The shift from non-local to local lithic resources is believed
to be a reflection of the shift from a more mobile settlement
strategy during the middle-to-late Archaic Period in central
and south Florida to one of reduced residential mobility and a
more restricted territory during the post-Archaic (cf. Austin
1996, 1997). We hesitate to state that post-Archaic peoples in
the Peace River region were sedentary, but certainly their
world was getting smaller, and this is reflected by the replace-
ment of non-local, high-utility cherts by local, lower-utility
cherts. This shift towards a dependence on the use of local
lithic resources was accompanied by a technological shift that
emphasized tool expediency. Analysis of post-Archaic tools
and debitage indicates that Peace River cherts were used most
often for the manufacture of flake tools and the small, triangu-
lar Pinellas point, which was used in a bow-and-arrow
technology (Janus Research 1996, 1997). These small bifaces
represent an elaboration of flake-tool technology since their
production involves pressure retouch of small flakes into a
standardized, triangular shape. The few large, stemmed
bifaces that appear in the post-Archaic assemblages were
almost all made from non-local cherts.
The inferred reduction in territorial range appears to have
occurred sometime around AD 100. Table 4 shows raw-
material data from two post-Archaic sites, Three Dog
(8P04777) and Hickory Hammock (8P04778), that are located
on opposite banks of a small creek that flows into the Peace
River. The data have been partitioned by occupational zones
that were identified on the basis of differences in artifact
content, ceramic type frequencies, and radiocarbon dates
(Janus Research 1997). AMS dates obtained on carbon residue
from the exterior of two pot sherds recovered from Hickory
Hammock indicate that the latest occupation dates to cal A.D.
1285-1305 while the middle occupation dates to about cal A.D.
90-225 (Janus Research 1997:Table 14). The ceramic type
frequencies in these occupation zones are consistent with these
dates. The lowest, and earliest, zone pre-dates A.D. 100 and
is characterized by a reliance on non-local lithic resources. A
similar pattern is exhibited at 8P04777. At both sites, the
later occupations exhibit the greatest dependence on local,
Peace River cherts while earlier occupations are dominated by
non-local cherts (Figure 9).




Archaic Sites
Peace River Region

8HR68 8HR71 8HR92

Post-Archaic Sites
Peace River Region

8HR44 8HR48 8P04777 8P047778

Figure 8. Temporal variation in the use of local (Peace River) versus non-local cherts at a) Archaic Period and b) post-
Archaic Period sites in the Peace River Basin.


2000 VOL. 53(2-3)





1-5 6-10


1-4 5-7 8-12

Figure 9. Temporal variation in the use of local (Peace River) versus non-local cherts at two post-Archaic sites in the
Peace River Basin: a) Three Dog, 8P04777; b) Hickory Hammock, 8P04778.





The purpose of this paper was to illustrate by example the
utility of employing raw-material provenance analysis in
archaeological research. Although the examples we have
presented here were designed to examine general spatial and
temporal trends, they have led us to consider more complex
issues related to understanding the causes for the patterns that
we have observed. These include factors of differential
settlement mobility, access to chert resources, exchange
systems, territoriality, and political circumscription and
control of resources. All of these issues can be addressed by
incorporating raw-materials analysis into archaeological
research programs (cf. Archaeological Consultants, Inc./Janus
Research 1994; Austin 1995a, 1995b, 1996, 1997, 1998;
Estabrook 1992, 1997; Estabrook and Williams 1992; Janus
Research 1995, 1997). Moreover, the geographic areas that
offer the greatest potential for examining these issues through
lithic data are not necessarily those areas where chert is
abundant. Instead, it is in those areas where chert is non-
existent or is limited in availability that such research will be
most profitably conducted, since these are the areas where
archaeological assemblages are small enough to be completely
analyzed and will display the variability in raw-material
composition that is necessary to address these questions

' Site and assemblage totals differ because three sites (8HI18, 8HI20,
and 8111678) have both Archaic and post-Archaic components and the
post-Archaic Pineland Site complex consists of three site numbers
(8LL33, 8LL36, and 8LL37).
2 The Upchurch chert collection is presently housed at the Florida
Museum of Natural History in Gainesville (FLMNH accession
numbers 93-1 and 93-15). Large collections of chert samples also
have been assembled by the authors and are stored at the offices of
Southeastern Archaeological Research, Inc. and Panamerican
Consultants, Inc.
SSilicified coral is another common raw-material type that occurs
with moderate to abundant frequency throughout central and south
Florida. Silicified coral, particularly the species Siderastrea, is
present in both the Arcadia and Suwannee Limestone formations.
Assignment of individual artifacts manufactured from silicified coral
to a specific quarry cluster is presently not possible due to an absence
of observable diagnostic features. Therefore, all silicified coral
artifacts have been included in the "Other" category in Table 2, and
were not included further in our analysis.
4 A recent survey in eastern Manatee County has recovered Peace
River cherts in small amounts from several post-Archaic sites there
(Janus Research 1998). To our knowledge, these represent the
western-most distribution of Peace River cherts so far documented.
5 It should be noted that cherts from several other quarry clusters
located in north-central Florida, including cherts derived from the
Eocene-age Ocala Limestone, are represented in several of the site
assemblages reported here, albeit in very low percentages. While
their presence at sites in central and south Florida is important for
documenting direct and/or indirect procurement of chert from more
northerly source areas, a full treatment of these data was beyond the
scope of this paper. See Austin (1995a, 1996, 1997) for extended
discussions on this topic.

The authors wish to thank Donna Ruhl and Christine Newman for
organizing the session on Current Research and Trends in Florida
Archaeology at the 1997 Southeastern Archaeological Conference
meeting in Baton Rouge, and for inviting us to participate. This
paper benefitted from the comments of Al Goodyear, Barbara Purdy,
Donna Ruhl, Dana Ste. Claire, Barry Wharton, and Kathleen
Hoffman, but only the authors can take responsibility for any errors
of logic or interpretation. Dawn Van DePutte drafted the quarry
cluster map in Figure 1 and Scott Mitchell illustrated the fossils in
Figure 2. Some of the data used in this paper comes from CRM
surveys and test excavations conducted by the authors while em-
ployed at Janus Research. We thank Ken Hardin for allowing us to
use these data here.

References Cited

Archaeological Consultants, Inc./Janus Research
1994 Phase IIIMitigative Salvage Excavation at the Muck Pond
East Site (8HI515) Located Within the Proposed SR 574
ImprovementArea, Hillsborough County, Florida. Report
prepared for the Florida Department of Transportation,
District Seven by Archaeological Consultants, Sarasota and
Janus Research, St. Petersburg.
Austin, Robert J.
1995a Lithic Acquisition and Use at Pineland. To be published
as Chapter 14 in The Archaeology ofPineland: A Coastal
SouthwestFlorida Village Complex, A.D. 100-1600, edited
by Karen J. Walker and William H. Marquardt. University
of Florida, Institute ofArchaeology and Paleoenvironment-
al Studies, Gainesville.
1995b Stone Tool Technology and Use. In Yat Kitischee: A
Prehistoric Coastal Hamlet 100 B.C. A.D. 1200, edited
and compiled by Robert J. Austin, pp. 176-218. Report
prepared for the Board of County Commissioners of
Pinellas County Florida by Janus Research, St. Petersburg.
On file, Pinellas County Planning Department, Clearwater
and Florida Division of Historical Resources, Tallahassee.
1996 Prehistoric Chert Procurement and Settlement Strategies on
the Lake Wales Ridge. The Florida Anthropologist
1997 The Economics of Lithic-Resource Use in South-Central
Florida. Ph.D. dissertation, Department ofAnthropology,
University of Florida, Gainesville.
1998 Technological and Functional Analysis of Stone Artifacts
from the Narvaez/Anderson Site (8PI54), with Comments
on Time Allocation and Craft Specialization. In The
Narvaez/Anderson Site (8PI54), A Safety Harbor Culture
Shell Mound and Midden, A.D. 1000-1600, edited by
Terrance L. Simpson, pp. 87-200. Central Gulf Coast
Archaeological Society, Tampa.
Chance, Marsha A.
1981 Wetherington Island: A Lithic Procurement Site in
Hillsborough County. The Florida Anthropologist 34:109-
Chance, Marsha A., and Elizabeth J. Misner
1984 Archaic Lithic Procurement Behavior at Wetherington
Island, Hillsborough County, Florida. Report prepared for
the Florida Department of Transportation and the Federal
Highway Administration by the Florida Division of Ar-
chives, History, and Records Management, Bureau of
Archaeological Research, Tallahassee.


2000 VOL. 53(2-3)



Cole, W. Storrs
1945 Stratigraphic and Paleontologic Studies of Wells in
Florida No. 4. Geological Bulletin No. 28, Florida
Geological Survey, Tallahassee.
Cushman, J. A.
1948 Foraminifera: Their Classification andEconomic Use (4th
Edition). Harvard University Press, Cambridge.
Dunham, R. J.
1962 Classification of Carbonate Rocks According to
Depositional Texture. In Classification of Carbonate
Rocks, edited by W. E. Ham, pp. 108-121. American
Association of Petroleum Geologists Memoir 1, Tulsa.
Estabrook, Richard W.
1992 It's Gotta Come From Somewhere: The Acquisition and
Use of Chert in the Central Highlands. Paper presented at
the 44th annual meeting of the Florida Anthropological
Society, St. Augustine.
Estabrook, Richard W., and J. Raymond Williams
1992 Analysis of Lithic Materials from the Rattlesnake Midden
Site (8HI980), Tampa Bay, Florida. The Florida Anthro-
pologist 45:39-51.
Goad, Sharon I.
1979 Chert Resources in Georgia. Laboratory of Archaeology
Series Report No. 21, University of Georgia, Athens.
1984 Chert Utilization in Georgia. In Prehistoric Chert Exploi-
tation: Studies from the Midcontinent, edited by Brian M.
Butler and Ernest E. May, pp. 71-85. Center for Archaeo-
logical Investigations Occasional Paper No. 2. Southern
Illinois University, Carbondale.
Janus Research
1995 Phase II Archaeological Investigations at Three Prehis-
toric Sites (8HR68, 8HR71, & 8HR92) on IMC-Agrico
Company's Fort Green Mine Southern Reserves, Hardee
County, Florida. Report prepared for IMC-Agrico Com-
pany by Janus Research, St. Petersburg. On file, Florida
Division of Historical Resources, Tallahassee.
1997 Archaeological Investigations at Two Prehistoric Sites on
Cargill Fertilizer, Inc.'s South Fort Meade Mine, Polk
County, Florida. Report prepared for Cargill Fertilizer,
Inc. by Janus Research, St. Petersburg. On file, Florida
Division of Historical Resources, Tallahassee.
1998 A Cultural Resource Assessment Survey of IMC-Agrico
Company's Four Corners Mine DRI Amendment Areas,
Manatee County, Florida. Report prepared for IMC-
Agrico Company by Janus Research, St. Petersburg. On
file, Florida Division of Historical Resources, Tallahassee.
Johnson, William Gray
1986 The Archaeology of the Green Swamp: A Predictive Model
for a Swamp Environment. M.A. thesis, Department of
Anthropology, University of South Florida, Tampa.
Luedtke, Barbara E.
1979 The Identification of Sources of Chert Artifacts. American
Antiquity 44:744-756.
1992 An Archaeologist's Guide to Chert and Flint. Institute of
Archaeology, University of California, Archaeological
Research Tools 7, Los Angeles.
Palmer, Jill, James Dunbar, and Danny Clayton
1982 Report on Phase II Underwater Archaeological Testing at
the Fowler Avenue Bridge Mastodon Site (8Hi393c/uw),
Hillsborough County, Florida. Interstate 75 Highway
Phase II Archaeological Reports Number 5, Florida Divi-
sion of Archives, History, and Records Management,
Bureau of Archaeological Research, Tallahassee.

Piper, Harry M., and Jacquelyn G. Piper
1980 An Archaeological Survey of the Levee Corridors of the
Proposed Green Swamp Flood Detention Area, Polk,
Sumter, and Lake Counties, Florida. Report prepared for
the Southwest Florida Water Management District by Piper
Archaeological Research, Inc., St. Petersburg. On file,
SouthwestFloridaWaterManagementDistrict, Brooksville
and Florida Division of Historical Resources, Tallahassee.
Rapp, George, Jr.
1985 The Provenance of Artifactual Raw Materials. In Archaeo-
logical Geology, edited by George Rapp, Jr. and John A.
Gifford, pp. 353-375. Yale University Press, New Haven.
Simpson, J. Clarence
1939 Aboriginal Stone Quarries of Hillsborough County and
Sources of Abrasives and Pigments. In Third Biennial
Report, Biennium Ending December 31, 1938, pp. 31-32.
Florida State Board of Conservation, Geological Division
and Archaeological Division, Tallahassee.
1941 Source Materials for Florida Aboriginal Artifacts. Pro-
ceedings oftheFloridaAcademy ofSciences for 1940 5:32-
Taylor, T. N.
1981 Paleobotany: An Introduction to Fossil Plant Biology.
McGraw-Hill, New York.
Upchurch, Sam B.
1980 Chert Origins and Availability, Pinellas and Hillsborough
Counties, Florida. In Holocene Geology and Man in
Pinellas and Hillsborough Counties, Florida, compiled by
Sam B. Upchurch, pp. 48-55. Southeastern Geological
Society Guidebook No. 22, Tallahassee.
Upchurch, Sam B., Richard N. Strom, and Mark G. Nuckels
1981 Methods ofProvenance Determination ofFlorida Cherts.
Report prepared for the Florida Division of Archives,
History, and Records Management, Bureau of Historic
Sites and Properties, Tallahassee.
Wharton, Barry R.
1984 Archaeological Resources of the Upper Hillsborough
Flood Detention Area, Pasco and Hillsborough Counties,
Florida. Southwest Florida Water Management District,




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.



'Bureau ofArchaeological Research, 500 S. Bronough St., Tallahassee, FL 32399-0250
E-mail: rwheeler@mail.dos.state.fl.us
2Bureau ofArchaeological Research, C.A.R.L. Archaeological Survey, 504 17h St., St. Augustine, FL 32095
E-mail: cnewman@mail.dos.state.fl.us
35814 NW 31" Terrace, Gainesville, FL 32653

Acquisition of property along the eastern side of the St.
Johns River, south of Lake George and north of Lake Dexter,
by the State of Florida's Conservation and Recreation Lands
(C.A.R.L.) program lead to the creation of the Lake George
State Forest in 1993. The C.A.R.L. Archaeological Survey is
a program within the Bureau of Archaeological Research
designed to work with agencies managing state lands in an
effort to identify, manage, and preserve cultural resources.
Research indicated that Mount Taylor (8V019) and Bluffton
(8VO22, 8V023), two sites that have played an important role
in our ideas about the culture history of the St. Johns River
area, were located on the Lake George State Forest property
(Figure 1).
Despite investigations at these two important sites (Bullen
1955; Moore 1893), review of the literature revealed that site
plans had never been made and the plotted locations of the
sites were unclear. There also were questions regarding the
current condition of the sites and whether any intact deposits
still existed. This paper presents the results of the C.A.R.L.
Archaeological Survey work at the two sites, which has
included nomination of Mount Taylor to the National Register
of Historic Places.' This information augments our current
knowledge of the pre-pottery Archaic Mount Taylor Period,
which also is discussed here.

