Table of Contents
 Editor's page - Brent R. Weism...
 8WL38, A protohistoric village...
 Early aerial photography : A remote...
 A cache of 55 points : Taylor County,...
 What do you call this point? Louis...
 Clubheads, bola stones, or what?...
 Of fields and streams : A tribute...
 Commentary : An engraved bone artifact...
 Painting Florida's vanished Indian...
 Featured photographs
 Join the Florida Anthropological...
 Back Cover

Group Title: Florida anthropologist
Title: The Florida anthropologist
Full Citation
Permanent Link: http://ufdc.ufl.edu/UF00027829/00053
 Material Information
Title: The Florida anthropologist
Abbreviated Title: Fla. anthropol.
Physical Description: v. : ill. ; 24 cm.
Language: English
Creator: Florida Anthropological Society
Conference on Historic Site Archaeology
Publisher: Florida Anthropological Society.
Place of Publication: Gainesville
Frequency: quarterly[]
two no. a year[ former 1948-]
Subject: Indians of North America -- Antiquities -- Periodicals -- Florida   ( lcsh )
Antiquities -- Periodicals -- Florida   ( lcsh )
Genre: periodical   ( marcgt )
Summary: Contains papers of the Annual Conference on Historic Site Archeology.
Dates or Sequential Designation: v. 1- May 1948-
 Record Information
Bibliographic ID: UF00027829
Volume ID: VID00053
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
        Page 231
    Editor's page - Brent R. Weisman
        Page 232
    8WL38, A protohistoric village site on Choctawhatchee Bay - Gregory A. Mikell
        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
        Page 250
        Page 251
        Page 252
        Page 253
        Page 254
        Page 255
        Page 256
        Page 257
        Page 258
        Page 259
        Page 260
        Page 261
        Page 262
        Page 263
        Page 264
        Page 265
        Page 266
        Page 267
        Page 268
    Early aerial photography : A remote sensing technique used to detect prehistoric earthworksin the Kissimmee river basin - William Gray Johnson
        Page 269
        Page 270
        Page 271
        Page 272
        Page 273
        Page 274
        Page 275
        Page 276
        Page 277
        Page 278
        Page 279
    A cache of 55 points : Taylor County, Florida (8TA203) - Clark Hardman, Jr., and Marjorie H. Hardman
        Page 280
        Page 281
        Page 282
        Page 283
        Page 284
        Page 285
        Page 286
    What do you call this point? Louis D. Tesar
        Page 287
        Page 288
        Page 289
        Page 290
        Page 291
        Page 292
        Page 293
        Page 294
    Clubheads, bola stones, or what? Louis D. Tesar
        Page 295
        Page 296
        Page 297
        Page 298
        Page 299
        Page 300
        Page 301
        Page 302
        Page 303
    Of fields and streams : A tribute to Hub Chason, Sr., a Florida river diver - James Dunbar
        Page 304
        Page 305
        Page 306
        Page 307
        Page 308
        Page 309
        Page 310
        Page 311
        Page 312
    Commentary : An engraved bone artifact from the Summer Haven site: A clarification of Wheeler's account - Randy V. Bellomo
        Page 313
        Page 314
        Page 315
        Page 316
        Page 317
        Page 318
        Page 319
    Painting Florida's vanished Indian peoples - Theodore Morris
        Page 320
    Featured photographs
        Page 321
        Page 322
    Join the Florida Anthropological Society (FAS)!
        Page 323
        Page 324
        Page 325
    Back Cover
        Page 326
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.



Volume 47 Number 3
September 1994

Page Number

Editor's Page. Brent R. Weisman 232
8WL38, A Protohistoric Village Site on Choctawhatchee Bay. Gregory A. Mikell 233
Early Aerial Photography: A Remote Sensing Technique Used to Detect Prehistoric Earthworks
in the Kissimmee River Basin. William Gray Johnson 269

A Cache of 55 Points: Taylor County, Florida (8TA203). Clark Hardman, Jr., and Marjorie H. Hardman 280
What Do You Call This Point? Louis D. Tesar 287
Clubheads, Bola Stones, Or What?. Louis D. Tesar 295

Of Fields and Streams: A Tribute to Hub Chason, Sr., A Florida River Diver. James Dunbar 304


An Engraved Bone Artifact From the Summer Haven Site: A Clarification of Wheeler's Account. Randy V. Bellomo 313
Milanich, Archaeology of Precolumbian Florida. Reviewed by Gordon R. Willey 315

Sassaman, Early Pottery in the Southeast. Reviewed by Nancy Marie White. 318

Painting Florida's Vanished Indian Peoples. Theodore Morris 320
"The Trampled Track," Osceola National Forest, submitted by Keith Lawrence 321

Cistern Excavation at the Historic Roesch House, Brevard County, submitted by Vera Zimmerman 322
Join the Florida Anthropological Society 323
Cover: Highlands County Legacy, from a painting by Theodore Morris.
Copyright 1994 by the
ISSN 0015-3893



Brent R. Weisman

One of the benefits of being the journal editor is the
pleasure of reading complimentary copies of FAS chapter
newsletters sent in from all regions of the state. Although all
newsletters are informative as to chapter activities, two
newsletters in particular consistently stand out in terms of the
quality of information presented and overall attractiveness.
First there is the newsletter of the St. Augustine
Archaeological Association, edited by Clara Gualtieri and now
in its ninth year. This newsletter often contains feature articles
on some aspect of recent archaeological activities (the current
issue, July 1994, has a lead article on excavation of the Cubo
Line, also featured in the June 1993 issue of the journal), short
book reviews, and other brief pieces, all free of jargon and
numbing technical language. The second newsletter, in its
tenth year, is from the Southwest Florida Archaeological
Society, based in Naples. Editor Art Lee skillfully combines
feature pieces on local archaeology, notices on local events
(often with commentary), interviews, book reviews, and
reports on significant archaeological findings worldwide into
what can truly be called a newsletter par excellence. There can
be no doubt that FAS chapter newsletters, especially those
modeled after the St. Augustine and Southwest Florida
examples, have a tremendous potential to reach readers of the
general public interested in archaeology and thus are of great
importance to the future of Florida archaeology. To Clara,

Art, and the other newsletter editors, keep up the good work!
Greg Mikell's article in this issue once again
demonstrates that Florida Gulf coast archaeology truly is an
embarrassment of riches. The abundance of artifacts recovered
from the 8WL38 site, their presence in discrete archaeological
features and strata, and the excellent preservation of plant and
animal remains should draw the attention of shell midden
archaeologists everywhere. It is unfortunate that all of the
sites reported on by Greg in recent years have been or will
soon be impacted by construction activities. One must wonder
what would be produced from these sites if the archaeologists
were not always working just one step ahead of the bulldozer.
It certainly seems as if the evidence of ritual feasting and the
discovery of European items in a probable aboriginal
nonmortuary context as described in this issue should stimulate
long-term research and preservation efforts in this somewhat
overlooked area of panhandle Florida.
I almost grouped the articles by the Hardmans and Tesar
into a special section called "Lithic Conundrums." Each
author presents a riddle--preform or point? bola stone or club
head? knife or point?--and leaves the reader to ponder the age-
old questions of form and function. Clearly, as Tesar
demonstrates, no matter how far we have yet to go, we are
well beyond the days when intuition alone was enough to
assign functions to artifacts.



Vol. 47 No. 3



Gregory A. Mikell

During June and July of 1993 two archaeologists and a
motley crew of 20 high school students undertook the daily
trek to 8WL38, a Fort Walton-period village site in Walton
County, Florida. As it turned out, the village site contains an
impressive protohistoric component. During the hectic four-
week field school, much was accomplished with the excavation
of 44 1 m square units and a wealth of information was
Okaloosa-Walton Community College, through funding
provided by a Florida Department of Education grant, offers
gifted and talented high-school students opportunities in
diverse areas unavailable in the local school system's curricula.
The annual field school program known as "We Dig Northwest
Florida" is designed to provide a unique and educational
experience to high school students as well as to allow for
research to be conducted on important archaeological sites
located on private property threatened by development and loss
to erosion.
Not only did the 1993 field school initiate the first
systematic investigations on the site, but the excavation
program allowed for thorough sampling of the village area,
documentation of several cultural features, and the recovery of
vital data that would have been lost to the bulldozer's blade.
With the recovery of a small number of Spanish artifacts as
well as Fort Walton cultural material from securely dated
protohistoric contexts, 8WL38 has also yielded a picture of life
as it was just prior to and during the initial arrival of the
Spanish invasion.