Mount Taylor

Mount Taylor (8V019) is a large snail and mussel shell
midden located on the east side of the St. Johns River, about
1.8 km south of Volusia and State Route 40 (Figure 1). The
site is 178 m east of the river. Access is best made, via roads
and trails, through the pine woods to the east. The site was
subjected to shell mining from the late 1890s through the
1920s, and eight trenches from prior excavations and looting
mark the surface of the midden. Despite damage from mining
and looting the site retains its original outline and considerable
portions of the midden remain undisturbed. Limited auger
testing revealed archaeological wet site deposits extending to
over a meter below the ground surface. Excavations in the site
by C. B. Moore (1893:12-13) revealed that the midden
accumulated prior to the use of pottery in the region. John
Goggin (1949:22), based on the site's presumably early age,
designated the pre-pottery Archaic of the northern St. Johns

area as the Mount Taylor Period (6000 BP- 4000 BP).

Environmental Setting

Mount Taylor is located in a cypress swamp within the
flood plain of the St. Johns River, south of Lake George and
north of Lake Woodruff and Lake Dexter. Many places along
the river above Lake George have expansive cypress swamps.
The existence of large middens in these swamps suggest the
river was once broader, with sites accumulating on river
islands or along the banks. The latter is probably the case with
Mount Taylor. Sedimentation, changes in the river's course,
and/or lower water levels have allowed the development of
cypress swamps within the river flood plain.
Vegetation in the surrounding river swamp is dominated
by cypress and water oak, with numerous ferns on higher
areas. The vegetation on Mount Taylor is quite different,
consisting of live oak, sweetgum, buckeye, red cedar, cabbage
palm, hornbeam, saw palmetto, toothache tree, and other
hardwood hammock plants.

Present Condition

Mount Taylor is a large aboriginal midden-mound com-
posed of freshwater snail and mussel shell, animal bone, bone
artifacts, shell artifacts, and stone tools. The height of the
mound has been significantly reduced by shell mining, though
the horizontal extent is clearly visible. The site runs roughly
east-west, at an angle of 117 degrees west of north. A map
madebythe C.A.R.L. Archaeological Survey indicates that the
site is ovoid in shape and measures 130 m along its longer
east-west axis (Figure 2). The site is 55 m along its north-
south axis. The site rises to approximately 3 m above sea level
at its highest point at the north edge of the site. Profiles made
from old excavation and vandal trenches reveal distinct
stratification and buried site features.

Previous Research

Clarence B. Moore (1893:12-13, 113-115) described
Mount Taylor as one of the highest shell mounds on the St.
Johns River, with deposits 25 to 27 ft thick. Moore dug four
excavation trenches in the site, and his accounts are repro-
duced below:


VOL. 53(2-3)



Excavation I
West end of ridge on summit, 7x4x7 1/2 feet
deep. At a depth of 1 foot 6 inches, an
arrow-head with bones of lower edible
animals, was found. Two feet down were
fragments of turtle shell, and half a foot
lower, a fragment of a bone awl was met
with. At 5 feet, 6 inches, from the surface
was a well defined fire-place, with bones of
the turtle. The shells are almost exclusively
Paludinae (cf. Viviparus georgianus), with
occasional Unionidae and a fewAmpullariae
(cf. Pomacea sp).

Excavation II
North-east side of mound about ten feet
vertically from the base. Besides the usual
fire-places and bones of edible animals, a
fragment ofaFulgur [nowBusycon sp.] was
found at a depth of four feet.

Excavation iI
On summit near center of ridge, 9x9x1 1 feet
deep. No clearly marked strata were met
with, the excavation being carried through
shell with a considerable percentage of
broken shell and sand. The following ob-
jects were found:-flint flake at a depth of 6
inches; fragment of bone awl 1 foot down;
small pendant ornament of shell 1 / feet
from surface, with bone implement; shell
gouge (broken) 2 feet down; rude arrow-
head on fire-place, 4 V feet down; flint flake
5 feet down; arrow-head lying on fire-place,
at a depth of 6 feet from surface; portion
of arrow-head on fire-place, bearing marks
of fire at a depth of 8 feet; fragment of bone
awl 9 feet 9 inches down; rude bone awl 10
feet down; bones of edible animals, mainly
of alligator, deer and turtle, at all depths.

Excavation IV
At margin of base north-east portion of
mound. Trench 11 1/2x10x6 feet deep at
end. A piece of coquina, smooth on one
side, evidently used for polishing, at a depth
of 2 % feet, and a fragment of arrow-head
(see Figure 5j) with the usual bones of
edible animals, were the only objects of
interest met with [Moore 1893:113-114].

For over 50 years Moore's excavated material represented
the most complete picture of the Mount Taylor Period. Goggin
(Florida Master Site File [FMSF] n.d.), in the 1940s, recorded
the site as V019 in the Archeological Site Survey of the
University of Florida, which formed the basis for the Florida
Master Site File. Goggin's description of the site is drawn
from Moore's account, but also includes information on the
excavated material, which Goggin examined at the Heye
Foundation, Museum of the American Indian in New York.2
Mount Taylor was visited by archaeologist L. Scott Nidy in
1973 who described the mining damage suffered in the 1920s

Figure 1. Location of Mount Taylor and Bluffton sites in the Lake George
State Forest.

when shell was carried off by convict labor in wheel barrows
and then barged to Jacksonville for road fill (1973 FMSF
8V019). Nidy noted that three of Moore's excavation trenches
were still visible at the time of his visit.

Current Research

Information compiled by the C.A.R.L. Archaeological
Survey in preparation for a National Register nomination
includes a plan map of the site (Figure 2), documentation and
profiles of old excavation trenches (Figure 3), auger testing for
wet site deposits, and a limited surface collection that pro-
duced a Strombus gigas shell celt (Wheeler and Newman
1997). Eight old excavation trenches and their associated


2000 VOL. 53(2-3)







SAuger 3

Trench 2

Trench rench 4


Auger I


C.A.R.L. Archaeological Survey
Bureau of Archaeological Research

S 10 20 30 40 m



Figure 2. Plan of Mount Taylor (8V019).


APRIL 9, 1997

I Humus
II Viviparus Midden
I' Viviparus Midden/Brown Soil
HI Brown Midden
IV Black Soil
V Apple Snail/Crushed Shell
VI Concreted Viviparus
VII Apple Snail/Crushed Shell

JANUARY 14, 1997

Figure 3. Profiles of Trench 2 and Trench 6, Mount Taylor (8V019).

300 cm


2000 VOL. 53(2-3)



0 2.5S


0 1 2 3 4 5 6 7cm
| i f i l i

0 1 2 3cm

- -- r.


o !2 3 ? Sc!

Figure 4. Artifacts from Mount Taylor (8V019): a) shell tube; b) Strombusgigas shell bannerstone; c-g) Florida Archaic
Stemmed points; h-i) decorated bone artifacts;j) Strombusgigas shell celt. a-b) redrawn from photographs in the Goggin
Collection, P.K. Yonge Library of Florida History, University of Florida, Gainesville; c-j) redrawn from Goggin
(1952:Plate 7j-m); g) redrawn from Moore (1894b:Figure 3); h-i) redrawn from sketches by John Goggin, on file, P.K.
Yonge Library of Florida History;j) collected during C.A.R.L. Archaeological Survey visit



Table 1. Artifacts from Mount Taylor in the University of Pennsylvania Museum of Archaeology and Anthropology.

Material Species/Element Description Dimensions' Weight Catalog No.
antler deer, antler decorated, rectangular object 29.2 x 17.9 x 13.8 6.0 L563-6142
bone deer, metapodial bipoint 84.9 x 11.9 x 6.8 5.5 L563-6134
bone deer, long bone hafted point 100 x 15.6 x 7.1 8.5 L563-6135
bone deer, metapodial point/shaft of tool 76.3 x 10.6 x 6 4.2 L563-6136
bone deer, long bone hafted bone point 96.2 x 13.2 x 5 5.7 L563-6137
bone deer, metapodial hated bipoint 71.6 x 8.6 3.9 L563-6138
bone deer, metapodial point/shaft of tool 99.6 x 10.9 x 8.2 6.9 L563-6139
bone deer, metapodial point frag. 68.9 x 8.1 4.1 L563-6140
bone deer, metapodial haft end frag. 42.8 x 10.4 2.4 L563-6141
antler deer, antler cut section 43.6 x 26.3 x 21.1 14.1 L563-6143a
bone deer, ulna cut and score marks 92.6 x 38 x 10 20 L563-6143b
antler deer, antler cut section 110.6 x 21.2 x 17 24.8 L563-6143c
shell Busycon carica, columella beachwom, unworked 115 x 17.1 72.8 L563-6132
limestone Anastasia coquina limestone beachwom, unworked 58 x 43.6 x 6 37.8 L563-6133
limestone coquina limestone flat, fragmentary 79.4 x 38.1 81.3 cg-94-2-20c
chert Archaic stemmed point, base 40.2 x 40.8 x 9.6 13.8 cg-94-2-20c
shell Strombus gigas celt 67.8 x 43.6 x12.6 47.1 cg-94-2-20c
chert flake 65.3 x 27.8 x 7.2 11.6 cg-94-2-20a
chert flake, heat altered 26.2 x 21.7 x 3.6 1.8 cg-94-2-20a
chert crude point/drill 34.8 x 11.7 x 8.6 3.5 cg-94-2-20a
bone deer, metapodial reduction debitage 48.8 x 16.9 4.6 cg-94-2-20
bone deer? sliver 61.2 x 9.2 x 3.9 2.1 cg-94-2-20a
shell Dinocardium robustum valve frag. 53 x 31.2 x 4 10 cg-94-2-20
a Measurements in millimeters and grams. Length/height given first, then width and thickness.

spoil were documented during the mapping project in early
1997. Five of these trenches must have been made since 1973
ifNidy's count is accurate. A profile obtained from Trench 6
(see Figure 3) reveals the following strata:

Stratum A Viviparus snail shell and crushed mussel shell.

Stratum B Viviparus shell and black earth midden.

Stratum C Apple snail and crushed shell in a matrix of loose
soil. This stratum intrudes down into Stratum E, the con-
creted layer, and may be a solution feature in the concretion
that was filled with midden.

Stratum D Black earth with some crushed mussel shell.

Stratum E Concreted Viviparus shell that would seem to
underlie much of the site. This concreted shell layer is
present at the bottom of many of the St. Johns River middens.

Stratum F Sandy clay with some shells. This is probably the
original river bank on which midden accumulation was

Trench 2 also was profiled during the C.A.R.L. Archaeological
Survey investigations (Figure 3). This profile revealed a

number of complex strata and features. As noted above, a
concreted shell layer (Stratum E) is present in the lower
portion of the profile, and contains what may be solution
features that have filled with shell midden material.
Three auger tests made near the western end of the site
revealed wet site deposits to a depth of over 1 meter below the
ground surface in that area.3 Material brought up with the
auger included Viviparus snail shell midden with well-pre-
served wood. The wet site component of Mount Taylor
probably holds some of the site's most significant resources.


Artifact collections from Mount Taylor are primarily those
of C. B. Moore (1893:113-115; 1894:18). Goggin
(1952:Plates 7j-m; FMSF 8V019) mentions and illustrates
some of the materials now at the National Museum of the
American Indian. Another collection of material from Mount
Taylor is present in the University of Pennsylvania Museum of
Archaeology and Anthropology.4 This material is part of
Moore's original collection from the site, which was appar-
ently divided amongst several museums. Some of the objects
mentioned or described by Moore (1893) are present in this
collection, including bone artifacts, marine shell artifacts,


2000 VOL. 53(2-3)


) I 1''Ii

a!l ii


50 mm

i 1

b a



~I Je ll il



ki I

Figure 5. Bone, shell, and stone artifacts from Mount Taylor (8V019), in the collection of the University of Pennsylvania
Museum of Archaeology and Anthropology: a) hafted bone bipoint, L563-6138; b) bone point fragment, L563-6140; c)
bone point, L563-6136; d) bone bipoint, L563-6134; e) bone tool, haft end fragment, L563-6141;J) hafted bone point, L563-
6137; g) hafted bone point, L563-6135; h) bone point, L563-6139; i) Strombus gigas shell celt, cg 94-2-20; j) Florida
Archaic Stemmed point, cg 94-2-20; k) decorated bone object, L563-6142



; ;
i I
i '




projectile points, and limestone artifacts, though it is difficult
to match artifacts with those in Moore's description. Human
remains also were present in this collection, but have not been
examined. Table 1 summarizes the material examined in the
University of Pennsylvania Museum.
Moore (1894b:18, 21; Goggin 1952:Plate 7j-m) recovered
at least six chert projectile points from Mount Taylor, and
notes the relative position of three of these in his discussion.
Drawings of five of these points (see Figure 4) were made from
illustrations in Moore (1894b:Figure 3) and Goggin
(1952:Plate 7j-m). The base of another point was present in
Moore's collection at the University of Pennsylvania Museum
(see Figure 5j). All six points conform to the Florida Archaic
Stemmed series as described by Bullen (1975:32), but are
difficult to assigned to particular types (e.g., Putnam,
Five bone artifacts reported by Moore (1893:113-114)
include bone awls, fragments of awls, and unidentified
implements. Goggin (1952:42) notes that two decorative bone
pins cataloged from Mount Taylor are in the Heye Foundation
collection (see Figure 4h-i). One is a fragmentary pin with
incised geometric decoration, the other is a complete pin with
carved decoration (FMSF 8VO19). This latter specimen is
identical in form to late Middle Archaic bone pins from the
lower Ohio-central Mississippi river valley region (Jefferies
1996:227-228). The University of Pennsylvania Museum
collection includes eight bone artifacts produced from deer
metapodials, as well as two cut sections of antler, and several
fragments of bone tool production debitage (see Table 1 and
Figure 5). The deer metapodial implements have wear
patterns indicating that they were hafted in handles and used
in textile or mat production.
Moore recovered several shell artifacts fromMount Taylor,
including a shell pendant ornament (Moore 1893:Figure 4), a
shell tube, and a shell gouge made from a Busycon carica
(Goggin 1952:Plate 8j). Similar pendants and tubes of shell
are known from Tick Island (Jahn and Bullen 1978:Figures
25, 47). A diminutive Strombus gigas celt is present in the
University of Pennsylvania Museum collection (see Figure 5i).
Moore (1898) also reports on a bannerstone made from a
Strombus gigas conch shell, found during mining at the site
(see Figure 4b). This appears to be a unique form of the
Archaic bannerstone type, which probably was used in
conjunction with an atlatl or dart thrower. A Strombus gigas
celt was found on the surface of the site during a 1996 visit
(see Figure 4j).
Several miscellaneous objects in the University of Pennsyl-
vania Museum collection are interesting, including fragments
of unmodified, beach-worn, marine shell and coquina lime-
stone. These types of items are commonly encountered today
on Atlantic beaches and suggest the riverine Mount Taylor
peoples may have been visiting the coast, as it seems unlikely
that these were items of exchange.


The Bluffton site was once a large shell mound with
associated shell fields and burial mounds. Wyman (1875:37)
notes that the site occupied 17 acres (6.9 hectares), stating that
the site "consists of a high rounded shell mound rising twenty
feet above the river ... and a long ridge extending in a south-
erly direction, nearly parallel with the river. There are two
burial mounds connected with it." Wyman (1875:37) dug in
one of these mounds and found human remains. Moore
(1894a:48) notes that Bluffton was one of the largest shell
mounds on the river, and estimated that shell deposits occu-
pied 35 acres (14.2 hectares), with a thickness of 25 ft (7.6 m)
in one area. Moore found few sherds of pottery in the area to
the north of the burial mound, while the area to the south, in
the orange grove, had a great deal of pottery; this observation
suggests some horizontal temporal differences in site areas.
Bullen (1955) made excavations at the site in 1951 during the
shell mining operation and recovered valuable information on
the structure and dating of the site.
Moore (1894a:45), like Wyman, describes two mounds at
Bluffton, "a conical mound of sand and shell ... and a mound
of sand somewhat more oblong in form than the usual trun-
cated cone." The former mound was located behind (to the
east) the main shell midden, and was roughly 100 ft (30.5 m)
in diameter by 16 ft (4.9 m) high. Both Sears (1960:57) and
Moore (1894a:44-48) noted that the bulk of the mound was
composed of basket-loaded shell midden. Below this was the
core mound of sand, under which was buried a single individ-
ual. The entire mound was constructed over older shell
midden deposits. Several intrusive burials were made in the
upper portion of the mound.

Environmental Setting

Unlike Mount Taylor, which is located in a cypress swamp,
Bluffton is situated directly along the St. Johns River, approxi-
mately 1.5 km north of Lake Dexter. The site is bordered to
the north and south by swampy areas or sloughs, and to the
east by mixed hardwood and pine, and pine flatwoods. Figure
1 illustrates the position and area occupied by the Bluffton site

Present Condition

Mining for shell in the late 1940s and 1950s severely
diminished the site, and actually created an "island" of midden
material in the St. Johns River. This island would appear to
be the remains of the shell ridge described by Wyman. Mining
largely ended when a dense layer of concreted shell midden
was reached and much of the surface of the site today is
composed of this concretion layer. It is likely that wet site
deposits lie below the concretion, and in the adjoining part of
the river. For this reason the underwater portion of the site