The Site

8WL38 is a large Fort Walton village and midden site
known to archaeologists and the local community since its
initial discovery and recording by Dr. Charles Fairbanks in
1958. William Lazarus (n.d) and Scott Nidy (1974:5) have
also made visits to the site and recorded information about it,
but there are no other recorded investigations of the site.
8WL38 was initially recorded as two sites: 8WL38 and
8WL39, but is now considered to be a single site. 8WL38 is
located on the shores of the Choctawhatchee Bay on the west
side of Four Mile Point north of Sandestin.
The site is situated primarily in a mixed live oak, water
oak, hickory, and magnolia hammock surrounded by pine
flatwoods. 8WL38 is bisected by a small, unnamed creek that

drains an extensive pine flatwoods swamp located to the east
and southeast of the site in the central portion of the point.
The water in the creek occasionally runs clear, indicating the
presence of a spring somewhere along it course. This spring
may have been important to the inhabitants of 8WL38 because
potable water is scarce on Four Mile Point. Local residents
report that as recently as 50 years ago, oysters could be
harvested from the area where the creek enters Choctawhatchee
Bay. The same was probably true 400 to 700 years ago as
Fairbanks (1958) reported what he believed to be a "shell
ring enclosure" present on the site. Scott Nidy reported a
similar feature on the site in 1974, but he describes it as four-
foot high, 50-foot long "crescent-shaped shell midden." The
present investigations indicate that Nidy's description is more
accurate, but his observation of a crescent shape also appears
somewhat exaggerated. The midden mound is more like a
linear ridge, but it does have a slight, west-facing crescent
shape. In any case, this midden mound is a central feature of
the site. Although two additional, but much lower midden
ridges were documented and partially excavated with positive
results, excavations in the midden mound proved to be the
highlight of the field school.
As illustrated in Figure 1, 8WL38 is a large area
containing distinct shell midden deposits centered around the
midden mound area (Feature 1). The site lies along a low,
sandy rise located on the west side of Four Mile Point, 2 to 3
meters above the mean high tide line of Choctawhatchee Bay.
The Bay forms the actively eroding western edge of the site.
The site extends up to 150 m inland from the bay shore.
Numerous discrete shell middens are recognizable across the
site. They contain concentrations of ceramics and higher
densities of cultural materials compared to other areas of the
site. For the most part, these middens represent domestic
refuse deposits or distinct activity areas that are, in some cases,
temporally distinct. With the exception of the midden mound,
each midden area investigated is generally 25 to 50 cm thick,
composed of black to dark grayish-brown sand and oyster shell
along with a few other shellfish species, and is rich with
vertebrate faunal remains. While numerous pottery sherds are
present in each midden area, only Fort Walton and Pensacola
ceramic series types are present in the overall assemblage of
Although shovel testing and limited test excavations had



Vol. 47 No. 3


been previously completed in portions of the site, with the
field school the entire site area was investigated and several
new discoveries were made. First, 8WL38 has an extensive
protohistoric component. The site was a Fort Walton village
was occupied between A.D. 1200 and A.D. 1550. A small
number of Spanish metal artifacts were recovered from, and in
areas adjacent to, the midden mound. The midden mound and
associated features within it produced the carbonized remains
of maize, a huge assemblage of faunal remains, and large
amounts of pottery. There is evidence that the midden mound
materials may be the remains of ritual feasting and ceremonial
activities. Finally, several shell beads, ceramic disks,
decorated and worked pieces of bone and shell, and items of
personal adornment were recovered from various areas of the

Field Methods

During the present investigations, 44 1 m by 1 m units
were excavated. Prior testing of the site consisted of the
excavation of 50 cm by 50 cm shovel tests in a 30 m grid along
the eastern and southern portions of the site area. Based on the
shovel test results and the presence of exposed shell midden,
excavation units were placed primarily in areas that appeared
to have potential for producing useful data. Each 1 m square
excavation unit was combined with others to form small blocks
of contiguous units ranging from 1 m by 2 m to 2 m by 4 m in
size. Units 1 m by 2 m in size were used in various areas to
test for "intactness" of subsurface deposits. Larger excavation
blocks were utilized to sample areas known to contain intact
deposits. A 1/8 inch steel probe was utilized to determine the
horizontal spatial limit of shell middens identified by
subsurface testing or the presence of surface materials.
Standard archaeological methods were used in conducting
and recording excavations. The majority of the excavated
matrix was screened through 1/4 inch mesh, but all features
and midden deposits were sampled for flotation and
radiocarbon-datable materials. Excavation proceeded in 10 cm
levels, with features and distinct midden deposits excavated
and documented separately. Features were recorded, sectioned,
and excavated as subunits immediately following recognition in
excavation floors or on unit walls. Aside from shell samples
taken for later laboratory analysis, shellfish remains were
sorted by species and/or other analytical category, weighed,
noted, and discarded in the field as backfill.

The Excavation Units

Figure 1 shows the location of all excavation units
completed during the most recent phase of site investigations
(Units 1-44) as well as previously excavated shovel tests (not
numbered). All cultural deposits documented through our
excavations are summarized in Table 1 (Appendix A). A more
detailed discussion of the nonmidden cultural features and the

various classes of artifacts and other data then follows. In
Table 1, proportionality of shellfish remains is based on shell
weight for the various species and categories.
The midden mound is a central feature of 8WL38 and as
such will be further detailed along with other features in the
following section. For now, suffice it to say that the mound
represents a locus of intense utilization and deposition of
refuse. It is clearly either an intentionally constructed or
intentionally accumulated area of the site that has a unique set
of characteristics, whether it is simply a conspicuous trash pile
or a platform for a structure.

The Features

Aside from shell midden and black sand midden deposits,
which may be considered cultural features themselves, seven
nonmidden features were documented. The seven features
consist of the midden mound and a distinct fire pit or hearth
within its summit, one fire pit or trash pit, four postmolds, and
an ash lens. Aside from the midden mound, each of these
features was located beneath a shell midden deposit. No other
features were discernable within midden deposits, but a few
distinct concentrations of artifacts were noted. Although the
concentrations of artifacts noted within shell midden deposits
do not constitute formal archaeological features, they will be
duly noted in this section.

The Midden Mound

Feature 1 is the midden mound located roughly in the
central portion of the site (Figure 1). The midden mound is
considered as a distinct feature because it is a unique physical
feature on the site and it contains data and artifacts which
imply that it represents the remains of a distinct activity area
within the village site.
the mound measures approximately 30 m (north-south) by
10 m (east-west) and rises 75 to 85 cm above its immediate
surroundings (see Figure 8). An area of least 40 m by 40 m on
the east- central portion of the midden mound has been
disturbed by shell removal. The mound is situated on a
crescent-shaped rise that curves around to the west to the
southwest of the mound itself. On the western terminus of this
rise there is a smaller raised midden area that was documented
with Excavation Block 5. Whether or not the crescent-shaped
rise is a natural feature or a constructed base on which the
midden mound and smaller midden were placed is not certain,
but there is a gap in shell midden deposits between the mound
and the smaller midden and the mound is much more
stratigraphically complex.
The character and content of the stratified deposits that
make up the midden mound are described in Table 1 and
illustrated in Figures 2 and 3. A brief summary, however,
will aid in an understanding of how the mound accumulated.
As illustrated in Figure 3, the uppermost midden deposits


, /
/ / i

0 0


e 41-42 ': "


0 40

Figure 1. Map of 8WL38.

1 mxl m Test Unit with recovery
50cm x 50cm shovel test with recovery
50cm x 50cm shovel test, no recovery
Shell midden
Raised shell midden




. 0



consist of an 8 to 12 cm thick midden layer of crushed and
whole oyster shell in black, organically rich sand (Stratum 2)
and a 12 to 16 cm thick shell midden deposit that contains
primarily whole oyster shell (Stratum 3). These two layers of
shell midden are considered contemporaneous because they are
differentiated only by characteristics of the oyster shell in each.
Each stratum contained extremely dense deposits of vertebrate
faunal remains, charred plant remains, and pottery sherds, as
well as numerous other types of artifacts. Stratum 3 also
contained an intrusive fire pit or hearth-like feature (Feature
Three distinct shell midden strata make up the remainder
of the mound deposits. Stratum 4 (Figure 3) is a 8 to 23 cm
thick, organically rich, black sandy midden deposit that
contains a moderate amount of oyster and scallop shell and
vertebrate faunal remains. This deposit also contains a large
amount of broken pottery and other artifacts. Stratum 4 is the
upper-most cultural strata to cover the entire mound. Stratum
5 (Figure 3) is a relatively thick layer of shell made up of
whole oyster shell for the most part. Organic materials, soil,
and artifacts are very sparse in Stratum 5, at least relative to
the surrounding matrix strata. It appears that this layer of shell
was laid down over relatively short period of time and may
have been shell brought in from other areas to initially provide
a platform for building up the mound. This layer of "clean"
shell is confined to the mound and the relative absence of
charred plant and faunal remains and the low density of pottery
within it lend support to such a contention. Stratum 6 is the
deepest cultural stratum encountered in the mound. Stratum 6
is an 11 to 17 cm thick layer of black, sandy shell midden rich
in charred plant and vertebrate faunal remains. Shell in this
midden deposit is primarily whole oyster and it contains a
moderate amount of broken pottery and a few other types of
artifacts. The density of cultural material in this midden does
not begin to approach the density of materials in Stratum 2
through Stratum 4. With the exception of a few scattered
pieces of shell and pottery, cultural materials abruptly
disappear at the base of the midden deposits as they grade into
sterile sand subsoils.
Whether or not the strata present in the midden mound
represent building stages or are simply distinct periods of
accumulation is not entirely clear. They probably represent
both. A summary of the mound stratigraphy from initial to
deposition may be instructive here. Stratum 6 is the basal
midden deposit in the mound structure and surrounding area.
Charcoal from Stratum 6 produced the earliest radiocarbon date
associated with the mound, approximately A.D. 1283. This
deposit appears to be a midden that began to accumulate prior
to the building of the mound. The lowest midden deposits in
Excavation Blocks 4 and 5 (see Figure 1) are very similar to
Stratum 6 in sediment color and texture and in terms of shell
remains density, species composition, and grain size (see Ford
1992:283-324). Other cultural materials recovered from these
middens are also similar in character and the radiocarbon date