2000 VOL. 53(2-3)




~~~1-~ ~ ~ __________
S.. ... ...- .

'' .'.' .- ..
30IO0 40110 SOLO OLIO 70LIO 80O10 90IO IOLIO

Figure 6. Plan and profile of Bluffton Burial Mound (from Sears 1960:Figures 1 and 2).


Figure 7. Photograph of excavation, Bluffton Burial Mound, by William Sears in 1959. Core mound profile, stake 70L20
at upper left. Collections of the Anthropology Department of the Florida Museum of Natural History, Cat. No. PN94.228.456.

should be included in the recorded extent of the site. The site
is roughly 600 m long and 350 m wide.

Previous Research

Wyman (1875:37) briefly describes his work at the site,
which he refers to as "Orange Bluff." This consisted of
excavations into one of the two burial mounds, noting that
previous diggers had already removed several burials. Wyman
(1875:37) simply notes that he found several human skeletons
in the upper portion of the mound, with no accompanying
artifacts. Excavations into the large shell mound produced
fragments of pottery and stone artifacts, including a large
biface (Wyman 1875:49, Plate II, Figure 1).
Moore (1894a:44-45) also mentions the twoburial mounds,
noting Wyman's excavations into one of them. Moore
(1894a:45) excavated the oblong sand mound, which was built
on a shell ridge. Moore dug a trench in 1892, 38 ft (11.6 m)
long and varying in width from 5 ft (1.5 m) to 11.5 ft (3.5 m),
on the eastern side of the mound. The construction sequence
included a smaller sand mound of brown sand and shell, with

an outer mantle added later, producing the irregular shape of
the completed mound. Only superficial burials were encoun-
tered, along with 3 broken projectile points, a tubular shell
bead, 3 chert flakes, and a worked deer antler. The shell
midden base upon which the mound was constructed was
apparently of the pre-pottery Archaic, as tests by Moore
(1894a:48) produced no ceramics.
Sears (1960:55-56) made additional tests in the burial
mound investigated by Moore. The dimensions of the mound,
100 ft (30.5 m) in diameter and 16 ft (4.9 m) high, compare
well with the description given by Moore, as do the cross-
sectional diagrams generated by both authors (Figure 6 is
Sears's site plan and cross-section diagram). Figure 7 is a
photograph of the excavated burial mound. Note that the
mound was capped with a layer of shell midden material,
apparently borrowed from adjacent parts of the pre-pottery site.
Fill associated with intrusive burials made into the top of the
mound contained sherds, but the original construction of the
mound predates the St. Johns I Period (Sears 1960:58), and
may be Archaic in age. Interestingly, Piatek's (1994:114)
excavations at the Tomoka Mound Complex indicated that


2000 VoL. 53(2-3)


Tii ,



~ ~-
.... i i
'.I ~
'' '' '~'_: ~ .- .. 1.-
i.' ~-~f ~ J~ r ~ urrws-r ~I
L .'. .-! -.
,-t4QWI1~. *I;-i..'~''' -- I

S. -- .. .:..' '

Figure 8. Looking to the east at the excavation of the large Bluffton shell mound. Bill
Edwards, of the Florida Geological Survey, is shown at top with worker in the vicinity
of Test H. Collections of the Anthropology Department of the Florida Museum of
Natural History, Cat. No. PN94.228.784.

midden material was used in construction of some of the
Archaic mounds there as well. The
Excavations into the large shell mound by the Florida
Geological Survey and Florida Park Service in 1951 are The Mount Taylor
summarized by Bullen (1955). Figures 8 and 9 show the tal in the development
excavations, which occurred while the site was being mined era in the St. Johns R
for shell. Bullen (1955:2) dug a 25 ft (7.6 m) long trench, their significance little
comprised of 5 units, through the southern part of the mound, sites, and in the cas
This trench varied in width from 5.5 to 7 ft (1.7 to 2.1 m), and material culture and
was excavated to a depth of 16 to 17.5 ft (4.9 to 5.3 m). Six work with the Mount
main layers, some with various substrata and features, were like Groves Orange
identified. Layers I and II contained fiber-tempered pottery (McGee and Wheeler
and date to the Orange Period. Layers III, IV, V, and VI were analog of this culture.
deposits of the pre-pottery Mount Taylor Period. integrate information
number of sites, wit
Current Research Taylor and Bluffton s

The C.A.R.L. Archaeological Survey work at Bluffton has
focused on producing a map of the site, prior to the extensive
mining, in an effort to locate the relative positions of the site
components noted above. This was accomplished using
enlargements of black-and-white aerial photographs and the


Goggin (1952:40-41) considered the Mount Taylor Period
to include everything prior to the appearance of pottery.
Revisions in chronology have placed the Mount Taylor Period


descriptions of the site given by vari-
ous writers (see Figure 10). Attempts
have been made to assess the condi-
tion of the remaining portions of the
site and determine the present condi-
tion of the site components. The
vicinity of the larger burial mound
was located and some human bone
was noted on the surface. It would
appear that the "island" of midden
material contains some intact depos-
its, as does the area of the old "shell
fields." Much of the site, however,
has been mined down to the concreted
shell midden.


Numerous artifacts, typical of the
Mount Taylor and Orange periods
were recovered by Bullen (1955), and
additional examples are reported from
private collections by Neill (1954) and
Webster (1970). The Mount Taylor
Period levels of the site excavated by
Bullen (1955) produced Strombus
gigas celts, Busycon adze/gouge
blades, a fragment of a steatite
bannerstone, as well as fragments of
bone implements.
Sherds of St. Johns Plain and
fiber-tempered pottery were noted
during the C.A.RL. survey visit, as
were fragments of faunal bone and
broken marine shell tools.5

Mount Taylor Period

and Bluffton sites have been instrumen-
t of ideas about the pre-pottery Archaic
iver Basin and adjacent areas. Despite
e was known about the overall plan of the
e of Mount Taylor, information about
stratigraphy was limited. More recent
Taylor Period has focused on wet sites,
midden (8VO2601) at Lake Monroe
1994), and sites representing the coastal
The following discussion is intended to
About the Mount Taylor Period from a
h an emphasis on placing the Mount
ites in context.6


Figure 9. Looking to the north at the main trench, Trench A, excavated in the large
Bluffton shell mound. Ripley Bullen and workers at side in cut, Bill Edwards and
worker at top. Note the numerous strata exposed in the excavation. Collections of the
Anthropology Department of the Florida Museum of Natural History, Cat. No.

within the Middle and Late Archaic, dating from 6000 B.P. to
4000 B.P. Interestingly, Miller (1992:102) suggests that the
first occupation of the St. Johns River basin coincided with
rising sea levels between 6000 and 4000 years ago, which
prompted the formation of the river and its freshwater shellfish
resources. This date is consistent with radiocarbon determina-
tions for the lowest levels of 8VO2601, which represent the
early phases of the Mount Taylor Period (McGee and Wheeler
1994:342). Bullen (1958:110; 1960) fixes the end of the
Mount Taylor Period and the beginning of the Orange Period
at 4000 B.P with the first appearance of fiber-tempered
pottery.7 The appearance of fiber-tempered pottery at
8VO2601 around 4100 B.P. confirms this (McGee and
Wheeler 1994:342).8 It seems likely that more discreet
temporal subdivision within the Mount Taylor Period may be
possible with further work, especially at sites with wet site
Pre-pottery era sites on the Atlantic Coast have tended to
date toward the end of the Mount Taylor Period, though it is
possible that earlier sites are drowned or have been destroyed
through erosion. Coastal pre-pottery components are present

deposits in larger sites.

at the Tomoka Mound Complex
(Piatek 1994:114-115) and the
Douglass Beach site (Murphy
1990:27), both of which indicate
occupation around 4600 B.P. One
southern exception is the Windover
site, which dates from 8000 to 7000
B.P. and has distinctively Early Ar-
chaic projectile-point types (Doran
and Dickel 1988:264). Windover is
approximately 8 km from the coast
and 7.5 km from the St. Johns River
Basin. At the northern end of the
area, Russo (1992:110; 1993:10)
reports on a coastal site dated to 5210
B.P., confirming contemporary occu-
pation of the coast and river basin
during the earliest part of the Mount
Taylor Period.

Settlement Pattern

Little is known about Mount Tay-
lor Period settlement systems. Figure
11 illustrates sites with Mount Taylor
Period deposits. Note the concentra-
tion of sites along the upper reaches
of the St. Johns River. Sites with pre-
pottery deposits on the Atlantic Coast
and Oklawaha River drainage also are
included, along with some sites in the
Indian River area. Study of Mount
Taylor Period settlement remains a
major area for further research, with
a need to identify smaller sites of this
period, and better identify pre-pottery

Site Arrangement and Form

Habitation sites dating to the Mount Taylor Period have
several general configurations. A typical form is the ovoid or
elliptical midden-mound, like that observed at Mount Taylor
and Harris Creek (8VO24) (Aten 1999:135-136), and like that
described for the Bower's Bluff (8LA88) sites (Bullen and
Bryant 1965:5, 11). The Kimball Island site (8LA89) is a pair
of conjoined ovoid midden-mounds (Bullen and Bryant
1965:17-18). Sites of the Mount Taylor Period also appear as
ridges of shell midden (e.g., 8V07 or the "swamp shell ridge
near Morrison's Creek" described by Moore 1893:11-12; and
8VO214 onHontoon Island describedby Wyman 1875:26-27).
a general rise in water levels, may be responsible for the
occurrence of sites deep in river swamps. The convoluted
course of the St. Johns River lends itself to the development of
oxbows, and it is possible that these morphological changes in


2000 VOL. 53(2-3)

P rat, *


Figure 10. Plan map of Bluffton site complex, ca. 1942 (based on United States Department of Agriculture aerial photograph

the river are responsible for the above phenomenon. This is
supported by the existence of later, St. Johns Period midden-
mounds in swamps (e.g., Otter Mound in the extensive cypress
swamp to the south ofHontoon Island).
Multicomponent sites are often more complex, with shell
mounds, shell ridges, shell fields, as well as shell and sand
burial mounds, many of which were used in Mount Taylor,

Orange, and subsequent periods. Bluffton is a good example
of this type of site, which was occupied from the Mount Taylor
through St. Johns I periods. The Old Enterprise (8VO55) site
complex has a similar configuration, with a large shell mound,
adjacent shell ridge, and shell fields (Wyman 1875:19-20; see
Figure 12). The Harris Creek site (8VO24) at Tick Island also
has this configuration, with a large shell mound, shell fields,



Figure 11. Mount Taylor Period sites, and coastal Archaic sites mentioned in text (GIS map generated by Florida Master
Site File).-4

Site No.


Site Name
Hampton Lake No. 1
Oxeye Island
Pepper Island
McGundo Midden
Rollins Bird Sanctuary
Spencers Midden
Durbin Ridge
Crescent Beach
Mound 2 mi. N. of Palatka
Horse Landing
Rollestown Midden
Piney Bluff Landing
Salt Springs
Kauffman's Island
Ft McCoy Midden

Piney Island Midden
Stokes Feny Bridge
Silver Glen Springs
Duval Midden
Alexander Springs

Mound 1 mi. above Astor
Manhattan Landing
Ocala Forest No. 1
Ocala Forest No. 2
Kimball Island

Old Town/St Francis

Crows Bluff

Hawkinsville Shell Mound
Hitchens Creek Midden
Morrison's Creek Midden

shell ridges, and later St. Johns Period middens and sand
mounds (Aten 1999:134-136; Bushnell 1960; Moore
1893:605-607). What these different site configurations
represent is unclear, though similar features occur regularly at
different sites, suggesting some intentional construction and
shared ideas on village layout, refuse disposal, and mound

Concretion in Sites

The presence of concreted shell midden material is a
distinctive feature of many of the St. Johns River shell mounds
(Wyman 1875:38, 41). Concretion is often found over large
areas in the basal layers of Mount Taylor Period sites, though

Site No.


it is known at later sites as well. Bluffton and Mount Taylor
both are underlain by extensive deposits of concretion, as
indicated in the profiles collected at these sites (see Figure 3).
Palmer and Williams (1977) give an excellent explanation of
this phenomenon, noting that it is the result of natural chemi-
cal processes. They explain as follows:

water percolating through humus combines with detrital
carbon dioxide to form carbonic acid. This in turn dissolves
calcium carbonate shell in the upper part of the midden to
form calcium bicarbonate solution. As the solution perco-
lates downward, it comes in contact with more alkaline soil
and begins to precipitate calcium carbonate, which forms the
cementing matrix for the assorted large particles of shell
[Palmer and Williams 1977:25].

Site Name
Mount Taylor
Dexter Point
Harris Creek
DeLeon Springs Midden
Tomoka Park Midden
Oak Hill Midden
Thursby Midden
Hontoon Island Midden
Hontoon Island North Midden
Mount Joy
Good's Shell Pit
Lake Monroe Outlet Midden
Gemini Springs Midden
Old Enterprise/ Groves
Orange Midden
Stone Island Midden
Thornhill Lake Mound
Raulerson's No. 2 Mound
McDonald's Farm
Ginn's Grove
Buzzards Roost
Orange Mound
Mulberry Mound
Hull Island
Moccasin Island
Persimmon Mound
Duda Ranch Shell Mound
John D. Wright

Douglass Beach
Coontie Island


2000 VOL. 53(2-3)

Figure 11. Mount Taylor Period sites, and coastal Archaic sites mentioned in text (see caption facing page.)

La.~)u PredM ty fte he M Ste Re




Figure 12. Old Enterprise (8V055) site plan. This plan is based on information
from aerial photographs, the Sanford (U.S.G.S. 1988) and Osteen (U.S.G.S. 1980)
quadrangle maps, Piatek (1984), and Wyman (1875:19-21). 8V02601, the wet site
component of the site, also is shown.

Coastal andRiverine Expressions ofthe Mount Taylor Culture

The current discussion has focused on the riverine aspect
of the Mount Taylor culture, though direct and indirect
evidence indicates the existence of a similar and probably
related culture on the Atlantic coast. Traditionally, scholars
believed that the Atlantic coast had been largely unoccupied
prior to the beginning of the Late Archaic and the higher sea
level of that time, which would have created suitable ecosys-

teams for human exploitation (see discus-
sions of these ideas in Piatek 1994:111-
113 and Wheeler and McGee 1994b:373-
374). However, the presence of marine
products at riverine sites, like shells and
shark teeth, suggest that the Atlantic coast
was being exploited by people of the
Mount Taylor Period. Excavations earlier
in the century hinted that there were pre-
pottery components at coastal sites
(Goggin 1952:41; Nelson 1918:92-93).
More recent work has demonstrated that
there are sites and site components on the
coast dating to the pre-pottery era, includ-
ing the offshore Douglass Beach site
(8SL17), which had feature comprised of
wooden stakes dating to 4630 +/- 100
B.P., as well as shell and bone imple-
ments, and a Newnan point (Murphy
1990:35-39). Sites with pre-pottery de-
posits at Crescent Beach (Bond 1992:152)
and the Tomoka Mound Complex (Piatek
1994:115-117) produced artifacts very
similar to those of bone, marine shell, and
shark teeth known from the riverine
Mount Taylor Period sites. Ste. Claire
(1990:191-192) discusses several other
Atlantic coast sites that may date to the
pre-pottery Archaic, includingFutch Cove
(8BR181), Norris Mound (8BR89), and
McDonald Farms (8V02570).
Russo (1993) reports twenty sites in
the Timucuan Ecological Preserve that
have or may have pre-pottery compo-
nents.9 Radiocarbon dates from four of
these sites range from 4500 B.P. to 3730
B.P., and a date from Spencer's Midden
(8DU5623) is even older at 5210 B.P.
(Russo 1992:110; 1993:10). Artifacts
from some of these sites include Florida
Archaic Stemmed points and fired clay
objects, suggesting a technology related to
that of the riverine Mount Taylor culture.
Russo (1993:12) presents an argument
based on zooarchaeological data that
suggests some of these sites may have
been occupied year-round.

Material Culture

Detailed studies of Mount Taylor Period material culture
are available from the excavations at 8V02601, where
anaerobic preservation allowed the collection of a large sample
of well-preserved implements. The collection from 8V02601
and collections from other sites indicates that the material
culture of the St. Johns area was already well established early
in the Mount Taylor Period. It should be noted that Goggin


2000 VOL. 53(2-3)


Table 2. Material culture of the Mount Taylor Period.

Textile and Leather Working
* bone splinter awls
* bone awls
* deer ulna awls
* bone needles
* bone fids
* shell columella awls
* shell net mesh gauges
Wood and Bone Working
* bone splinter gouges
* bone splinter gravers
* bone gouges
* bone celts
* bone splinter burnishers
* shark tooth implements
* wooden handles for carving implements
Wood Working
* Busycon spp. shell adze/gouge
* Strombus gigas shell celts
* Busycon carica cutting-edge tools, type AX
* Busycon carica cutting-edge tools, type X
* wooden handles for hated shell tools
Lithic Working
* antler flakers

* b
* e;
* d
* de
* b
* sl
* sl
* w
* g
* Sc
* b
* w
* b
* F
* b
* B
* w
* wv
* d
* c:

had very little excavated material to work with from this
period, which explains the omission of some notable artifact
types from his description of the Mount Taylor Period (Goggin
Shell andBone Technology. The quality of preservation at
8VO2601 allowed for the recovery of a large and well-pre-
served collection of bone and shell artifacts. Analysis of
formal qualities and wear patterns indicates that many of the
bone artifacts were weaving tools, involved in the production
of mats and textiles (Wheeler and McGee 1994a). Information
from the burials at Windover confirm an extensive Archaic-era
textile tradition, and the excavators indicate that there are
seven different textile twining/weaving variants present at the
site (Doran and Dickel 1988:274). Many of the other tools,
especially those of shell, appear to have been involved in
woodworking, and this is confirmed by the shell tool marks
found on many pieces of wood at 8VO2601 as well as the large
number of wood chips found in the site (Figure 13).
Chipped Stone Artifacts. Projectile points recovered from
Mount Taylor Period sites are predominantly Florida Archaic
Stemmed types (cf. Jahn and Bullen 1978:Figures 55-59).
Clausen (1964:20-22) compares the stemmed projectile point
types of North Central Florida with those of the Mount Taylor
culture, noting the essential similarity of the types. This is not
surprising when one considers that sites with Mount Taylor
deposits extend well to the west of the St. Johns River. Purdy
(1994:390-392) notes that lithic material was probably a
valued commodity used in performing certain tasks, but was

not a major element of
Mount Taylor Period
nal Adornment technology. Thermal
one pins/fasteners alteration of chert was
expanding head bone pins practiced, and is indi-
ecorated bone cated by a pink or red-
ecorated antler dish coloration or a
one beads glossy surface. Often the
hell beads Archaic stemmed points
lell pins recovered from Mount
'ooden pins Taylor Period sites are
roundstone beads difficult to classify,
ing/Fishing though they seem to ad-
ocketed antler points here to some standard
onebipoints forms (cf. Aten
'ooden net floats 1999:152-154, Fig. 12).
annerstones/atlatl weights This may indicate that
lorida Archaic Stemmed points there is a regional
Preparation chipped-stone tradition,
aked clay cooking objects consisting of as yet un-
usycon spp. shell receptacles named types. The tem-
ooden stirring paddles poral significance of
wooden bowls? these unnamed variants
sportation may be useful in refining
dugout canoes the chronology of the
ugout canoes
anoc paddles Mount Taylor Period.
Woodworking. Many
of the shell and bone ar-
tifacts mentioned above
are indirect evidence of woodworking technology. These tools
and others have equivalents in a modern carpenter's tool kit.
Busycon cutting-edge tools and Strombus celts are ax analogs,
used in heavy wood cutting; evidence for this was found at
8VO2601 where the ends of planks had been cut with a
Busycon cutting-edge tool (Wheeler and McGee 1994b:386).
While saw blades were not recovered at 8VO2601, wood that
had been cut with a saw was recovered (Wheeler and McGee
1994b:386). Drills also were present, although no wood
showed any evidence of having been drilling. Shell and bone
gouges and adzes are used in the same manner as their modern
metal counterparts, as are bone or shell chisels. Chert and
shark tooth knives were used much the same way that chisels
and rasps are used to shape and smooth wood during final
finishing. Other finishing tools used to shape and smooth
wood are coquina blocks (as rasps), sandstone (for sanding),
and limestone (for fine sanding). One other implement
necessary in the type of woodworking represented at 8VO2601
was a chert wedge (Purdy 1994:391).'0
The primary woodworking method identified at 8V02601
was the production of planks or billets of wood. Planks were
produced by cutting a groove across a log at either end of the
piece to be removed. The plank was then split off using a
wedge or series of wedges-the valuable chert wedge was
probably used to start the split, and then wooden or antler
wedges were used to complete the process. Many of the
wooden artifacts from 8V02601 were produced by this
method, including planks, a canoe paddle, stirring paddles,



0 1 2 3
I m cm

Figure 13. Shell artifacts from 8V02601: a) Busycon carica type X cutting-edge tool; b) Busycon carica type AX cutting-edge
tool; c) Busycon carica type AX cutting-edge tool; d-e) Busycon adze/gouge tools (from Wheeler and McGee 1994a:Figures
18,20,22). a-c) arrows indicate hafting holes; d-e) arrows indicate bit/blade end. Along with Strombusgigas celts, these shell
tools are specialized wood working implements, characteristic of the Mount Taylor Period.

tool handles, and other straight-grained wooden artifacts
(Wheeler and McGee 1994b:380-382; Figure 14). Isolated
finds of dugout canoes or wooden containers dating to the
Mount Taylor Period are more direct indicators of the kinds of
wood products made and used." It should be emphasized that
woodworking in the Mount Taylor Period was not restricted to
the use and modification of radial branches and poles.
Virtually all of the native woodworking tools, techniques, and
products that were in use at the time of European contact had

been developed during this early era.
Baked Clay Cooking Objects. Fragmentary objects of clay
are frequently encountered in Mount Taylor Period sites, but
often have been overlooked by excavators. Excavations at
8V02601 produced a large quantity of partially complete clay
objects of various forms (Wheeler and McGee 1994a:369-
372;see Jahn and Bullen 1978:Figures 21-22 for an illustration
of the range of forms encountered at Tick Island; see Figure
15). This type of object is known from other areas, and their

2000 VOL. 53(2-3)



- -2

\ -- *.**- ...J .^^ ^

0 1 2 3


Figure 14. Wooden artifacts from 8V02601: a) tenoned tool-handle of radial branch; b) tool handle of cross-cut wood;
c) incised and tenoned tool-handle of cross-cut wood; d-e) notched net floats (from Wheeler and McGee 1994b:Figures
1, 2, 4).






5 cm

Figure 15. Baked clay object forms. Examples of typical forms encountered at 8V02601 and other St. Johns area sites: a)
grooved-cylinder; b) biconical; c) biconical-grooved; d) biscuit-shaped.

function has been speculated upon (Bunn 1974; Sassaman
1993:130-135; Small 1966; South 1970:6-7). McGee
(1995:79-102) undertook a systematic technological study of
these objects and concluded that while they were suitable for
use in earth ovens and shellfish steaming pits, they probably
had been used in boiling freshwater snails in wooden bowls."2
McGee's study indicated that the various shapes were used for
different heating tasks (e.g., grooved cylinders and biscuit
forms brought water to a boil rapidly, then cooled quickly,
while biconical forms released heat slowly). Coupled with the
use of heat in improving chert, this information suggests that
the people of the Mount Taylor culture possessed a sophisti-
cated knowledge of pyrotechnology, which undoubtedly was
important in the later development of pottery.


Cumbaa (1976) was the first to quantify the dietary
importance of shellfish in the Archaic of the St. Johns area.
Data from early sites were lacking and zooarchaeological
techniques were unrefined, resulting in the conclusion that,
despite the large numbers of freshwater snails in mounds, they
made only a small contribution to the diet (Cumbaa 1976).
More recently Russo et al. (1992:104) used zooarchaeological
evidence from late Mount Taylor/early Orange period deposits
at 8VO2601 to demonstrate that shellfish represented 98 % of
the dietary meat weight, with very small contributions by fish,
mammals, and turtles. Further research at 8VO2601, which
utilized larger samples, indicated that freshwater shellfish
contributed from 33 % to 87 % of dietary meat weight in
earlier Mount Taylor Period levels, with larger contributions
from fish and other aquatic vertebrates (Wheeler and McGee
Wet site preservation at 8VO2601 allowed for the only
archaeobotanical analysis of materials from a Mount Taylor
Period site (Newsom 1994). Newsom (1994:406-407) identi-
fied a diverse array of plant remains from the site, including
edible and other useful species. Edible nut types include

hickory, live oak, and swamp white oak, as well as seed and
wood remains from plants that bear pulpy, edible fruit,
including black gum, prickly pear cactus, saw palmetto,
maypop, wild plum, blackberry, persimmon, red mulberry,
elderberry, and wild grape. Also found in the analysis are
species with small starchy seeds, fresh greens, tubers, and
other miscellaneous edible plants and fungi. Pokeweed,
toothache tree, and other related species may have been used
for medicinal purposes, and native cane and gourds would
have been used for utilitarian items, such as housing material
(cane), and containers, net floats, and rattles (gourds). As with
other components of subsistence and technology, these results
demonstrate that people of the Mount Taylor Period were well
adapted to their environment and very familiar with a variety
of plant resources, even during the earliest phases of the

Exchange Networks

Peoples of the Mount Taylor Period were evidently in-
volved in long-distance exchange networks that involved other
groups throughout Florida, as well as cultures in other parts of
the Southeast. These exchange networks were a mechanism
for the redistribution of raw materials, as well as finished
ornaments and tools.
Implements made from the strong lip of the Strombusgigas
shells are evidence of contact and exchange with peoples of
southeastern Florida, since the geographic range of this
mollusk is restricted to the lower Atlantic Coast and Florida
Keys (Keegan 1982:81-82; Wheeler 1993). Strombus gigas
celts were found at Mount Taylor by Moore (see Table 1) and
during the C.A.R.L. Archaeological Survey visit. This tool
type also is known from 8VO2601 on Lake Monroe (Wheeler
and McGee 1994a:363, Figure 19), Bluffton (Bullen 1955:4,
11-12), and Tick Island (Jahn and Bullen 1978:Figure 43j-n),
to name a few. It persists through the Orange Period in the St.
Johns area, but then is rarely found outside of southeastern
Florida (Keegan 1982:81-82; Wheeler 1993). The unusual


2000 VOL. 53(2-3)


Figure 16. Ground-stone beads, Coontie Island (8SJ13). Photograph from the John M.
Goggin Papers, courtesy of the Manuscript Collection, George A. Smathers Libraries,

University of Florida.

bannerstone made of the Strombus gigas lip from Mount
Taylor is interesting, as it represents an exotic form from the
north executed in an exotic media from the south (Moore
Ground-stone artifacts, like celts, bannerstones, pendants,
and beads, are indicative of exchange with groups outside of
Florida (Jefferies 1996). These types of items occur at a
number of riverine Mount Taylor Period sites, including
Bluffton (Neill 1954:13-15), Tick Island (Jahn and Bullen
1978:Figures 42g, 46), and Thornhill Lake Mounds (Moore
1894a:167-173), as well as pre-pottery-era coastal sites like
Coontie Island and the Tomoka Mound Complex (Clausen
1964:28, 36-37, Figures 14-15; Piatek 1994:111-112). Figures
16 and 17 illustrates some of the ground stone artifacts from
Coontie Island (8SJ13).
These exchange networks help explain the similarity of
decorated bone artifacts throughout Florida, where similar
designs are found throughout the state (Wheeler 1994). It also
is likely that these networks were instrumental in the spread of
technology, including the fiber-tempered pottery of the Late
Archaic. It is interesting to note that exotic goods, including
artifacts of Strombus gigas and the ground stone items, are
absent from the Early Archaic Windover site, despite other
artifactual and stylistic similarities with the later Middle and
Late Archaic Mount Taylor cultures. Aten (1999:177) notes
that no ground stone artifacts were recovered at the Harris
Creek site (8V024) on Tick Island. This may mean that these

regional and long-distance ex-
change networks had not yet de-
veloped in the Early Archaic, but
came into being sometime early
in the Mount Taylor Period.

Decorative Arts

Carved and incised examples
of bone and antler give the best
picture of Mount Taylor Period
art (Figure 18). These objects,
which include pendants, pins,
and hair ornaments, are similar
in form to those from later peri-
ods. The typical designs, how-
ever, are distinct from those of
later periods, and are easily dis-
tinguished from the decorated
bone work of the subsequent Or-
ange Period (Wheeler 1994:47-
49). Interestingly, the designs
found in the Mount Taylor Pe-
riod are shared by many other
pre-pottery Archaic cultures of
Florida, and suggest a broad ar-
tistic tradition throughout the

state (cf. decorated antler from
Republic Groves in Wharton et
al. 1981:66, Figures 12-13).
This widespread aesthetic tradition may be a counterpart to the
equally widespread distribution of Florida Archaic Stemmed
points, as noted above." Typical designs are organized around
diamond motifs, and often incorporate nested chevrons, nested
triangles, and nested rectangles. Hatching and cross-hatching
occur as primary motifs, or are used to highlight other design
elements. The designs are usually symmetrical and very
formal, in contrast to the flamboyant designs associated with
the decorated ceramics and bone work of the Orange Period
and known especially from Tick Island (see examples in Jahn
and Bullen 1978).
Other decorative objects known from the Mount Taylor
Period include marine shell disk beads, tubular beads cut from
bird bone, and undecorated bone and shell pins. Decorated
and undecorated bone pins were probably used as hair or
clothing ornaments, and wear patterns on these artifacts are
consistent with this interpretation (Wheeler and McGee
1994c:354, 358). One group of bird bone beads recovered
from 8VO2601 had been strung on a fine twisted and knotted
cordage, with knots separating individual beads (Wheeler and
McGee 1994c:358, Figure 12). Bone pins and shell beads form
the basis for decorative systems of later cultures in the St.
Johns area, as well as many other Florida and Southeastern
cultural traditions.



ated with a village midden
nearby, and some included burial
goods like tubular shell beads,
bone implements, decorated ant-
ler hair ornaments, as well as
atlatl weights and hooks (Jones
and Carr 1981:83-84). The
Gauthier cemetery material has
not been thoroughly analyzed,
but may parallel the Harris Creek
Mound Building. There
would appear to be increasing
evidence for the Archaic-era con-
struction of sand burial mounds.
Moore (1894a: 167-173) describes
excavations at the Thornhill Lake
mound group, which produced a
number of burials and associated
ground stone implements and
Ornaments, including
AiWs .r --, AG Tl : bannerstones, miniature
a2 3 4 bannerstones, catlinite (or red
jasper) and shell beads, and pen-
dants made from broken
Figure 17. Ground-stone bannerstones, Coontie Island (8SJ13). Photograph from the John bannerstone wings. The burial
M. Goggin Papers, courtesy of the Manuscript Collection, George A. Smathers Libraries, mound excavated by Moore was
University of Florida. built of brown sand, atop a shell
midden with the typical concre-
tion layer encountered in this
Mortuary Patterns region (Moore 1894a:88). Mitchem (1999:30) indicates that
Thornhill Lake may represent a Mount Taylor Period site.
Mortuary patterns of the Early and Middle Archaic exhibit Piatek (1994) reports on pre-pottery Archaic earthen burial
some variation, indicating that several burial traditions were mounds at the Tomoka Mound Group, a coastal site that has
being practiced, or that there was some experimentation produced ground stone bannerstones and pendants. Recent
regarding burial. On a broader scale it appears that there were work at the site has led Piatek (1994:115-117) to conclude that
regional or subregional mortuary traditions, the site was occupied by pre-pottery Archaic peoples ca. 4060
Shell Mound Burials. Cemetery-style interment in pre- B.P., who were well adapted to the coastal region and interred
pared sand platforms on shell mounds also is characteristic of their dead in earthen burial mounds. Aten (1999:177) suggests
the Mount Taylor culture. Bullen (1962; Aten 1999; Jahn and that the Tomoka Mounds may represent a later form of burial
Bullen 1978:20-22; Russo 1994:94-96) excavated 175 burials that succeeds the shell mound cemeteries, like those at Harris
that were made in two distinct mortuaries on the platform of Creek (8VO24) and Orange Mound (80R1).
the Mount Taylor Period shell mound at the Harris Creek site Mortuary Ponds and the Middle Archaic. Cemetery-style
(8VO24) on Tick Island. Dates from the site indicate the interment in shallow ponds seems to be an ancient form of
interments were made around 5400 B.P. (Aten 1999:Table 4). burial that occurs in the Early and Middle Archaic, primarily
Aten (1999) gives a detailed description of the burials, associ- in southern Florida. Windover, the earliest known mortuary
ated features, and details of mound construction, providing the pond, is located in northern Brevard County not far from the
most extensive portrait of Mount Taylor Period mortuary east coast nor the Indian River, and just to the south of the St.
behavior. Moore (1893:617-623; see Bullen 1962:480) Johns area (Doran and Dickel 1988). Burials at this site
describes a similar Mount Taylor Period shell mound cemetery produced artifacts similar to those described for the Mount
at Orange Mound, and argues rather convincingly that the Taylor culture, and it seems likely that this site is a precursor
burials encountered were made in a prepared earthen mound, to the coastal aspect of the Mount Taylor culture. Middle
which later became encased in the larger shell-heap. Archaic mortuary ponds are present in southern Florida at Bay
Gauthier (8BR193) is another cemetery site dating to the West (Beriault et al. 1981), Republic Groves (Wharton et al.
Middle and Late Archaic, essentially on the St. Johns River 1981), and Ryder Pond (FMSF 1998). While elements of
and approximately 18.5 km from the coast (Jones and Carr material culture and style of interment seems to be similar
1981). Burials at this site were made in a sandy ridge associ- among the Mount Taylor and other Middle Archaic sites, the


2000 VOL. 53(2-3)



' e

0 1 2 3
I cm

0 1 2 3 4 5cm

Figure 18. Decorated bone and antler: a) carved and incised antler hair ornament, Gauthier; b-c) incised bird bone tubes,
Windover; d-h) decorated bone artifact fragments, 8V02601 (see Wheeler 1994; Wheeler and McGee 1994b).

contrast between pond burial and burial in sandy ridges or
shell mounds is striking. It may be that there is a geographic
distinction between the mortuary patterns of the Archaic
populations of the St. Johns area and southern Florida, though
there would appear to be some experimentation regarding
mortuary patterns during this era.


Origins of the Mount Taylor Culture

Scholars had originally conceived of an interior, uplands
origin for the Mount Taylor culture, with hunter-gatherer
populations of north-central Florida seasonally occupying
riverine areas (Milanich and Fairbanks 1980:150).14 More
recent research has demonstrated that adequate resources were

present in the St. Johns River basin and adjacent coastal region
to support local populations by 6000 years ago. Evidence of
well-adapted technological and subsistence systems suggest
that these early populations comprised a primarily aquatic-