obtained from Feature 3 (A.D. 1238), which is located just
below the middens in Block 4 overlaps with the range of the
date obtained from Stratum 6 for the base of the mound
(Appendix A, Table 2).
A layer of "clean" shell (Stratum 5) was then apparently
piled on top of the existing midden. Stratum 5 is the only
deposit that appears to be primarily a building stage and may
be shell carried in to initially provide a platform to build up
the mound. The layer of shell was then covered by a mantle of
shell midden refuse. The mantle of midden material, Stratum
4, is the uppermost cultural stratum to cover the entire mound.
Like the basal midden layer, Stratum 4 appears to extend into
the immediate surrounding area as it is very similar in
composition to the upper shell midden deposits documented in
Excavation Blocks 4 and 5. The top of the mound contains
two layers of shell midden (Stratum 2 and 3). Both are
confined to the summit of the mound and appear to be either a
cap or the final large depositional episode.
No human remains or evidence of interments have been
encountered in the midden mound, so it is apparently not a
small burial mound. Such a possibility cannot be ruled out,
however, until more of the mound has been excavated. Given
the present evidence, it appears as though the mound is a
prominent accumulation of refuse that was intentionally piled
up over an existing midden deposit. The mound may have
held a structure, but no evidence of one was discerned within
or below the midden deposits that make up the mound.
Artifacts recovered from the midden mound, especially those
recovered in the upper three midden layers, provide some
evidence as to its nature and function. Other features, to be
discussed later, may also be pertinent to the formulation of the
Excavation Block 1 (8 square meters) produced a plethora
of cultural remains from within the midden mound,
particularly the upper 40 cm. Innumerable pieces of burned
and unburned animal bone, 669 pottery vessel fragments,
ceramic discs, a ceramic adorno, a Busycon shell drinking cup,
a Busycon shell spatula, shell and bone tools, shell beads and
pendants, a greenstone celt fragment, a stone disc (possibly a
crude chunkey stone), a few stone tools, chert and gar-scale
projectile points, a brass buckle, and a hugh amount of shell
were excavated there. The artifacts that make up this long list
are detailed below. These midden deposits lay on top of a
layer of "clean" shell (a possible mound platform) and contain
the accumulated refuse of intensive and varied activities, some
of which may have involved ritual and ceremony.
The density of shell, vertebrate faunal remains, and
charred wood and other plant remains within the upper 30 to
35 cm of the mound is simply incredible. I have worked many
shell middens, but the midden mound at 8WL38 contains the
densest deposits of faunal materials that I have ever seen. To
illustrate this point, Table 3 (Appendix A) summarizes the
materials recovered from a 100 liter flotation sample taken
from Stratum 3 in Test Unit 35, adjacent to Feature 2 (Figure

Feature 7

0 40

North Wall Profile

TU 1




Figure 2. North wall profile, Unit 1/2.

Figure 3. North wall profile, Excavation Block 1.


North Wall Profile

TU 36 TU 35 Stratum

Feature 2


j ;:'.:~.:-:::~: ~~ ~':r-;- Vill

0 80

TU 2


Figure 4. North wall profile, Excavation Block 2.

TU 22


0 40

West Wall Profile
TU 20

.. ~ ~ . ... .

Figure 5. West wall profile, Excavation Block 3.

North Wall Profile

TU 15 TU 13 TU 11



o 80
2 E m

TU 18

I 1

=-K= C


Figure 6. West wall profile, Excavation Block 4.

Figure 7. East wall profile, Excavation Bock 4.

West Wall Profile
TU 28 TU 27 Stratum


Feature 6 4IV

0 40



Block 4
Block 1 Bk
3 365 282 2
80o" |5 6 33

disturbed area

Block 5 1 2

100cm 0N


0 10

Figure 8. Midden Mound area contour/plan map (Feature 1).


3). As the table illustrates, oyster is the dominant shellfish
species present and fish represent the majority of the faunal
remains. Fish remains were often encountered in
concentrations. It seemed as if the concentrations were the
"bottom of the pot" residues of many fish stews dumped on to
the mound. In contrast to interior riverine Mississippian sites
where there is ample evidence of maize cultivation, the
presence of maize in the mound at 8WL38 is also worthy of
note. The occurrence of maize in small quantities is not
entirely unexpected given the fifteenth to sixteenth century
time frame for the upper portion of the mound and the growing
body of evidence for the use of maize at Fort Walton and
Pensacola sites on the Florida Gulf Coast (Lee and Joy 1989;
Mikell 1990; Thomas and Campbell 1990).
The pottery assemblage from the upper three midden
layers in the mound is typical for a middle to late Fort Walton
site on the Choctawhatchee Bay (Mikell 1992). A few late
types, such as Leon Check Stamped, Jefferson Plain ware,
Lake Jackson Plain (variety Jefferson), and Lamar Complicated
Stamped are present along with more common Fort Walton
types (Appendix A, Table 6). Pensacola Series pottery makes
up 15.9 percent of the assemblage recovered from the upper 50
cm of the mound. The pottery assemblage recovered during
the most current excavations compares favorably with
previously documented collections (Mikell 1992:59-61),
although it is now obvious that 8WL38 is dated to both the
Indian Bayou and Four Mile Point phases.
Also of note in the upper portion of the mound is the
presence of a fire pit or hearth (Feature 2). Feature 2 is an
intrusive, somewhat basin-shaped pit contained primarily
within Stratum 3, but extending 8 cm into Stratum 4 (Figure
2). As Figure 9 illustrates, the pit is a maximum of 51 cm
deep and approximately 110 cm wide in the east wall profile of
Block 1. The pit contains several facies of burned oyster shell
and charcoal, as well as concentrations of burned and unburned
animal bone and pottery sherds. Feature 2 appears to be a pit
that was dug as a fire pit and receptacle for refuse that was
burned. Oyster shell around the feature is charred and
thermally altered up to 12 cm around the confines of the top of
the pit. The various facies of charcoal and burned shell
indicate that it was used on more that one occasion or for more
than a single, short term event. In plan view, Feature 2 is
oval-shaped, measuring 38 cm by 107 as it was exposed in
Test Unit 35 at 13 cm below the mound summit. The
unexcavated remainder of the pit was shored up with plastic
lined plywood prior to back filling the excavation block.
Table 4 (Appendix A) details the cultural material
recovered from the Feature 2. The wood charcoal and other
botanical remains presented in Table 4 are those recovered in a
37 liter flotation sample taken from the fire pit or hearth. The
pottery and faunal remains figures include specimens from the
flotation sample as well as those recovered manually during