based culture (Russo et al. 1992; Wheeler and McGee 1994b).
Exchange networks involving shells from southern Florida and
groundstone artifacts from other parts of the Southeast, a
widespread tradition of decorated bone and antler, as well as
the construction of earthen burial mounds, points toward a
complex culture, comparable in many ways to subsequent
developments in the region. Changes in sea level have
drowned coastal Mount Taylor Period sites, and may have
seriously altered or destroyed many others. The early sites
reported by Russo (1993:10-12, 71-72, 91-95, 98-106) from
near the mouth of the St. Johns River may well have been
spared from destruction because of their locations in the river




marshes, away from high-energy environments immediately
adjacent to the Atlantic ocean. Artifact assemblages from
these sites, and other pre-pottery coastal sites, appear to be
identical to those found at the riverine sites. As early dates are
obtained from coastal sites, we may find that the riverine
Mount Taylor culture, which relied on shell and shark-tooth
tools and other marine products, originated on the Atlantic

Significance of Mount Taylor and Bluffton

The Mount Taylor and Bluffton sites are significant for
their role in our understanding of the Archaic era in the St.
Johns River basin. Moore's (1893) work at Mount Taylor
provided the framework for Goggin's (1952) formulation of the
pre-pottery Archaic in the northern St. Johns area. Likewise,
Bullen's (1955) excavation at Bluffton provided a stratigraphic
portrait of both the Mount Taylor and Orange periods. Both
sites have suffered from shell mining. Current work at these
sites has, however, indicated that significant deposits are still
intact and may be useful in further research on the Mount
Taylor and Orange periods. The discovery of wet site deposits
at Mount Taylor is particularly significant, since similar
deposits at other sites have generated a far greater spectrum of
material remains than their dry land counterparts.

Future Research

There has been a recent revival in interest in the Mount
Taylor Period and the pre-pottery cultures of the St. Johns
River and Atlantic Coast; there are, however, a number of
significant research topics that remain to be pursued. Notable
directions for further research include analysis of settlement
patterns, a refinement of the pre-pottery Archaic chronology,
a better understanding of the relationship of coastal and
riverine cultures, as well as study of the broader relationships
with other Archaic peoples in Florida and the Southeast.


' Mount Taylor was listed on the National Register of Historic
Places as of October 8, 1997.
2 Now the National Museum of the American Indian, Smithsonian
SA 2.75" sand bucket auger was used in this testing.
4 Apparently the original collection was housed at the Academy of
Natural Sciences of Philadelphia. Before this collection was
removed to the Heye Foundation-Museum of the American Indian
in New York, a group of objects was loaned or traded to the Uni-
versity of Pennsylvania Museum. Goggin did not study this mate-
rial when he examined other University of Pennsylvania Museum
collections in the 1940s. The senior author examined this collec-
tion in July, 1998.
5 During the forest fires of 1998 a small portion of the southern end
of Bluffton was damaged by a fireplow. Inspection of this area
revealed sand-tempered plain and St. Johns Plain sherds, fragments
of marine shell, the base of a Florida Archaic Stemmed point, and
a fragment of a decorated bone pin typical of the Orange Period (cf.
Jahn and Bullen 1978:Figure 34o).
6 Much of this discussion was formulated in conjunction with

research at 8V02601 by authors Wheeler and McGee and accord-
ingly this site figures prominently in the text.
7 Interestingly the dates for the beginning and end of the Mount
Taylor Period have largely been based on dates for contemporane-
ous cultures in other parts of the Southeast. The dates from
8VO2601 tend to support these previously accepted dates.
8 All radiocarbon dates discussed in this article are uncorrected,
uncalibrated dates.
9 Only some of these sites are shown in Figure 11.
10 We called this artifact a "handaxe" because of its shape.
" The oldest Florida canoes are from DeLeon Springs, along the St.
Johns River, dated to 5140 +/- 100 B.P. and 6050 +/- 60 B.P.
(Hartmann 1996:93; Purdy 1991:270).
12 Presumably these cooking objects could have been used for any
task that required boiling water.
'1 Jefferies (1997) reaches similar conclusions regarding interaction
during late Middle Archaic (6000 B.P.-5000 B.P.) in the southern
Midwest after examining bone pins from a number of sites.
" These interpretations are largely based on Cumbaa's (1976)
thoughts on shellfish in the Archaic diet.


The authors would like to thank the following individuals, who
have contributed their time and knowledge to this project. Cathy
Lowenstein and Mike Parker, of the Florida Division of Forestry,
participated in the survey of Lake George State Forest and showed
us around Mount Taylor and Bluffton. John Williams, of the
Volusia Anthropological Society, also participated in the survey.
Melissa Elsberry, Assistant Keeper of the American Section Col-
lections, University of Pennsylvania Museum of Archaeology and
Anthropology, helped in accessing C. B. Moore's Mount Taylor
collection. Elise LeCompte, Registrar of the Anthropology Depart-
ment, Florida Museum of Natural History, provided access to
photographs of Bluffton, and Joyce Dewsbury, Photo Archivist at
the Department of Special Collections, George A. Smathers Librar-
ies, University of Florida, aided in obtaining copies of the Coontie
Island artifacts. Lawrence E. Aten provided a copy of his manu-
script on Ripley Bullen's work at the Harris Creek site on Tick
Island. Chip Birdsong, of the Florida Master Site File, did an
excellent job preparing the map of Mount Taylor Period sites.
Alan Nelson, of the Florida Bureau of Archaeological Research,
helped in improving one of the site maps. Thanks also are due to
Donna Ruhl, Bob Austin, and the anonymous reviewers for provid-
ing comments on this paper.

References Cited

Aten, Lawrence E.
1999 Middle Archaic Ceremonialism at Tick Island, Florida:
Ripley P. Bullen's 1961Excavation at the Harris Creek
Site. The Florida Anthropologist 52:131-200.
Beriault, John, Robert Carr, Jerry Stipp, Richard Johnson, and Jack
1981 The Archeological Salvage of the Bay West Site, Collier
County, Florida. The Florida Anthropologist 34:39-58.
Bond, Stanley C., Jr.
1992 Archaeological Excavations at 8SJ42, the Crescent Beach
Site, St. Johns County, Florida. The Florida Anthropolo-
gist 45(2):148-161.
Bullen, Ripley P.
1955 Stratigraphic Tests at Bluffton, Volusia County, Florida.
The Florida Anthropologist 8:1-16.
1958 More Radiocarbon Dates and Their Significance. The

2000 VOL. 53(2-3)



Florida Anthropologist 9:97-109.
1959 The Transitional Period. Southeastern Archaeological
Conference Newsletter 6:43-62.
1962 Indian Burials at Tick Island. Year Book of the American
Philosophical Society, 1961. Pp. 477-480.
1975 A Guide to the Identification ofFlorida Projectile Points.
2nd edition. Kendall Books, Gainesville.
Bullen, Ripley P., and William J. Bryant
1965 Three Archaic Sites in the Ocala National Forest,
Florida. Report No. 6. The William L. Bryant
Foundation, Orlando.
Bunn, Jennings W., Jr.
1974 Clay Balls: Ceremonial or Utilitarian? The Florida An-
thropologist 27:4748.
Bushnell, Francis F.
1960 The Harris Creek Site, Tick Island, Volusia County. The
Florida Anthropologist 13:25-31.
Clausen, Carl J.
1964 The A-356 Site and the Florida Archaic. Master's thesis,
Department of Anthropology, University of Florida,
Cumbaa, Stephen L.
1976 A Reconsideration of Freshwater Shellfish Exploitation
in the Florida Archaic. The Florida Anthropologist
Doran, Glen H., and David N. Dickel
1988 Multidisciplinary Investigations at the Windover Site. In
Wet Site Archaeology, edited by Barbara A. Purdy, pp.
263-289. Telford Press, Caldwell, New Jersey.
Goggin, John M.
1947 A Preliminary Definition of Archaeological Areas and
Periods in Florida. American Antiquity 13(2):114-127.
1948 Florida Archeology and Recent Ecological Changes.
Journal of the Washington Academy of Sciences
1949 Cultural Traditions in Florida Prehistory. In The Florida
Indian and His Neighbors, edited by John W. Griffin, pp.
13-44. Inter-American Center, Rollins College, Winter
1952 Space and Time Perspective in Northern St. Johns Arche-
ology, Florida. Publications in Anthropology 47. Yale
University Press, New Haven.
Florida Master Site File
n.d. Site form for Mount Taylor (8VO19). On file, Florida
Master Site File, Bureau of Archaeological Research,
Florida Division of Historical Resources, Tallahassee.
1998 Site form for Ryder Pond (8LL1850). On file, Florida
Master Site File, Bureau of Archaeological Research,
Florida Division of Historical Resources, Tallahassee.
Hartmann, Mark J.
1996 Development of Watercraft in the Prehistoric Southeast-
ern United States. Ph.D. dissertation, Department of
Anthropology, Texas A&M University, College Station.
Jahn, Otto L., and Ripley P. Bullen
1978 The Tick Island Site, St. Johns River, Florida, edited by
Adelaide K. Bullen and Jerald T. Milanich. Florida An-
thropological Society Publication 10, Gainesville.
Jefferies, Richard W.
1996 The Emergence of Long-Distance Exchange Networks in
the Southeastern United States. In Archaeology of the
Mid-Holocene Southeast, edited by Kenneth E. Sassaman
and David G. Anderson, pp. 222-234. University Press of
Florida, Gainesville.

1997 Middle Archaic Bone Pins: Evidence of Mid-Holocene
Regional-Scale Social Groups in the Southern Midwest.
American Antiquity 62(3):464-487.
Jones, B. Calvin, and Robert S. Carr
1981 Florida Anthropologist Interview with Calvin Jones Part
II: Excavations of an Archaic Cemetery in Cocoa Beach,
Florida. The Florida Anthropologist 34(2):81-89.
Keegan, William F.
1982 A Biological Introduction to the Prehistoric Procurement
of the Strombus gigas. The Florida Anthropologist
Le Baron, J. Francis
1884 Prehistoric Remains in Florida. In Smithsonian Institu-
tion Annual Report for 1882, pp. 771-790, Washington,
McGee, Ray M.
1995 Fired-Clay Objects: Cooking Technology from the
Florida Shell Mound Archaic at Groves Orange Midden
(8V02601) Florida. Master's thesis, Department of
Anthropology, University of Florida, Gainesville.
McGee, Ray M., and Ryan J. Wheeler
1994 Stratigraphic Excavations at Groves' Orange Midden,
Lake Monroe, Volusia County, Florida: Methodology and
Results. The Florida Anthropologist 47(4):333-349.
Milanich, Jerald T.
1994 Archaeology ofPrecolumbian Florida. University Press
of Florida, Gainesville.
Milanich, Jerald T., and Charles H. Fairbanks
1980 Florida Archaeology. Academic Press, New York.
Miller, James J.
1992 Effects of Environmental Changes on Late Archaic Peo-
ple of Northeast Florida. The Florida Anthropologist
Mitchem, Jeffrey M.
1999 Introduction: Clarence B. Moore's Research in East
Florida, 1873-1896. In The East Florida Expeditions of
Clarence Bloomfield Moore. University of Alabama
Press, Tuscaloosa.
Moore, Clarence B.
1893 Certain Shell Heaps of the St. John's River, Florida,
Hitherto Unexplored. American Naturalist 27:8-13, 113-
117, 605-624, 709-733.
1894a Certain Sand Mounds of the St. Johns River, Florida.
Journal of the Academy of Natural Sciences ofPhiladel-
phia 10(Parts 1 and 2).
1894b Certain Shell Heaps of the St. John's River, Hitherto
Unexplored. American Naturalist 28:15-26.
1898 Recent Acquisitions: Shell "Banner Stone" from the St.
John's River, Florida. Journal of the Academy of Natural
Sciences ofPhiladelphia 11:185-188.
Murphy, Larry E.
1990 8SL17: Natural Site-Formation Processes of a Multiple
Component Underwater Site in Florida. Professional
Paper No. 39. Southwest Cultural Resources Center,
National Park Service, Santa Fe.
Neill, Wilfred T.
1954 Artifacts from the Bluffton Midden, Volusia County,
Florida. The Florida Anthropologist 7:11-17.
Nelson, Nels C.
1918 Chronology in Florida. American Museum of Natural
History, Anthropological Papers 22(2):75-103. New
Newsom, Lee




1994 Archaeobotanical Data from Groves' Orange Midden
(8VO2601), Volusia County, Florida. The Florida An-
thropologist 47:404-417.
Palmer, Jay, and J. Raymond Williams
1977 The Formation of Goethite and Calcareous Lenses in
Shell Middens in Florida. The Florida Anthropologist
Piatek, Bruce J.
1984 Preliminary Assessment of the "Enterprize Midden" Site
8Vo55-Enterprize, Florida. Typescript and maps on
file, Florida Master Site File, Division of Historical
Resources, Tallahassee.
1994 The Tomoka Mound Complex in Northeast Florida.
Southeastern Archaeology 13(2):109-118.
Purdy, Barbara A.
1991 The Art and Archaeology ofFlorida's Wetlands. CRC
Press, Boca Raton, Florida.
1994 The Chipped Stone Tool Industry at Groves' Orange
Midden (8VO2601), Volusia County, Florida. The
Florida Anthropologist 47(4):390-392.
Rouse, Irving
1951 A Survey ofIndian River Archeology, Florida. Publica-
tions in Anthropology No. 44, Yale University Press,
New Haven.
Russo, Michael
1992 Chronologies and Cultures of the St. Marys Region of
Northeast Florida and Southeast Georgia. The Florida
Anthropologist 45(2):107-126.
1993 The Timucuan Ecological and Historic Preserve Phase
IIIFinal Report. National Park Service, Southeast Arche-
ological Center, Tallahassee.
1994 Why We Don't Believe in Archaic Ceremonial Mounds
and Why We Should: The Case from Florida. Southeast-
ern Archaeology 13(2):93-109.
Russo, Michael, Barbara A. Purdy, Lee A. Newsom, and Ray M.
1992 A Reinterpretation of Late Archaic Adaptations in
Central-East Florida: Groves' Orange Midden
(8VO2601). Southeastern Archaeology 11(2):95-108.
Sassaman, Kenneth E.
1993 Early Pottery in the Southeast: Tradition and Innovation
in Cooking Technology. University of Alabama Press,
Sears, William H.
1960 The Bluffton Burial Mound. The Florida Anthropologist
Small, James F.
1970 Poverty Point Baked Clay Objects. The Florida Anthro-
pologist 19:65-76.
South, Stanley
1970 Baked Clay Objects from the Site of the 1670 Settlement
at Charles Towne, South Carolina. The Institute ofAr-
cheology and Anthropology: Notebook 2(1):3-16.
Ste. Clarie, Dana
1990 The Archaic in East Florida: Archaeological Evidence
from Early Coastal Adaptations. The Florida Anthropolo-
gist 43:189-197.
United States Department of Agriculture
1942 CZC-1C-38, dated 3-10-42, vicinity ofBluffton. On file,
Map and Imagery Library, University of Florida,
United States Geological Survey
1980 Osteen 7.5" quadrangle map. Reston, Virginia.

1988 Sanford 7.5" quadrangle map. Reston, Virginia.
Webster, William J.
1970 A New Concept for the Busycon Shell Receptacle. The
Florida Anthropologist 23:1-7.
Wharton, Barry R., George R. Ballo, and Mitchell E. Hope
1981 The Republic Groves Site, Hardee County, Florida. The
Florida Anthropologist 34:59-80.
Wheeler, Ryan J.
1993 Spatial and Temporal Distribution of Shell Tools from
the East Okeechobee Area. Paper presented at the 501
annual meeting of the Southeastern Archaeological Con-
ference, Raleigh.
1994 Early Florida Decorated Bone Artifacts: Style and Aes-
thetics from Paleo-Indian through Archaic. The Florida
Anthropologist 47(1):47-60.
Wheeler, Ryan J., and Ray M. McGee
1994a Technology of Mount Taylor Period Occupation, Groves'
Orange Midden (8VO2601), Volusia County, Florida.
The Florida Anthropologist 47(4):350-379.
1994b Wooden Artifacts from Groves' Orange Midden. The
Florida Anthropologist 47(4):380-389.
1994c Report of Preliminary Zooarchaeological Analysis:
Groves' Orange Midden. The Florida Anthropologist
Wheeler, Ryan J., and Christine L. Newman
1997 An Assessment of Cultural Resources at the Lake George
State Forest, including Mount Taylor and the Bluffton
Site. C.A.R.L. Archaeological Survey, Bureau of Archae-
ological Research, Florida Division of Historical Re-
sources, Tallahassee.
Wyman, Jeffries
1875 Fresh-Water Shell Mounds of the St. John's River,
Florida. Memoir 1, No. 4. Peabody Academy of Science,
Salem, Massachusetts.


2000 VOL. 53(2-3)



Florida Museum of Natural History, University of Florida, Gainesville, FL 32611

Most human-induced changes in natural or pristine
ecosystems of the Americas are thought to have occurred as a
result of European exploration and colonization. Twenty years
of interdisciplinary work in environmental archaeology is
modifying our understanding of the role and the degree to
which pre-Columbian peoples modified their environments.
This work suggests that human-induced changes may not be
a consequence of cultural complexity, but rather of the
behavior of the human species in general. Evidence from
Pacific and Caribbean islands shows that people dramati-
cally altered plant and animal communities along with the
landscape before A.D. 1492 (Lepofsky et al. 1996).
Extinction and extirpation of species represent the long-
term effects of human alteration of these biotic communi-
ties. Although less severe, demographic changes in
animal populations also seem to have occurred.
Steadman's (1995) work on tropical Pacific Islands
documents the prehistoric extinction of an estimated 2000
species of birds representing 20% of the world's avifauna.
This occurred island by island as people colonized
Oceania. In Polynesia most of the extinctions occurred in
the past 3000 years (Steadman 1995:1124). In western
Polynesia, bats and lizards accompany the extinction of
the birds (Steadman 1995:1124). Human predation,
habitat loss, introduction of predators, competitors, or
pathogens are the likely causes of the loss of so many
species (Steadman 1995). Since birds are plant
pollinators, seed dispersers, and control insects, habitat
degradation was further exacerbated by their extinction.
Wing (1999) reports changes in terrestrial and marine
animals in the zooarchaeological records of some islands
of the Lesser Antilles and the Virgin Islands. Through
time there appears to have been a shift in animal use from
terrestrial crabs to marine mollusks and terrestrial verte-
brates. The decline in the numbers of land crabs al
(Geocarcinidae) is accompanied by an increase in the
frequency of hermit crabs (Coenobiidae). At the same time,
carnivorous reef fishes decline in frequency, while reef
omnivores and herbivores increase. There also is a clear
reduction in the body sizes of territorial reef fishes, such as
parrotfishes, groupers and snappers. These trends accompany
a three-to-four fold increase in site size and number in places