A sample of charcoal taken from Feature 2 produced a
calibrated radiocarbon date range of A.D. 1468 to A.D. 1552
(Appendix A, Table 2). This range of dates is of interest not
only because it establishes the age of the fire pit and its
contents (primarily the pottery types), but it also indicates a
protohistoric context for the upper midden deposits in the
mound. The aforementioned brass buckle was recovered less
than 12 cm west of the edge of Feature 2 at 23 cm below the
mound summit. Given the sixteenth century date of Feature 2,
the association between the buckle and the fire pit indicate that
the buckle is Spanish. This date also suggests that the buckle
can be most likely associated with the presence of either the
Panfilo de Narvaez or the Hernado de Soto expeditions in
northwest Florida as a source of (Hann 1988c). The original
source of the buckle will always be a matter of speculation, of
course, since it is not a specific temporally diagnostic artifact.
A rust encrusted, thin iron band fragment and a small hand-
wrought nail were recovered from the base of Stratum 2 in
Excavation Block 4 in Units 27 and 28, less than 10 meters
from Feature 2. The iron band and hand-wrought nail are
quite likely Spanish as well. Clarence B. Moore (1918: 537-
539) also recovered Spanish artifacts at the "Aboriginal
Cemetery at Hogtown Bayou" (8WL50) which is located
approximately 2 km to the southeast of 8WL38.
Curiously, a group of 22 stingray (Dasyatidae) spines, an
engraved ceramic disc, two Olive shell beads, an altered
alligator scute, a small ceramic disc, and a cut and burned
freshwater turtle shell fragment that may be a piece of a gorget
(Figure 10) were found in a cache-like association with the
brass buckle. It is possible that this group of artifacts,
including the buckle, is the remains of a medicine bundle or is
some sort of ceremonial cache. As can be seen in Figure 10,
the ceramic disc is engraved with what is either an equal-armed
cross or a Christian cross. The association of the disc and the
buckle may be taken to suggest that the disc is engraved with a
Christian cross, but as with the source of the Spanish
materials, I can only speculate.
The lower midden layer (Stratum 6) in the mound
produced the type and density of artifacts more comparable to
other middens on the site. Two characteristics of the lower
midden deposit in the mound are noteworthy here. First, a
charcoal sample taken from charred wood recovered in the
lower midden yielded a calibrated date range of A.D. 1278 to
A.D. 1387. This date does not overlap with the date obtained
from Feature 2 and indicates that the initial deposits in the
midden mound area accumulated at least 80 to 100 years prior
to the final use of the mound. Second, although it would be
rewarding to see some notable variation between the ceramic
assemblages in the upper and lower portions of the mound,
there are no significant differences. There are, however, no
late Fort Walton pottery types in the mound below Stratum 5
and the amount of Pensacola Series pottery falls off only
slightly. It would appear, based on the ceramic evidence, that


Figure 9. Profile of Feature 2 in Block 1, East Wall.


1* L" *" s N 41 .' t .1 -_

Figure 10. Artifacts recovered as a cache adjacent to Feature 2. Top row: stingray spines; middle: sharks tooth, small ceramic
disc; bottom row (l-r): turtle shell gorget fragment, brass buckle, engraved ceramic disc, altered alligator scute, Olive shell beads.

East Wall Profile
TU 35 Stratum


burned shell



0 20

, o =




Block 4
Plan View at 35cmbs TU 25 TU 23

TU 27

.- Features
TUTU 2 8 TU 26 TU 24

Figure 11. Excavation Block 4 plan view at 35 cmbs.
TU 28TU 26TU 2
o: 60:~
I: :cm-

Figure~ ~ ~~' 11.r Exaato lok4la iw 3 cas

Figure 12. Selected sand-or grit-tempered pottery. Top row (l-r): Point Washington Incised var. Point Washington, var. Hogtown
Bayou; bottom row (l-r): Point Washington Incised var. Point Washington (2), var. unspecified, var. Griffith, var. unspecified.



the same people built the midden mound from sterile sand up.
Feature 3 is a 34 cm deep, basin-shaped fire pit or trash
pit located at the base of and immediately below the midden
deposits in the east end of Excavation Block 4 (TUs 23 & 24).
A profile of Feature 3 is illustrated in Figure 7. The charcoal
rich pit is roughly circular in plan view and measured 64 cm
by 62 cm where it became recognizable in the basal portion of
the midden, narrowing to 34 cm by 38 cm near its flattened
base. The black sandy pit fill was full of charred wood and
other plant remains, burned or partially burned animal bone, a
small number of pottery fragments, and a scattering of burned
shell. Table 5 (Appendix A) presents a summary of the
materials recovered from a 35 liter flotation sample and all of
the pottery taken from Feature 3.
The primary significance of Feature 3 is that it is
associated with four postmolds also uncovered in Block 4 and
that a charcoal sample taken from the pit yielded a calibrated
date ranging between A.D. 1209 to A.D. 1310 (Appendix A,
Table 2). The date range for Feature 3 overlaps with the date
range of A.D. 1278-A.D. 1387 associated with the base of the
midden mound and, in part, suggests that the two are part of
the same depositional episode or are at least temporally related.
Features 4, 5 and 6 are postmolds associated with Feature
3 at the base of the midden in Block 4 (Figure 11). Feature 4
is a 36 cm deep, circular postmold measuring 23 cm in
diameter where it became discernible, which tapered to a
rounded end. The postmold itself was a very dark brown to
black stain with a dark gray shadow. The stained matrix,
particularly the gray shadow, contained oyster shell that may
have been used as chinkingg" as well as a single piece of grit
tempered plain pottery and charred wood fragments. The
charred wood in this and the other postmolds may indicate that
the posts burned while still in the ground. Feature 5 is 40 to 44
cm deep, oval-shaped, double postmold measuring 23 cm by
38 cm, that is otherwise similar to Feature 4 in color and
texture. The double postmold shadow contained two plain
sherds and the darker core contained charred wood. Feature 6
is a 38 cm deep circular postmold measuring 22 cm in
diameter. This postmold contained two plain sherds and
charred wood. A profile of Feature 6 is illustrated in Figure 6.
Features 3 through 6 may represent, in part, the remains
of a structure. The postmolds form a line that may represent a
wall and as Figure 11 illustrates, a shell midden deposit lies to
one side of the line formed by the postmolds. If these features
are part of a structure and if the fire pit is located on its
interior, the midden, as may be expected, lies outside the
structure. A large Point Washington Incised rim sherd was
also recovered from what would be the interior of the
structure. No wall trench was discernible in the sandy soil into
which these features were excavated. The absence of a wall
trench may argue against the presence of a structure. A
determination of whether or not the features actually do
represent a structure will have to await further excavations.

Feature 7 is an ash lens, oval in plan view (61 cm by 43
cm), and 12 cm thick at its maximum. Figure 2 illustrates this
feature in profile. The ash lens was encountered at the base of
the shell midden on the slope of the midden mound in Test
Unit 1. The ash lens contained wood charcoal and charred
cranial and vertebral fragments from a alligator that were
covered by sherds from a large fragment of a Lake Jackson
Incised (variety Blounstown) pottery vessel. A sample of
charcoal taken from the ash lens yielded a radiocarbon date
ranging between A.D. 1452 and A.D. 1537, a date consistent
with the upper portion of the midden mound.
An altered and charred alligator dermal scute was also
recovered from the shell midden adjacent to Feature 7. The
scute had been ground along its edges into a roughly circular
shape prior to being partially burned (Figure 19).
Interestingly, other similarly altered dermal scutes were
recovered in Blocks 1 and 3. The altered scutes and charred
alligator bone in Feature 7 certainly invites interpretation as
being a part of some sort of ritual activity in which alligators
had some importance.

The Artifacts Recovered

Long-term prehistoric occupation of 8WL38 is not
indicated by either the artifact assemblage or the radiocarbon
dates. As a single component village that, according to the
radiocarbon dates, may have existed for about 300 years
(Appendix A, Table 2), 8WL38 offers a look at life on the
Choctawhatchee just before and during the early days of the
Spanish invasion of Florida and the Southeast. As such, the
materials from the site deserve careful consideration. Shellfish
remains, vertebrate faunal remains, and pottery make up the
vast majority of artifacts recovered, but many other types were
also recovered. A few notes on analysis methods are discussed
as an introduction to the description of each class of artifacts


A total of 4,676 pottery sherds and vessel fragments was
recovered from the site during the field school investigations.
An account of all ceramics recovered from the controlled
excavations is presented in Table 6 (Appendix A). Table 6 is
organized by excavation block as well as by more specific
proveniences such as level, and feature. Selected examples of
pottery from 8WL38 are illustrated in Figures 13 through 16.
The ceramic analysis is based on two existing type-variety
typologies that are, in part, applicable to the Choctawhatchee
Bay region. Fuller and Stowe's (1982) typology for shell
tempered ceramics in the Mobile Bay and Mobile-Tensaw
Delta region covers the Pensacola and related ceramics quite
well. John Scarry's (1985) type-variety system for Fort
Walton ceramics in the Tallahassee Hills and Apalachicola