such as the island of Nevis (Wilson 1989).
Compared to island environments, changes in the
zooarchaeological record of continental sites may be less
apparent due to greater biological input from a larger reservoir
(MacArthur and Wilson 1967). This appears to be especially

true of highly productive environments such as estuaries.
Shifts in species composition, ontogenetic age structure, and
changes in animal body size related to over-exploitation may
require greater human harvesting pressure in such environ-
ments before they become clear in the zooarchaeological
record. When these changes are observed they could be
attributed to factors other than over-exploitation, such as
human selection or changes in the environment. For example,

gure 1. Outline map showing the location of the five
chaeological sites analyzed in this study.

the role that humans played in the extinction of the North
American mammoth and mastodon at the end of the Pleisto-
cene is hotly debated. Some have attributed their extinction to
environmental conditions associated with the onset ofthe early
Holocene, while others have suggested that human hunting
pressure was a major contributing factor (Webb 1985:212-
The population dynamics of an organism relates to
ontogenetic age, growth rate, mortality and recruitment.
Human over-exploitation can perceptibly alter species popula-
tion dynamics. Since size is not always a proxy for age,


VOL. 53(2-3)



independent evidence for determining the ontogenetic age
structure of animal populations is an important tool for
identifying the over-exploitation of some animal resources
(Cerrato 1980; Hallam 1972). Bivalve mollusks are sensitive
indicators of human harvesting pressures. Furthermore, in
coastal sites molluscan remains tend to be the most numerous
zooarchaeological constituents, thus sample sizes are large. In
this paper we apply a technique that can be used to determine
the ontogenetic age structure of hard clam Mercenaria spp.
populations (Deevey 1947). In so doing we will show that
people living at five pre-Columbian archaeological sites

Translucent incr
Opaque incr


9 ,



in the heavy isotope of oxygen 80 (Figure 3), indicating shell
formation during the warm part of the year (Jones and
Quitmyer 1996; Quitmyer et al. 1997). The white or opaque
increment is relatively enriched in 'O, indicating cooler water
conditions. These data show that a couplet of one opaque and
one translucent increment constitutes one year's shell-growth
and subdivides the continuous shell-growth record into
successive, annual components (Figure 2). A count of the
translucent (or opaque) growth increments provides the
ontogenetic age in years of the animal at the time of death. It
is notable that hard clams are a long-lived species that may

nt -Annual growth cycle


Ventral ,
margin 5cm

Figure 2. Radial cross-section of the hard clam (Mercenaria spp.) exposing the annual shell-growth increments. The numbers
identify each year of life, while the lines indicate size (shell height) at age.

(Figure 1) in the southeastern United States collected these
animals at a rate that significantly altered the age structure of
the exploited clam beds.

Materials and Methods

Mostbivalve shells contain macroscopic growth-increment
patterns that form with an annual periodicity (Carter 1990;
Quitmyer et al. 1997). These are analogous to tree rings. As
bivalve mollusks grow, their shells increase in size by the
addition of calcium carbonate along the shell margin.
Through the annual growth cycle environmental factors such
as temperature, or factors related to temperature, influence
changes in shell microstructure, density, and chemistry. This
biogeochemical recording of the environment and climate is
present in the concentric rings, bands, or annuli formed in the
shells of many species (Rhoads and Lutz 1980).
In hard clam shells the incremental growth structures are
viewed best in radial cross-sections of the shell (Figure 2). For
hard clams in the southeastern United States, successive
micro-samples of calcium carbonate drilled along the shell-
growth axis (youngest to oldest) show that the translucent or
dark increment (as seen in reflected light) is relatively depleted

attain ages of 40 years.
Once the ontogenetic age is determined for each specimen,
the rate of survivorship can be expressed as the percentage of
the population surviving each year of life. A complete plot of
these dynamic data can be expressed in survivorship curves
thatfacilitate the comparison of modern and zooarchaeological
assemblages (Cerrato 1980; Deevey 1947; Hallam 1972).

Zooarchaeological Hard Clams

Hard clams were excavated from five archaeological sites
in south Georgia and east-central Florida (Table 1, Figure 1):
a) Marsh Shell Ring (9GN57), St. Simons Island, GA
(Marrinan 1979, Quitmyer et al 1985); b) Poisonberry and
Artesian Well areas of the Kings Bay Site (9CAM171), GA
(Quitmyer et al. 1985); c) Marsh and South Bunker areas of
the Devil's Walkingstick site (9CAM177), GA; d) Edgewater
D (8V0115) and Edgewater B (8VO1705), FL (Russo 1989);
and e) Snyders Mound and Fiddle Crab Mound at Seminole
Rest (8Vo124), FL (Horvath 1995).

2000 VOL. 53(2-3)


ement -
4 3

Table 1. Archaeological site, site areas, and dates of the archaeological site components containing hard clams analyzed in
this study.

Archaeological Site and Site Area Archaeological Period and Date Reference


Marsh Shell Ring (9GN57)
Kings Bay -- Poisonberry Area (9CAM171)
Kings Bay -- Artesian Well Area (9CAM171)
Devils Walkingstick -- Marsh Area (9CAM177)
Devils Walkingstick --South Bunker Area (9CAM177)

Edgewater D (8V0115)
Edgewater B (8V01705)
Seminole Rest -- Snyders Mound (8V0124)
Seminole Rest -- Fiddle Crab Mound (8VO124)


Late Archaic 2500 -1000B.C.
Swift Creek A.D. 300 -- A.D. 700
Swift Creek A.D. 300 -- A.D. 700
Savannah A.D. 1000 A.D. 1500
Savannah A.D. 1000 -- A.D. 1500

St. Johns IA IB -- A.D. 460 -- A.D. 650
St. Johns IA IB A.D. 460 A.D. 650
St. Johns I II -A.D. 670-1010
St. Johns I- II --A.D. 670-1010

Marrinan 1975
Quitmyer et al. 1985
Quitmyer et al. 1985
Quitmyer et al. 1985
Quitmyer et al. 1985

Russo et al. 1989
Russo et al. 1989
Quitmyer 1995
Quitmyer 1995

Modem Hard Clams

Hard clams were collected from Kings Bay, GA and the
Indian River, FL during the course of two studies that corre-
lated incremental shell formation with the time of the year
(Arnold et al. 1998; Quitmyer et al. 1997). These specimens
were randomly collected through at least two annual periods.

Results and Discussion
At Kings Bay, GA and St. Simons Island, GA
we observed a shift in the age structure of hard -1.50
clams from middens representing the Archaic -1.00
(Figure 4), Swift Creek (Figure 5), and Savannah C
periods (Figure 6) (Quitmyer et al. 1985). During h- -0.50
this 4,500-year interval pre-Columbian sites grew
in number and in density (Caldwell 1971). The Q 0.00
hard clam population from the Archaic period 0.
(2500 B.C. 1000 B.C.) Marsh Ring, St. Simons 0.
Island, GA had an average age of 6.8 years at the 1.00
time of death (Quitmyer et al. 1985:37). The
survivorship curve shows mortality declining with 1.50
age and the greatest mortality occurring between
one and eight years of age (Figure 4). Notably,
the assemblage contains a few individuals that F
reached 25 years of age. This declining mortality Charlote
rate is similar to the pattern observed for the variation.
Kings Bay, GA Artesian Well and Poisonberry carbonate s
middens that date to the Swift Creek period (A.D. e nc
300 A.D. 700) (Figure 5) (Quitmyer 1985:36). s re
In contrast, age structure of the Kings Bay Savan- y more n
nah period (A.D. 1000 A.D. 1500) hard clams
is quite different from those of earlier times (Figure 6). The
average age of Savannah period hard clams is around 4.5
years. The greatest mortality occurs before age five years and
there are no individuals older than 14 years of age (Quitmyer
et al. 1985:35).
When the survivorship profile of modern Kings Bay hard

clams (Figure 7) is compared with the zooarchaeological
assemblages, the greatest similarity is with the Archaic and
Swift Creek populations (Figure 6). The mean age at time of
death for the modern hard clams is 10.5 years and this is not
significantly different from the Archaic or Swift Creek
assemblages (Quitmyer 1985:38). However, the mean age of
the Savannah period samples is significantly younger

Charlotte Harbor, FL

I 11 1
Sample Number

Marginal portions of a cross-sectioned hard clam valve from
harbor, FL, showing the micro-sampling locations and the 86O0
T or 0 is placed with each data point, indicating whether the
sample was taken from the translucent (dark) or opaque (white)
nent, respectively. Warmer water temperatures are indicated
gative O80 values.

(Quitmyerl985:38). It is noteworthy that human harvesting
has not influenced the age structure of the modern assemblage.
Access to Kings Bay has been prohibited by the military for
the last 45 years because it served as a military munitions
loading facility and later as a Trident Submarine support base.
Modern hard clam harvesting with heavy equipment is further




Archaic Period-- 2500-1000 B.C.

0 2 4 6 8 10 12 14 16

18 20 22 24

Age (Years)
Figure 4. Survivorship curve for zooarchaeological hard clams (Mercenaria spp.) excavated from the Archaic Period
(2500-1000 B.C.), Marsh Shell Ring (9GN57), St. Simons Island, GA.

Swift Creek Period -- A.D. 300-700

o 80 Artesian Well
= 6 -&- Poisonberry
j 60


0 2 4 6 8 10 12 14 16 18 20 22 24
Age (Years)

Figure 5. Survivorship curves for zooarchaeological hard calms (Mercenaria spp.) excavated from the Swift Creek period
(A.D. 300-700), Artesian Well and Poisonberry areas of the Kings Bay Site (8CAM171), GA.








2000 VOL. 53(2-3)

Savannah Period -- A.D.1000-1500
,100 -
o -v- Marsh
S-- South Bunker
(/ 60


0. O
0 2 4 6 8 10 12 14 16 18 20 22 24
Age (Years)
Figure 6. Survivorship curves for zooarchaeological hard calms (Mercenaria spp.) excavated from the Savannah period
(A.D. 1000-1500), Marsh and South Bunker areas of the Kings Bay Site (8CAM171), GA.








0 2 4 6 8 10 12 14 16 18 20 22 24

Age (years)
Figure 7. Survivorship curves for randomly collected live hard calms (Mercenaria spp.) from Kings Bay, Georgia and
Indian River, FL.




St. Johns IA-IB A.D. 460-650






0 2 4 6 8 10 12 14 16 18 20 22 24
Age (Years)

St. Johns I II A.D. 670-1010


. 80

S 60

I 20

0. o

0 2 4 6 8 10 12 14 16 18 20 22 24
Age (Years)
Figure 8. Survivorship curves for zooarchaeological hard calms (Mercenaria spp.) excavated from St. Johns IA and IB
periods (A.D. 460-650), Edgewater D and Edgewater B Sites, FL and from St. Johns I and IIA periods (A.D. 670-1010)
Snyders Mound and Fiddle Crab Mound, Seminole Rest, FL.


2000 VOL. 53(2-3)











Von Bertalanffy Growth Curve

2 4 6 8
Age (Years)

10 12

Figure 9. Von Bertlanffy growth curve showing the theoretical maximum shell height (mm) vs. age (years) of measured
hard clams (Mercenaria spp.). The two vertical lines show the theoretical 8 mm difference in shell height of a five-year-old
and 10-year-old hard clam. The analyzed specimens were randomly collected live from Kings Bay, GA (Jones et al. 1990).

Mercenaria -- Kings Bay, GA


0 2 4 6 8 10 12 14 16 18 20 22 24
Age (Years)

Figure 10. A scatter plot of the shell height (mm) vs. age (years) of living hard clams (Mercenaria spp.) randomly collected
from Kings Bay, GA. (Quitmyer et al 1985). The box encompasses specimens that are between 5 and 20 years of age,
showing that size is not a good proxy for age.


o o

- - --




limited by the shallow waters of the creeks and bays of the
south Georgia coast (Walker and Tenore 1984).
Hard clams excavated from Edgewater, Snyders and
Fiddle Crab Mounds (Figure 8) provide evidence of the over-
exploitation of hard clams along Florida's Mosquito Lagoon
during the St. Johns period. Edgewater mounds D and B date
between A.D. 380 and 710, while the Snyders and Fiddle Crab
Mounds date between A.D. 670 and 910 (Horvath 1995; Russo
etal 1989). Zooarchaeological quantification of these middens
shows that at Edgewater hard clams were the second most
abundant species in the midden, while at Snyders and Fiddle
Crab Mounds they accounted for over 99% of the identified
species (Kozuch 1995). The excavation at Snyders Mound, for
example, exposed a 6 m deep, stratified section of hard clam
shells that accumulated in less than 140 years (Horvath 1995).
Test excavations from the 1.2 hectare site indicate that the
mound is almost entirely composed of hard clam shells.
The survivorship curves from the four St. Johns period
sites (Figure 8) are similar in structure to the Kings Bay
Savannah period curves (Figures 6). They show that the
greatest mortality took place during the first six years of life.
At Edgewater B, 80% of the specimens never exceed three
years of age and the older portion of the population is missing
from the assemblages (Figure 8). When compared to the
modern age structure of hard clams from the nearby Indian
River (Figure 7), all of the St. Johns period hard clam samples
are significantly younger, except for the Snyders Mound clams
(Quitmyer 1995:129). Unlike today's Kings Bay hard clams,
the average age (years) of the modern Indian River population
has been lowered by intensive commercial harvest. This study
suggests that the St. Johns people were exerting greater
cropping pressure than modem commercial harvesters.
The question might arise-could the diminished age
structure of mollusk populations observed in this study be the
result of humans selecting clams of a given size rather than
over-exploitation? This seems unlikely. First, hard clams are
infaunal animals and size-class selection would be difficult
because they are blindly extracted from the substrate. Further-
more, the theoretical size at each year of life, estimated by the
Von Bertalanffy growth curve (Figure 9), shows that there is
only an 8 mm difference between a 5-year-old hard clam and
one that is 10 years old. Third, size is not always a proxy for
age. For example, a five-year-old clam can be just as large as
a 20-year-old clam (Figure 10). In other words, the harvest of
clams that range between 70 and 85 mm (shell length) would
contain individuals that range between 5 and 20 years of age.
In the zooarchaeological assemblages that have been impacted
by intensive harvesting, the older members of the population
are missing from the zooarchaeological assemblage.


Age determination of hard clam shells using annual shell
growth increments provides evidence for the over-exploitation
of this resource at five prehistoric sites in coastal southeast
Georgia and coastal east-central Florida. The data show that

hard clams were collected at such a rate that the average
ontogenetic age of their populations was significantly reduced.
This seems to occur in those sites where there is evidence for
dense human populations, evidence for sedentism and in
places where hard clams are a major part of the faunal
The pre-Columbian over-exploitation of hard clams
extended over a wide temporal and geographic range. It is not
likely that the people living at these sites were of the same
cultural affiliation. The decisions that resulted in the over-
exploitation of the hard clams appear to be more intrinsic to
the human species rather than associated with a specific
cultural, economic, subsistence or political system.

References Cited

Arnold, W.S., T.M. Bert, I.R. Quitmyer, and D.S. Jones
1998 Contemporaneous Deposition of Annual Growth Bands in
Mercenaria mercenaria (Linnaeus), Mercenaria
campechinesis (Gmelin), and Their Natural Hybrid Forms.
Journal of Experimental Marine Biology and Ecology
Caldwell, Joseph R.
1971 Chronology of the Georgia Coast. Southeastern Archaeo-
logical Bulletin 13:88-92.
Carter, Joseph G. (ed.)
1990 Skeletal Biomineralization: Patterns, Processes and
Evolutionary Trends, Volume I Van Nostrand Reinhold,
New York.
Cerrato, Robert M.
1980 Demographic Analysis of Bivalve Populations. In Skeletal
Growth of Aquatic Organisms: Biological Records of
Environmental Change, edited by Donald C. Rhodes and
Richard A. Lutz, pp. 417-463. Plenum Press, New York.
Deevey, Edward S.
1947 Life Tables of Natural Populations of Animals. Quarterly
Review of Biology 22:283-314.
Hallam, Anthony
1972 Models Involving Population Dynamics. In Models in
Paleobiology, edited by Thomas J.M. Schopf, pp. 62-81.
Freeman, Cooper, and Co., San Francisco.
Horvath, Elizabeth A.(ed.)
1995 Final Report on the Archaeological Investigations at the
Seminole Rest Site (CANA-063/8V0124), Canaveral
National Seashore, Volusia County. National Park Service,
Southeast Archaeological Center, Tallahassee.
Jones, Douglas S., and Irvy R. Quitmyer
1996 Marking Time with Bivalve Shells: Oxygen Isotopes and
Season of Annual Increment Formation. Palaios
Jones, Douglas S., Irvy R. Quitmyer, William S. Arnold, and Dan. C.
1990 Annual Shell Banding, Age, and Growth Rate of Hard
Clams (Mercenaria spp.) from Florida. Journal ofShell-
fish Research 9:215-225.
Kozuch, Laura
1995 Shellfish Gathers of Florida's East Coast. In Final Report
on the Archaeological Investigations at the Seminole Rest
Site (CANA-063/8V0124), Canaveral National Seashore,
Volusia County, edited by Elizabeth S. Horvath, pp. 59-91.
National Park Service, Southeast Archaeological Center,


2000 VOL. 53(2-3)


Lepofsky, Dana, Patrick V. Kirch, and Kenneth P. Lertzman
1996 Stratigraphic and Paleobotanical Evidence for Prehistoric
Human-Induced Environmental Disturbance on Mo'orea,
French Polynesia. Pacific Science 50(3):253-273.
MacArthur, Robert H., and Edward O. Wilson
1967 The Theory of Island Biogeography. Monographs in
Population Biology, Princeton University Press, Princeton.
Marrinan, Rochelle A.
1975 Ceramics, Mollusks, and Sedentism: The Late Archaic
Period on the Georgia Coast. Ph.D. dissertation, Depart-
ment of Anthropology, University of Florida, Gainesville.
Quitmyer, Irvy R.
1995 Season of Harvest and Age Structure ofMercenaria spp.
Excavated from St. Johns II(A.D. 590-1400) Archaeologi-
cal Contexts of the Mosquito Lagoon, Florida. In Final
Report on the Archaeological Investigations at the Semi-
nole Rest Site (CANA-063/8V0124), Canaveral National
Seashore, Volusia County, edited by Elizabeth S. Horvath,
pp. 92-132. National Park Service, Southeast Archaeologi-
cal Center, Tallahassee.
Quitmyer, Irvy R., H. Stephen. Hale, and Douglas S. Jones
1985 Paleoseasonality Determination Based on Incremental Shell
Growth in the Hard Clam, Mercenaria mercenaria, and Its