Valley was used for the Fort Walton Series pottery. It became
evident during the course of analysis, however, that there are
several varieties of the Fort Walton and Point Washington
Incised types that occur in the Choctawhatchee region that are
not defined by Scarry. A type-variety typology for the
Choctawhatchee region would be preferable, but for now a
meshing of the two applicable type-variety schemes is utilized.
One new type is proposed in this report because it cannot
be pigeonholed into an unspecified variety of any type. This is
a sand or grit tempered ware decorated with small, circular,
smoothed to somewhat pointed beads applied to or pinched up
from the exterior of the vessel while wet. The beads occur in
broad bands around the rim on small carinated and collared
bowls and jars or cover the entire exterior of small jars. Rim
sherd examples of this type are illustrated in Figure 13. I am
calling this pottery type Fort Walton Beaded, variety Walton,
for the beaded and dimpled appearance that the decorative
technique gives to the vessel surface and for Walton County,
Unless otherwise noted here, the classification of pottery
follows that of Willey (1949), Fuller and Stowe (1982), and
Scarry (1985). A few other aspects of the ceramic analysis
should be noted. First, I am referring to Fuller and Stowe's
Bottle Creek and Bear Point Complex pottery types and
varieties as Pensacola Series pottery. Although use of the term
"Series" is not appropriate in the type-variety-mode-complex
classification system that both Fuller and Stowe (1982) and
Scarry (1985) adopted from Phillips (1970), it is retained here
until a similarly detailed classification system for the
Choctawhatchee Bay region is developed. Second, Lake
Jackson Plain sherds are identified only when a rim segment
was present or when body sherds were obviously associated
with a vessel from which a rim segment was recovered.
Varieties of Lake Jackson Plain were sorted according to
Scarry's temper type and size criteria as were the shell
tempered Bell and Mississippi Plain types and varieties.
Following Fuller and Stowe (1982), Bell and Mississippi Plain
subsumes what I have formerly classified as Pensacola Plain
(Mikell 1992). Third, because no rim segments of grog-
tempered Jefferson Ware vessels were recovered (they are all
grit tempered), all grog-tempered sherds are classified as Lake
Jackson Plain, variety Tallahassee or variety Jefferson
following Scarry (1985:221). The distinction between many
small Fort Walton and Point Washington Incised (variety
unspecified) sherds that are otherwise not assignable to a
specific type-variety is based on the presence (Fort Walton) or
absence (Point Washington) of punctations as part of the
decorative motif. Fourth, many of the sherds identified as Lake
Jackson Incised (variety Blountstown) are fragments of
shallow, plate-like bowls with a single incised line executed on
the interior of the vessel inside the rim. Scarry (1985:220)
describes variety Blountstown as including "all bowls and
beakers decorated with one or more broad horizontal lines",
but at 8WL38 only two forms of this variety occur, the plate-

like bowls mentioned above and carinated, collared, cazuela,
flaring rim, and simple restricted bowls with two horizontal
incised lines on the exterior of rims or collars. Sixth, several
sherds classified as Marsh Island Incised, variety Columbia
follow Scarry's (1985:226) criteria except that they are
decorated with arcs in a manner similar to that described by
Fuller and Stowe (1982:56-60) for the Arnica and Mary Ann
varieties of D'Olive Incised. Finally, the category
"unidentified incised" consists of small or eroded sherds
lacking enough decoration to assign to a type classification
with any measure of confidence.
As Table 6 (Appendix A) illustrates, all of the pottery
recovered from controlled excavations at the site belongs either
to the Fort Walton or Pensacola Series. A previously analyzed
assemblage of 364 sherds indicates the same (Mikell 1992:59)
and all pottery that I have seen on other collections from the
site, is also Fort Walton or Pensacola. Fort Walton Series
types included in the assemblage are: Carrabelle Punctated,
Cool Branch Incised, Fort Walton Decorated, Fort Walton
Incised, Jefferson Ware, Lake Jackson Incised, Lake Jackson
Plain, Lamar Bold Incised, Lamar Complicated Stamped, Leon
Check Stamped, Marsh Island Incised, Ocmulgee Fields
Incised, Point Washington Incised, and residual plain.
Pensacola Series types include Bell and Mississippi Plain,
D'Olive Incised, Moundville Incised, Mound Place Incised,
Pensacola Incised, and Carthage Incised. Table 7 (Appendix
A) presents proportionality data for the 8WL38 assemblage.
With the exception of the distribution of a few relatively
infrequently occurring types, most of the types and varieties
appear to be fairly evenly distributed across the site.
Table 7 demonstrates that, in terms of both a proportion
of the entire assemblage and a portion of all decorated
ceramics, Fort Walton types and varieties account for
approximately 75 percent of the pottery recovered. Given the
protohistoric dates associated with certain features at 8WL38,
the relatively high percentage of sand and grit-tempered
pottery is somewhat surprising. Even when the protohistoric
deposit assemblages are separated from the overall assemblage,
Fort Walton pottery dominates shell-tempered Pensacola
wares. The pottery from 8WL38 certainly indicates the
persistence of the sand and grit tempered ceramic tradition that
characterizes Fort Walton (Mikell 1992).
A variety of vessel shapes are present in the 8WL38
ceramic assemblage. Although no whole vessels were
recovered from the site, enough large sherds, particularly rim
and collar sherds, were recovered to provide a list of vessels
forms. Carinated, collared, cazuela, flaring rim, simple
restricted, and shallow plate-like bowls, collared and everted
rim jars, beakers, and bottles are present in the assemblage.
Collared vessels, shallow bowls, simple restricted bowls, and
everted rim jars are the most frequently occurring forms.
Incised decorations occur primarily on the various forms of
bowls, while plain wares are usually jars. There are
exceptions to this statement, but the pattern generally holds


true for both Fort Walton and Pensacola Series types. The few
beaker and bottle sherds recovered are incised.
Several rim and appendage modes can also be identified
and described for the assemblage. Selected examples are
illustrated in Figures 13 through 16. Aside from vertical
nodes, lug handles, and strap handles commonly found on
Lake Jackson Plain, Lake Jackson Incised, and Cool Branch
Incised vessels in northwest Florida (Jones 1973; Scarry 1985)
and rim modes commonly associated with Leon-Jefferson
Complex assemblages (Smith 1951; Jones 1973), the more
frequently occurring rim modes from 8WL38 share a great deal
in common with the rim modes described by Fuller and Stowe
(1982:84-86) for shell tempered ceramics. Fuller and Stowe's
"Bear Point", "Gasque", and "D'Olive" rim modes are very
common within the 8WL38 assemblage, and they are found on
shell, sand, and grit tempered vessels. In a sense, and given
some latitude in terms of variation, these rim modes cross-cut
the Fort Walton-Pensacola dividing line.
The most common Fort Walton rim mode is the "ticked"
or notched rim originally described by Willey (1949). At
8WL38, notched rims are the most common rim mode for
incised and plain pottery. The range of variation in notching
runs from larger, cruder notches to the very fine notches more
often associated with the Bear Point mode. This range of
variation in notching is also present on Pensacola Series
ceramics, but the "classic" Bear Point rim mode is the most
common form on decorated, collared and plate- like, shell-
tempered ceramics at 8WL38.
The Gasque rim mode, as defined by Fuller and Stowe, is
also common at 8WL38. Gasque rim modes occur on
Pensacola and Mound Place Incised sherds, but are also quite
common on Point Washington Incised sherds. D'Olive rim
modes occur on both D'Olive Incised and Marsh Island,
variety Columbia sherds from plate-like vessels recovered from
the site. Plate-like vessels are apparently the only vessel forms
that have the D'Olive rim mode at 8WL38.
The Douglas rim mode (Fuller and Stowe 1982:86) shares
general similarities with some of the nodes, small lugs, or
pinching on Lake Jackson Plain, Lake Jackson Incised, Cool
Branch Incised, and Jefferson Ware vessels, but the Douglas
mode's range of variation is more limited than that for the Fort
Walton pottery. Vertical lugs and nodes on Lake Jackson
Plain and Incised sherds are fairly common at 8WL38 and a
few strap and loop handles also occur on Lake Jackson Plain
sherds. For shell-tempered pottery, the occurrence of lugs and
nodes is more limited and no sherds with loop or strap handles
were recovered. Two plain shell-tempered rim sherds are
Douglas rim modes and five shell-tempered sherds have more
widely spaced nodes; this compares with 16 various Fort
Walton type sherds that have some form of handle or nodes.
In the 8WL38 assemblage, rim and collar appendages
other than handles, lugs, and nodes consists of rim beading and
applique strips on Jefferson Ware and Lake Jackson Plain rim
sherds and six rim effigy adornos on Point Washington and

Fort Walton Incised sherds. The rim effigies present in the
8WL38 assemblage are fairly typical Fort Walton zoomorphic
rim adornos, consisting of stylized frog and bird heads, and a
stylized eagle head (Figure 16). In each occurrence, the rim
effigy points away from the interior of the vessel, which is
usually some form of bowl. Rim effigies on shell tempered
vessels usually face the interior of the vessel (cf., Fuller and
Stowe 1982:87). Thus far, no shell-tempered sherds from
8WL38 with rim effigies have been recovered.

Ceramic Discs

Eleven complete or partial ceramic discs, or pottery
discoidals were recovered from various proveniences at
8WL38 (Appendix A, Table 6). Excavation Block 1 in the
midden mound produced more of these discs than any other
area investigated, with four being recovered from the upper 30
cm of the mound. Of the 11 ceramic discs, eight were cut and
ground from grit and sand-tempered pottery fragments. The
remaining three are from shell tempered pottery. The ceramic
discs are not uniformly sized and range in diameter from 13 to
41 mm. All but two of the discs have plain, smooth surfaces.
The two largest discs have surfaces that have been engraved or
scratched. One shell-tempered example has numerous, roughly
parallel lines "scratched" on to both surfaces and the largest
disc has an equal-armed cross, reminiscent of a Christian cross,
engraved on its polished interior surface (Figures 10 and 16).

Ceramic Adorno

A curious ceramic adorno was recovered from the upper
portion of the midden mound (Figure 16). The ovate,
teardrop-shaped adorno is made from what appears to be a
smoothed and polished, red filmed or red slipped, plain grog
tempered sherd. The adorno measures 44 mm by 35 mm,
narrowing to 16 mm at its "top." The main body has a
roughly circular, 20 mm in diameter cutout and a smaller (4
mm), off-center hole drilled through its narrower portion. All
edges of this piece are rounded as if they were ground smooth
and reddish pieces of grog are visible on both surfaces. The
function or use of this adorno is not readily apparent, but it
may have been a pendant or some part of another form of
adornment. Its shape and cutout give it the appearance of an
eye and could even be part of a weeping eye motif that was
incorporated into a mask or headdress.

Shell Beads and Pendants

Six shell beads, consisted of three different types, and
four shell pendants were recovered from various contexts
during our excavations (Figure 17). Four of the shell beads
are Olive shells, measuring between 25 and 37 mm in length,
that have been drilled so that they can be strung. The
remaining shell beads consist of a single, semi-circular, center-


Figure 13. Selected sand -or grit-tempered pottery. Top row (l-r): Fort Walton Incised var. unspecified (4), var. Fort Walton;
middle row (l-r): Fort Walton Incised var. Choctawhatchee, variety Cayson, Marsh Island Incised var. Marsh Island (3), var.
Columbia; bottom row (l-r): Marsh Island Incised var. Columbia, Cool Branch Incised var. Cool Branch (2), var. unspecified,
Fort Walton Beaded var. Walton (3).

Figure 14. Selected sand-or grit-tempered pottery. Top row (l-r): Lake Jackson Plain (4), Lake Jackson Incised var. Blountstown
(2); bottom row (l-r): Lake Jackson Incised var. Blountstown (2), Jefferson Ware, Jefferson Ware rim with applique strip on rim ,
Leon Check Stamped, Jefferson Ware.


Figure 15. Selected shell-tempered pottery. Top row (l-r): D'Olive Incised var. unspecified (3), var. Dominic; middle row (l-r):
Pensacola Incised var. Jessamine, var. Pensacola (2), var. Gasque, Mound Place Incised var. Waltons Camp-Incised Tab Tail,
Pensacola Incised var. Moore; bottom row (l-r): Pensacola Incised var. Perdido Bay (2), Moundville Incised var. Snow's Bend,
Moundville Incised var. Bottle Creek.


Figure 16. Ceramic discs, ceramic adorno, and rim effigies. Top row: grit-or sand-tempered discs; middle row (l-r): shell-
tempered discs (3), engraved grit-tempered disc, adorno; bottom row (l-r):frog- like rim effigy, "popeye" rim handle/adorno, eagle
head rim effigy, ceramic vessel foot or rim handle.


drilled whelk shell bead (13-15 mm wide) and a drilled, short
(7 mm) tubular segment of an unidentified shell. Two of the
Olive shell beads were recovered from different domestic
refuse middens, while the other four were recovered from the
midden mound.
Two of the shell pendants are made from whelk or conch
shell columella and the other two are altered and drilled scallop
shells (Figure 17). The columella pendants consist of two
very different forms: a short (27 mm), ground and polished,
celt-shaped piece with a engraved line extending around its
exterior just above its center and a ground, 54 mm long piece
that has been drilled near its proximal end. The scallop shell
pendants were altered in a variety of ways. First, and most
obviously, the proximal end of each shell has been drilled with
one or two small holes and the shell hinge has been cut away.
Second, the proximal, lateral and distal edges have been
ground and smoothed. The proximal end was ground into an
arch-like shape after the hinge was removed. One of the
finished pendants was then rubbed or painted with hematite to
give it a reddened exterior finish. The scallop shell pendants
measure 51 mm in length by 57 mm maximum width and 59
mm (length) by 63 mm. All four pendants were recovered
from the upper 30 cm of the midden mound.

Shell Cup and Spatula

A broken whelk shell drinking cup and a whelk shell
spatula- like object were recovered from the upper portion of
the midden mound (Figure 18). The shell cup measures 209
mm by 150 mm and has no engraving on it, but its interior is
has a stained appearance compared to the exterior. The
exterior horns have been ground smooth. The major breakage
of this cup appears to be that which is evident along its medial
exterior edge. The cup was recovered from the midden
mound, Stratum 3, near Feature 2.
The spatula-like whelk columella and exterior shell piece
(Figure 18) has obviously been shaped by cutting and
grinding. What is not apparent is the function of this artifact.
It bears no evidence of being a pendant and is rather large for
one (192 mm long). The greatest amount of effort expended
on modifying the shell is that spent in shaping the exterior
shell. The horns have been ground off, the whorl has been
cut, and the columella shows no distinct sign of alteration.
Numerous holes made by boring marine clams are present on
the object, which is curious since it was recovered from a
"dry" context. If the holes were present when the shell was
being altered, they may have been taken advantage of for the
manufacturing of a spatula-like strainer.

Shell Tools

A small number of shell tools or shell tool manufacture

byproducts were recovered from various midden areas on the
site. The short list of the shell tools consists of three small,
semi- rectangular quahog shell wedges; a pointed and bevel-
edged quahog shell scraper or wedge; a delicate, 125 mm long
whelk columella awl or potter's tool; and an exterior whelk
shell adze fragment. Shell tool byproducts consist of several
pieces of whelk shell fragments and a broken and cut upper
whelk shell fragment from which the columella was extracted.
The columella awl or potter's tool and the broken and cut
upper whelk shell fragment were recovered from the upper
portions of the midden mound, while the remaining shell tools
and manufacturing byproducts were recovered from various
domestic refuse middens.

Bone Tools

Four classes of bone tools were recovered from 8WL38:
bone awls, bevel-edged bone wedges or chisels, bone projectile
points, and gar-scale projectile points (Figure 19). Three deer
metapodial awls; two mammal longbone wedges, one made on
a deer metapodial; two mammal longbone distal point
fragments; and two gar-scale "arrowheads" complete the list of
bone tools. Two of the bone awls have distal ends worn and
polished from use and the wedges evidence wear consistent
with gouging and pounding prior to discard or loss. Both gar-
scale points and one of the bone awls were recovered from the
midden mound.

Other Bone Artifacts

The category, "other bone artifacts" consists of non-tool
bone artifacts. Four types of artifacts in this category were
recovered: two alligator dermal scutes shaped by edge grinding
and smoothing, 11 unaltered individual shark teeth, a group of
stingray spines, and a burned fragment of a possible turtle shell
gorget recovered from the midden mound adjacent to Feature
2. The alligator scutes may be gaming pieces or "charms" and
their association with the stingray spines, turtle shell gorget
fragment, and the other cache artifacts suggests that they may
have been part of a medicine bundle or ceremonial cache.
Shark teeth were recovered only from the midden mound and
the shell midden in Excavation Block 3 (see Figure 1).

Projectile Points

A single chert projectile point was recovered during our
excavations. The point is a small (25 mm long), tan chert,
stemmed triangular arrow point (Figure 20). This type of
point is rare in the Choctawhatchee region and is similar in
form to Caddoan points from Louisiana, eastern Texas and
Oklahoma, and lower Arkansas (Perino 1968, 1971). The
point was recovered in Excavation Block 4, in an area adjacent


Figure 17. Shell beads and pendants. Top row (l-r): Olive (Olivella) shell beads (4), drilled whelk shell bead, drilled fossil bead;
bottom row (l-r): scallop shell pendants, columella pendants, possible cockle shell pendant fragment.

Figure 18. Shell cup and shell tools. Left to right: conch shell celt, shell scraper/wedge, whelk shell cup, columella awl/potter's
tool (below cup), whelk adze/scraper, whelk spatula.


to the midden mound.

Other Chert Tools

In addition to the chert arrow point, four other types of
chert tools were found. Three utilized tertiary flakes, a
denticulate blade, a microdrill or engraving tool, and a broken
stemmed blade that had been resharpened into a hafted scraper
make up this group of artifacts (Figure 20). The denticulate
blade and the micro drill or engraving tool were recovered
from the midden mound. The utilized flakes were recovered
from various domestic middens as was the hafted scraper.
Although the scraper appears to have originated as a late
Archaic Newnan cluster (Powell 1990:28) or Cotaco Creek
(Cambron and Hulse 1975:17) stemmed blade, it was
apparently resharpened, utilized, and discarded or lost in a
Fort Walton context. As with virtually all chipped stone
artifacts in the Choctawhatchee region, the chert items from
8WL38 are made from nonlocal materials.