Implications for Future Analysis of Three Georgia Coastal
Shell Middens. Southeastern Archaeology 4(1):27-40.
Quitmyer, Irvy R., Douglas S. Jones and William Arnold
1997 The Sclerochronology of Hard Clams, Mercenaria spp.,
from the Southeastern U.S.: A Method of Elucidating the
Zooarchaeological Records of Seasonal Resource Procure-
ment and Seasonality in Prehistoric Shell Middens.
Journal ofArchaeological Science 24:825-840.
Rhoads, Donald C., and Richard A. Lutz
1980 Skeletal Growth ofAquatic Organisms: Biological Records
of Environmental Change. Plenum Press, New York.
Russo, M., A. Cordell, L. Newsom, and R. Austin
1989 Phase III Archaeological Excavations at Edgewater
Landing, Volusia County, Florida. Prepared for
Randor/Edgewater Landing, Inc. by Piper Archaeological
Research, Inc., St. Petersburg, FL.
Steadman, David W.
1995 Prehistoric Extinctions of Pacific Island Birds:
Biodiversity Meets Zooarchaeology. Science 267:1123-
Walker, Randal L., and Kenneth R. Tenore
1984 The Distribution and Production of the Hard Clam,
Mercenaria mercenaria, in Wassaw Sound, Georgia.
Estuaries 7(1)19-27.
Webb, S. David
1985 Main Pathways of Mammalian Diversification in North
America. In The Great American Biotic Interchange,
Volume 4, edited by Francis G. Stehli and S. David Webb,
pp. 201-214. Plenum Press, New York.
Wing, Elizabeth S.
1999 Land Crab Remains in Caribbean Sites. Paper presented
at the XVIth International Congress for Caribbean Archae-
ology. Guadeloupe.
Wilson, Samuel M.
1989 The Prehistoric Settlement Pattern Nevis, West Indies.
Journal ofField Archaeology 16(4):427-450.



S. yes, but what was the

The exact, full wording of that reference is as close
as your phone:

Back issues of The Florida Antbhf pologist -- going
back close to a half century are available at the

Graves Museum of Archaeology
and Natural History

481 South Federal Highway
Dania, FL 33004

Phone (305)925-7770
FAX (305)925-7064

Sole agents for back issues of The Florida Anthropologist



National Park Service, SoutheastArcheological Center, 2035E. PaulDiracDrive, Johnson Building, Box 7, Tallahassee, FL 32310
Email: margo_schwadron@nps.gov

In January of 1997, the National Park Service conducted
archaeological investigations within De Soto National Memo-
rial in Bradenton, Florida. Previous to these investigations, no
comprehensive archaeological survey had ever been conducted
within the park. Archaeological testing was conducted to
determine the nature and origin of a series of curvilinear shell
ridges located within a mangrove swamp, as well as several
possible shell mounds. The investigation strategy included
survey and mapping of the park, the placement of two excava-
tion units in two of the possible
mounds, and shovel testing of
the shell ridges and middens.
Twenty radiocarbon dates, Alabama
archaeobiological studies, and
artifact analyses indicate that
the site was once an extensive
Manasota through Safety Har-
bor period village complex oc-
cupied from approximately
2510 +/- 60 B.P. 1100 +/- 60
B.P. (365 B.C. to A.D. 1395.)
This study demonstrates that
dynamic coastal settings may
have undergone complex cul-
tural and natural site-formation
processes that are observable in
the archaeological record.

Site Location and History

De Soto National Memorial
is located on the central gulf
coast of Florida (Figure 1),
within Manatee County, five
miles from the city of
Bradenton. The park is bor-
dered on the east by a calm,
shallow cove, and on the north
by the Manatee River, at the
southern mouth of Tampa Bay.
The coastal park is situated in a s 0 5
rich, estuarine environment,
with most of the park's 10.68
hectares consisting of a mature
mangrove community. Aprom- Figure 1. Location of the

inent spit of land, historically known as Shaw's Point, projects
into the Manatee River (Warner 1986).

Historic Site Descriptions of Shaw's Point

Prior to planning archaeological investigations at De Soto
National Memorial, an Archaeological Overview and Assess-
ment study was completed (Schwadron 1998). The study
uncovered many historic documents and early site descriptions
indicating that De Soto National Memorial contained the

Shaw's Point site (8MA7).


VOL. 53(2-3)




'41 '.

9f .dA

A" '

._E t) *~~

,r ," ', k *^ w'. '., *" .,..

Ic ir1it

Figure 2. The Shaw's Point shell mound, 1910. Courtesy of Manatee County Historical Society, Bradenton, Florida.

V ~~sd.
r ~ "

-( ~ ,~rs~ *^ -? ^ ^ y I-''' S -~ ^
;.,^ -,,-., ;-Cr *; ~*< "i-t^ ^ ^

" *-.- -.* r **^ T
,^^r ~I*I :*;^.:"-^*^
VI, ,
'C-r = ..,
msmI,, x


r; 4~




*ES T .

. V dk



Figure 3. Plan view drawing of the Shaw's Point site, from Walker 1880.

- Vh- oo I

Figure 4. Sketch map of the Shaw's Point site by Charles T. Earle, 1920.

Shaw's Point site (8MA7), a once extensive and significant
prehistoric archaeological mound complex (Schwadron 1998).
Most of the shell mound complex was destroyed around 1910
when its shell was mined for roadfill. Several photographs of
the Shaw's Point site dating from the early 1900s are archived
by the Manatee County Historical Society. The photographs
depict large shell mounds, probably only a few years before
they were mostly destroyed (Figure 2).

The earliest published reference to the archaeological site
at Shaw's Point was by Daniel G. Brinton in 1859. Brinton's
(1859:180) reference to a shell mound at the mouth of the
Manatee River describes it as "twenty feet in height... [with]
numerous fragments of coarse, ill marked pottery...mixed with
charcoal, ashes and dirt, and never more than three feet below
the surface." S. T. Walker described the Shaw's Point site as
an expansive series of shell middens and mounds extending


ne --~3 3


Figure 5. 1941 aerial photograph of the Shaw's Point site and area.

along the shore for 171.90 m, with a height exceeding 9.14 m
(Walker 1880; see Figure 3).
In the 1920s and 1930s, local archaeologist Montague
Tallant visited the Shaw's Point site and described it as "one
of the most important sites on the west coast" (Tallant n.d.).
Tallant described the Shaw's Point site as an extensive village
complex consisting of a large shell kitchen midden covering
2 or so hectares, and including five burial mounds, one
cemetery, and a roadway .80 km long. Tallant considered the
Pillsbury Mound (8MA31) to be one of the burial mounds of
the Shaw's Point village.
Other early investigators who visited, described, or
collected remains from the Shaw's Point site include Jeffries
Wyman (1870), Marshall Newman (Willey 1949:173),
William Plowden, John Goggin, and Ripley Bullen. Willey
(1949:342) considered Shaw's Point to be a significant site
with a long occupation, possibly dating from the Deptford to
the Safety Harbor period.

Shaw's Point Shell Ridges. While most researchers who
visited Shaw's Point described the extensive shell mounds and
burial mounds, only a few mentioned the presence of shell
ridges and shell middens hidden in the mangrove swamp.

As a young child in 1881, Asa Pillsbury settled with his
family just west of Shaw's Point. During a 1963 interview,
Pillsbury described the Shaw's Point site and the shell ridges
in the mangrove swamp to Manatee County Historian Carl

King There was a kind of causeway from here [Pillsbury
Mound] to Shaw's Point?
Pillsbury There was a whole lot to it. They had made
ridges with their shells. And in between the ridges they had
water for their canoes to come into back of there. So there
always been a harbor and always be handy to get in and out.
The ridges come to the shell mound. You can see them there
now. That is on the Mangrove side of it. From where the
gate is, that was all mangroves from there on down. And
from where the gate is, the ridge comes up there [Pillsbury in
King 1963:12].

From 1920 to 1927, Bradenton resident Charles T. Earle
documented development activities at Shaw's Point. Ed
Ballard, the owner of the Shaw's Point property, ventured to
develop the property into an island resort by cutting a canal
through the mangrove swamp and draining it. During this
time, Earle collected a large amount of artifacts from Shaw's


2000 VOL. 53(2-3)



0 50o 100 o ten
-- I

Figure 6. Archeological base map of De Soto National Memorial and the Shaw's Point

Point, donating them to the Smithsonian Institution (Willey
1949:341). Over a seven-year period, Earle corresponded with
Jesse W. Fewkes (then Chief of the Bureau of American
Ethnology), providing an invaluable narrative describing the
shell ridges and mounds at Shaw's Point.
Earle's description of the Shaw's Point site mentions a
large shell mound along the north side of the beach, a large
"muck pond," and describes the entire inland area as "muck
beds lying between low shell ridges" (Earle 1920). He states
that during preparation for the development, the entire swamp
was cleared of vegetation, and that the shell ridges, "being
cleared of all growth, are easy to examine and dig in" (Earle
1920). According to Earle, Ballard eventually became
sensitive to the significance of the site, and continued the
development of the property with efforts to avoid damaging the
shell ridges.
Earle's correspondence described the dredging of a 3 m
drainage canal through the swamp, which took a total of two
months to dig. While dredging, Earle noted that:

The digging proved the shell ridges or benches enclosed
water courts like Mr. Cushing described, as the muck
averaged two to three feet deep, covering a deposit of shell of
various depths mostly running deeper than the ditch (five
feet). This shell was not a natural oyster bar- it was evi-
dently deposited by the aborigines as [it] is exactly like the
shell in the mound, that is, the refuse from meals containing
broken bones, fragments of pottery, etc. One burial was
encountered, a badly decayed skeleton lying full length on

top of the shell strata, covered by two
feet of topsoil (sand and mud mixed).
I did not see it in position but the
workmen's detailed accounts were
descriptive. The bones crumbled
away and the teeth were carried off by
the men [Earle 1920].

A very significant part of the
Earle-Fewkes correspondence is two
sketch maps of the Shaw's Point site,
which include details of the shell
ridges, muck pond, shell mounds, and
the location of the drainage ditch
(matching the existing location of the
drainage ditch within the park) (Fig-
ure 4). Several other interesting fea-
tures that Earle noted on his maps are
an "ancient shoreline," and an "an-
cient canal." Earle's placement of the
shell ridges also corresponds very
well with the position of the ridges
today, and his maps indicate that an
effort was taken to avoid disturbing
the ridges. These maps are invaluable
for documenting the relationship of
the site features before park develop-

Present Appearance of the Shaw's Point Site

Most of the shell mound complex was destroyed around
1910 when its shell was mined for road fill. The area was
designated a National Memorial in 1948, after the United
States De Soto Commission found that the vicinity of Shaw's
Point was the most likely landing place of Spanish explorer
Hernando de Soto (Swanton 1985). A small portion of the
land was cleared and developed for a visitors' center and
parking lot. Today, with the exception of the two possible
shell mounds (Marker and Remnant), and the shell ridges,
middens and a small shell mound (Egret Mound) located in
the mangrove swamp, there is little visible evidence of the
expansive shell mound complex that once encompassed the
area (Figure 5).

Research Methods and Results

The research plan for archaeological investigations at De
Soto National Memorial was designed to meet several objec-
tives, including: (1) survey and mapping of the park to create
an archaeological base map and to delineate the shell ridges,
middens and mounds; (2) subsurface testing of shell ridges and
middens in the mangrove swamp; and (3) the excavation of
test units placed in the two possible mounds (Marker and
Remnant Mounds).


Table 1. Radiocarbon results from the shell ridges, middens and Egret Mound (Seac Acc. 1283).

Calibrated Calendar Calibrated Mean
Sample No. Provenience Material Conventional C-14 Age* Date** Date
Beta-103939 Ridge 3, ST 23, 40-45 cmbs Oyster 1220 + 60 BP A.D. 1050 1295 A.D. 1200
Beta-103938 Ridge 3, ST 25, 40-45 cmbs Oyster 1100 + 60 BP A.D. 1200 1395 A.D. 1295
Beta-103937 Egret Mound (Ridge 5), ST 31, 32 cmbs Charcoal 1610 + 70 BP A.D. 265- 290 and A.D. 435
A.D. 320 -615
Beta-103940 Egret Mound, ST 31, 46-56 cmbs Oyster 1710 + 60 BP A.D. 595 800 A.D. 685
Beta-103941 Ridge 6 ST 33 Oyster 2210 + 60 BP A.D. 15 280 A.D. 140
Beta-103942 Ridge 6 ST 33, 90-100 cmbs Oyster 2210 +60 BP A.D. 15 280 A.D. 140
Beta-103946 Ridge 7, ST 36, 85-90 cmbs Oyster 2360 + 60 BP 175 B.C. A.D. 110 20 B.C.
Beta-103947 Ridge 7, ST 37, 40-50 cmbs Fl. Fighting Conch 2510 + 60 BP 365 B.C.-. 55 B.C. 200 B.C.
Beta-103933 Midden 18, 60 cmbs Unburned Wood 110 + 50 BP A.D. 1670 1950 multiple
Beta-103934 Midden 18, 60 cmbs Oyster 2200 + 70 BP A.D. 0 330 A.D. 150
Beta-103932 Midden 26,40 cmbs Lightning Whelk 1790 + 70 BP A.D. 465 730 A.D. 635

Uncalibrated with one sigma (68%) confidence level
** Calibrated with two sigmas (95%) confidence level

Table 2. Radiocarbon results from the Marker Mound (Seac Acc. 1283).

Calibrated Calendar Calibrated Mean
Sample No. Provenience Material Conventional C-14 Age* Date** Date
Beta-103943 Zone G Oyster 2040 + 60 BP A.D. 205-475 A.D. 355
Beta-103944 Zone H, Level 14, 125-135 cmbd Oyster 2320 + 70 BP 155 B.C. A.D. 175 A.D. 25
Beta-103945 Zone I, Level 17, 155-165 cmbd Oyster 2190 + 70 BP A.D. 15 345 A.D. 160

Uncalibrated with one sigma (68%) confidence level
** Calibrated with two sigmas (95%) confidence level

Table 3. Radiocarbon results from the Remnant Mound (Seac Acc. 1283).

Calibrated Calendar Calibrated Mean
Sample No. Provenience Material Conventional C-14 Age* Date** Date
Beta-103935 Zone A, Level 4, 34 cmbd Charcoal 1560 + 50 BP A.D. 410 620 A.D. 535
Beta-103936 Zone A, Level 4, 36-40 cmbd Lightning Whelk 1630 + 60 BP A.D. 660 895 A.D. 760
Beta-103948 Zone B, Level 7, 60-70 cmbd Oyster 1930 + 70 BP A.D. 310 635 A.D. 465
Beta-103949 Zone C, Level 11, 102-110 cmbd Oyster 2110 + 60 BP A.D. 120 410 A.D. 260
Beta-103950 Zone D, Level 15, 140-150 cmbd Oyster 2260 + 70 BP 70 B.C. A.D. 250 A.D. 90
Beta-103951 Zone D, Level 16, 150-164 cmbd Oyster 2260 + 60 BP 45 B.C A.D. 230 A.D. 90

* Uncalibrated with one sigma (68%) confidence level
** Calibrated with two sigmas (95%) confidence level


Radiocarbon Sampling

Figure 7. Shell ridge with scatter of large lightning whelk


All shell ridges, middens and mound features were mapped
using a Trimble Pro-XL global positioning system (GPS) with
sub-meter accuracy. The use of GPS was invaluable for
producing a plan-view map of the shell features in the sensi-
tive mangrove swamp. Using GPS required no cutting of
vegetation, saving valuable field time and preventing damage
to the mangroves. The data were exported to geographic
information systems (GIS) software, and a plan-view map of
the ridges and shell features (Figure 6) was produced that very
closely matches the earlier aerial photographs that show these
features (see Figure 5).
While conducting the survey and mapping of features in
the mangrove swamp, a total of seven ridge components were
located and mapped. Other mapped features included four
small circular shell middens, one small shell midden mound
(Egret Mound), and 13 dredge spoil piles.

Since shell was prevalent throughoutthe site, shell samples
were taken from all proveniences for possible radiocarbon
dating. When carbon or burnt wood was encountered in situ,
the material was carefully removed with a clean trowel,
wrapped in aluminum foil, and then placed in a protective vial
or bag, which was labeled with the appropriate provenience
The shell radiocarbon samples taken from shovel tests in
the mangrove swamp were taken directly from the shovel test
hole from a specifically measured depth (not randomly).
When collecting a shell radiocarbon sample, all of the shell
selected was taken from one measured location within the test
(i.e., shell collected together from the center of the test at 50
cm depth) To maintain consistency between samples,
oysters-the most prevalent mollusk at the site-were selected
whenever possible. Robust oyster shells were chosen over
smaller, more fragile shells because they reduce the potential
for erroneous radiocarbon dates causedby secondary carbonate
components, and calcium carbonate re-carbonization (or re-
crystallization) of the shells. On occasion, other mollusks,
such as lightning whelk and Florida fighting conch also were
taken as samples. Approximately 115 grams of shell were
submitted for each radiocarbon sample. Clean, whole shells,
which showed no signs of re-carbonization, were selected for
Twenty radiocarbon samples were submitted to Beta
Analytic Inc., and the results are listed in Tables 1-3. All
radiocarbon dates were calibrated by Beta Analytic, Inc. and
are reported in this paper using two sigma date ranges (95%
confidence level). Radiocarbon ages (uncalibrated date) are
given in years before A.D. 1950 + one standard deviation
(68% confidence level). Corrections for C-13 and C-12 were
conducted on marine carbonates (shell samples), and the dates
adjusted for both global and local geographic reservoir effects.
Calibrations of the radiocarbon age determinations were made
using the Pretoria Calibration Procedure (Stuiver et al. 1993,
Talma and Vogel 1993:317-322, and Vogel et al. 1993:73-86).

Investigation of the Shell Ridges and Egret Mound

The site contains a series of long, curvilinear shell ridges
that are oriented north-to-south and run parallel to the cove.
Some of the shell ridges are vegetated with small cacti, brush,
and occasionally small buttonwood trees, but most are devoid
of heavy vegetation. All of the ridges are completely sur-
rounded by mangrove swamp. Very large scatters of robust
lightning whelk (Busycon sinistrum) shells cover areas of the
shell ridges, and Native American pottery and faunal material
also are present on the surface (Figure 7). Investigation of the
shell ridges was conducted to test the hypothesis that they are
cultural in origin, and if so, to determine their function, age,
and cultural affiliation.
Each ridge was tested with two 50 x 50 cm shovel tests. All
material removed from each test was screened using 6.4 mm
(1/4") hardware mesh. All artifacts and ecofacts recovered


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