Grinding Stones

Nine nondescript fragments of and spalls from grinding
stones and/or anvils made from locally occurring sandstone
were recovered from several domestic refuse middens. One
hand-sized, hemispherical, mano-like grinding stone (Figure
20) was also recovered from the western slope of the midden
mound. The mano- like tool is made from hard, fossiliferous,
crystalline limestone, a nonlocal stone.

Ground and Pecked Stone Artifacts

This category of artifacts includes two fragments or spalls
from ground and polished greenstone celts, a sandstone disc
that resembles a crude chunkey stone, and one-half of a
symmetrical disc made from a light gray pumice-like material
(Figure 20). The possible chunkey stone and a celt spall were
recovered from the midden mound. The occurrence of the
possible chunkey stone and the celt fragments in the mound are
of interest because of their potential as status or ceremonial
items. One celt fragment retains part of a bevelled edge that
has been scarred by heavy pounding and impact that may have
resulted in breakage.

Other Lithic Artifacts

Within this category are 12 chert and quartzite flakes
(debitage) and a piece of dark pumice. The piece of pumice
appears to be unaltered and its function is unknown. The
pumice-like piece and the debitage were recovered from
various middens, but not the midden mound. The assemblage
of flakes consists of three flakes with small amounts of cortex
and nine tertiary flakes. Six of the tertiary flakes are biface
thinning flakes. The chipped stone debitage indicates that

while tool maintenance occurred at 8WL38, these stone tools
were reduced from core materials and manufactured elsewhere.

Metal Artifacts

Three metal artifacts, of apparent Spanish origin given
their sixteenth century context, were recovered from the
midden mound and its eastern slope. These metal artifacts
include a brass buckle, a hand-wrought nail, and an iron band
fragment (Figure 21). The brass buckle, which is bent on one
side and is missing its tongue, is nearly square, measuring 27
mm (width) by 29 mm (length). The buckle compares
favorably with Spanish examples recovered in the Tallahassee
and Pensacola areas (Florida Museum of Natural History and
Florida Bureau of Archaeological Research staff, personal
communications, 1993). It is most likely an equipment strap
or accoutrement buckle. The buckle was recovered within a
few centimeters of Feature 2 (dated to A.D. 1468-1552), at 23
cm below the mound summit surface in association with a
cache of stingray spines, an engraved ceramic disc, Olive shell
beads, an altered alligator scute, small ceramic disc, and a
piece of a turtle shell gorget fragment that were described
In addition to the brass buckle, two other iron artifacts
were recovered from the eastern slope of the mound in
Excavation Block 4 (TUs 27 and 28). A small (4 mm wide by
28 mm long), hand-wrought iron nail was recovered in direct
association with a large Lake Jackson Plain rim sherd in Unit
28 at 19 cm below the surface. In Unit 28, at 22 cm below the
surface and 12 cm south-southeast of the iron nail, we
recovered a 54 mm long segment of a 24 mm wide iron band.
Given the protohistoric component of 8WL38, the
presence of Spanish artifacts does not come as a complete
surprise. Although the extent of the Spanish presence in the
Choctawhatchee Bay area during the middle sixteenth through
the seventeenth century is poorly documented and appears to
have been very limited, some Spanish artifacts show up in Fort
Walton sites around the Bay (Scarry 1990b). C.B. Moore
(1918: 537-539) recovered several Spanish artifacts from some
of the burials at a nearby Fort Walton cemetery (8WL50).
Moore reports finding glass beads, iron objects, and a pair
scissors in a number of burials, some of which he describes as
urn-burials (Moore 1918:538). The protohistoric components
and particularly the presence of metal artifacts at 8WL38 and
8WL50 certainly invites speculation as to the relationship of
the two sites.

Faunal and Botanical Remains

Invertebrate Fauna

Remains of numerous shellfish and estuarine fish species
comprise the vast majority of faunal remains at 8WL38.
Shellfish remains are usually a dominant feature of any shell


Figure 19. Bone artifacts. Top row (l-r): altered alligator scutes, turtle shell gorget fragment, gar-scale points, bone point
fragments; bottom row (l-r): bone wedges, bone awls.


Figure 20. Stone artifacts. Top row (l-r): Chunkey stone, stone disc, mano; bottom row (1-r): pumice, utilized flake blade,
denticulate blade fragment, drill/graver, stemmed triangular point, Archaic stemmed blade/end scraper.

Figure 21. Metal artifacts. Left to right: hand-wrought iron nail, brass buckle, iron band fragment.

midden site, but the extent to which shellfish "dominated" the
lives of the inhabitants of shell midden sites is questionable
(Waselkov 1982). The most common shellfish species
represented at 8WL38, by far, is Crassostrea virginica
(Virginia oyster). On average, oysters make up 81 percent, by
weight, of all shellfish remains sampled and it is readily
apparent that every shell midden on the site contains a majority
of oyster shell. There is little doubt that oysters were an
important food resource. The oyster shell assemblage can also
easily provide a good deal of information about the estuarine
environment in the vicinity of 8WL38.
Five samples of 100 whole or nearly whole oyster valves
(shells) were examined for various characteristics. The
samples were taken from five separate midden strata and each
sample consists of 50 upper and 50 lower valves. Some degree
of randomness was assured for the sampling in that the first
specimens meeting shell completeness and upper/lower valve
criteria in each sample unit were taken for data recording. Size
of the shell was not a criteria for sampling. The valves were
examined for the following: valve morphology, length and
width measurements, evidence of organisms that attach to
oyster shells (epibiont analysis), and evidence of preparation
techniques. Aside from finding that the oyster shells deposited
in the upper portions of the midden are, on average, quite a bit
larger than those sampled from other deposits, analysis shows
rather uniform sets of sample data. Therefore, the summary of
the oyster shell analysis is presented as a general discussion
rather than a sample by sample breakdown.
Two types of oyster are present at 8WL38: 1) sand
oysters, which are short and broad, inhabit shallow intertidal
waters, and are found in clusters on beaches and bars of
coarse, firmly packed sand; and 2) bed oysters, which are
generally larger and longer, inhabit mixed muddy sand, and
occur either singly or in loose clusters (Kent 1992). Bed
oysters are dominant in the samples taken; sand oysters
account for only 22 percent of the total of the five samples.
Both sand and bed oysters may inhabit the same brackish
estuary zone, but sand oysters always occur in the intertidal
zone or somewhat shallower and less muddy water than bed
oysters (Kent 1992).
Evidence of attaching organisms includes boreholes made
by Cliona truiiti-type sponges (35 percent of sample),
boreholes made by Polydora worms (9 percent of sample), and
the presence of barnacles and blue mussels (11 percent of
sample) that commonly occur on oysters. Nearly one-half of
the oyster shells sampled (45 percent) showed no sign of
attaching organisms. The presence of Cliona sponges and
Polydora worms on the oysters indicates they were taken from
brackish waters where salinity remained low throughout most
of the year (Hopkins 1956, 1962; Kent 1992). Evidently the
Choctawhatchee Bay was less saline and had more muddy
bottoms during the Fort Walton period than it does today.
Like many Fort Walton-era oyster middens on the
Choctawhatchee Bay, the oyster shell assemblage at 8WL38


suggests that a very healthy population was being exploited.
Compared to late Weeden Island sites in the area, average size
of the 8WL38 oyster shells alone indicates that they grew
rapidly and were free of disease and an overabundance of
parasites. There is no evidence to suggest that intensity of
harvest diminished the resource. There is simply no decline in
mean size or apparent quality when the variously dated samples
are compared. It is quite likely that the creek outflow area in
the bay adjacent to 8WL38 was the site of one oyster bed that
the inhabitants exploited.
Evidence for preparation techniques is limited and, in
fact, there is a preponderance of negative evidence suggestive
of preparation. Of the 500 valves examined, none show
evidence of shucking. Only 16 valves are broken in such a
way as to suggest cracking open with a hammerstone, only 37
are scorched as if they were roasted open, and only 51 are
burned. The absence of typical hammerstones and large stone
flakes for shucking also argues against these two means of
extraction, but according to Kent (1992:43), some of the flat
grinding stone fragments that we recovered may have been
suitable as oyster-cracking hammerstones. Boiling oysters
would leave little or no tell-tale signs. The lack of evidence
for preparation techniques may be indicative of oyster meat
extraction by either boiling or by preliminary roasting in such
a manner that the shells were not often scorched or burned.
The absence of shuck marks and the virtual absence of
evidence of cracking in a large sample is consistent with this
suggestion (Waselkov 1982; Kent 1992).
Although other species of shellfish are present within the
shellfish assemblage as well, they are relatively few in number
compared to oyster. Additional shellfish species that occur in
frequencies worth mentioning include scallop (Argopecten),
various conchs and whelks (Melongena, Busycon, Strombus,
and Fasciolaria), quahog (Mercenaria), moon shells
(Naticidae), and marsh clam (Rangia). These species, like
oysters, inhabit or occasionally inhabit shallow estuarine
water. Of these other shellfish, only scallop occurs in high
enough frequencies to make up more than 10 percent of any
given sample taken at the site (11 percent of total).
One additional marine invertebrate is represented at the
site. Nearly 300 burned blue crab (Callinectes sapidus) claw
fragments were recovered from the midden mound (Blocks 1
and 4) and in the Excavation Block 3 midden. Of the 278 claw
fragments, 211 were recovered from the midden mound "cap"
deposits and 46 were recovered from either Stratum 4 in the
mound or in Block 4. The remaining 21 claw fragments are
from the Block 3 shell midden. The fact that each claw
specimen is burned may indicate that they were roasted "in the
shell" or that only burned or calcined claw fragments are
preserved. The large number of claws found in the mound is
of also interest, especially if crab claws are considered to be a
food resource reserved for special people or special occasions,
but it should also be noted that they are not usually as
numerous on other Choctawhatchee Fort Walton sites.


Vertebrate Fauna

A large assemblage of vertebrate faunal remains was
recovered from various shell midden and feature contexts.
Although no quantification will be presented here, a cursory
analysis of several samples indicates that a rather homogeneous
assemblage is present in various contexts across the site.
These samples include materials recovered in 1/4 inch screens,
flotation samples, and 1/16th inch water-screened bulk
samples. The vertebrate faunal remains assemblage clearly is
dominated by estuarine fish species. White-tailed deer,
raccoon, opossum, rodent, turkey, cormorant, alligator,
freshwater turtle, freshwater fish, and snake remains are also
present. Deer and freshwater turtle bone is quite common on
the site, especially the midden mound, but other nonestuarine
fish bone is only sparsely represented. The only exception to
the homogeneity of the samples examined is, of course, the
extremely dense deposits of fish bone recovered from the upper
30 to 40 cm of the midden mound.
Tables 3, 4, and 5 (Appendix A) present a summary of
flotation samples analyses for the Features 1, 2, and 3,
respectively. Table 8 (Appendix A), presented below, lists the
vertebrate species identified from 8WL38. A more detailed
analysis would undoubtedly identify more species, but those
listed in Table 8 are the most numerous and most readily
A few characteristics of the assemblage should be noted
here. The majority of the saltwater fish present in the
assemblage are species that are habitually or occasionally
present in and around oyster beds and sandy intertidal zones.
A wide range in the size of various fish is present in the
samples, suggesting that a major means of taking fish was by
net or weir. The freshwater fish identified (gar and bowfin),
are found in brackish water. Only two pieces of deer longbone
and one piece of bird longbone show evidence of butchering or
cutting for some other purpose.

Botanical Remains

Botanical remains were recovered from both features and
shell midden contexts. As with the vertebrate faunal remains,
a formal quantification of botanical remains is not attempted
here. A cursory examination of 13 flotation and radiocarbon
samples resulted in the identification of several plant foods and
wood fuel sources (Appendix A, Table 9). Two domesticated
plant foods are identified: maize and squash or bottle gourd.
Fragments of carbonized maize cob, kernel, and numerous
cupules were recovered in seven of 10 flotation samples taken
from the midden mound and four other midden deposits. A
single burned seed fragment from Feature 2 in the midden
mound was identified as squash or bottle gourd. Carbonized
grape seeds, persimmon seeds, hickory nutshell, and sumac
seeds round out the list of plant food remains. Wood charcoal

identified to genus includes oak, hickory, pine, maple, cedar
and grapevine.
The plant remains identified are indicative of both
hardwood hammock and forest edge exploitation for plant food
and fuel. The presence of maize and squash at 8WL38, while
not unexpected, does raise the issue of whether or not these
cultigens were being grown around the site in the poor, sandy
coastal soils. It is an issue that cannot be addressed at this
time, but it could be that coastal Fort Walton populations were
acquiring these plant foods from elsewhere through trade
(Hann 1988a: 17).

Choctawhatchee Bay Area Fort Walton Culture and 8WL38

During the Fort Walton period (ca. A.D. 1000 A.D.
1600), permanent village settlements, mound sites, and other
ceremonial sites found along Florida's northwest Gulf Coast,
in the Apalachicola Valley, in the Tallahassee Red Hills, and
in the Marrianna Lowlands attest to a well-established and
complex Mississippian variant culture (Milanich and Fairbanks
1980:193; Brose 1984; Mikell 1992). Data collected from
8WL38 has provided some of the best documentation of Fort
Walton settlement on the Choctawhatchee Bay between A.D.
1200 and the Protohistoric period (A.D. 1500-1600). At
8WL38, Fort Walton villagers settled in a coastal hardwood
hammock where they collected from and fished the
Choctawhatchee Bay and the numerous bayous around Four
Mile Point, hunted the hammocks and pine flats, and may have
grown maize and squash. 8WL38 has provided information
concerning both the everyday lives of the villagers as well as
ceremonial aspects of the Fort Walton period.
The data and materials that we recovered from 8WL38
are important for several reasons. To begin with, they
compare favorably with data obtained from other nearby Fort
Walton sites. Although not reporting any European artifacts,
Thomas (1989) described a Fort Walton village site (8WL99),
located little more than a kilometer southeast of 8WL38, which
yielded dates of ca. A.D. 1250-1600 and a smaller, but
remarkably similar ceramic assemblage as that of 8WL38. As
mentioned earlier, C.B. Moore (1918) reported European
artifacts from burials at a cemetery on Four Mile Peninsula
(8WL50). Nidy (1974) also reported many other Fort Walton
sites on Four Mile Peninsula that probably related to 8WL38
in some way.
The European artifacts and the Protohistoric period dates
from 8WL38 are, of course, important because they
demonstrate that the Fort Walton culture was in place and
thriving on the Choctawhatchee during the early days of
Spanish exploration and that somehow, artifacts from these
explorations made their way to the site. Although Spanish
artifacts have been recovered at other Fort Walton cemeteries
in the Choctawhatchee Bay area (summarized in Scarry
1990b:95), there are no other reports in any available

literature, no matter how incomplete or brief, that place early
European artifacts in a midden or habitation site context in the
Bay area. Even though it is likely that other protohistoric
Choctawhatchee Fort Walton habitation sites contain, or at one
time contained Spanish artifacts, they would likely be few in
The association of European metal artifacts and aboriginal
materials in a midden mound setting dated to between A.D.
1450-1550 leaves three obviously important questions to be
addressed. First, who was the original source of the metal
artifacts; second, how did they get to 8WL38; and third, why
are they at the site? The original source of the artifacts cannot
be demonstrated because they are not specific temporally
diagnostic artifacts. Historic documentation and the
radiocarbon dates obtained from 8WL38, however, indicate
that it is quite likely that the artifacts came to the site through
indirect contact (trade) with the Alonso de Pineda (1519),
Panfilo de Narvaez (1528), or Hernando de Soto (1539-1540)
expeditions in northwest Florida (Harrisse 1892; Hann 1988a,
1988b). Even though it is possible that either Pineda and
Narvaez, or Maldonado (de Soto's lieutenant) or their men
were on the Choctawhatchee, we only know that they were
somewhere in the area. It is unlikely that the Spanish were
ever at 8WL38 or on Four Mile Peninsula.
Why these artifacts are at 8WL38 requires a more in-
depth discussion. Our excavations so far indicate that Spanish
artifacts at the site occur only in the midden mound. There is
no doubt that the mound is a unique deposit on the site and is
the remains of activities that were distinct from routine
domestic activities which resulted in the "usual" shell middens
found elsewhere on the site. The midden mound, especially its
upper portion, is unique for the following reasons: it is an
intentionally "constructed" and conspicuous feature, it is
located in the central portion of the site, it is among the largest
Fort Walton midden mounds or midden ridges documented in
the Choctawhatchee Bay area (cf. Thomas 1989), its upper
portion contains a wide range of artifact types, including large
deposits of pottery, "exotic" items, items of personal
adornment, items potentially used in rituals, a potential
ceremonial cache or medicine bundle, and it contains
subsistence remains that may be indicative of ritual feasting.
Feature 1 is a "midden" mound in every sense of the
word. With the exception of a possible shell platform that was
laid down on top of an existing midden deposit prior to the
final periods of use of the mound (Figure 3, Stratum 5), each
of the stratified or distinct layered deposits is midden-like in
nature. What makes this feature distinct from other shell
middens on the site is that the artifacts it contains are an
unusual mix of domestic refuse suggestive of ritual feasting
and other items associated with ceremony (Appendix A, Table
10). Although no architectural features, such as postmolds or
wall trenches were documented in Excavation Block 1, the
presence of a fire pit near the mound summit (Feature 2)
indicates that the mound is more than simply a refuse heap.


The central location of the midden mound within the site
layout may also be of significance. Although excavations at
8WL38 have not been of the scale required to firmly
distinguish discrete activity areas within the site in terms of
specific use of space, enough data is available to generate some
ideas about site structure. Following Oetelaar's (1993:662-
687) application of Portnoy's (1981) general model of
settlement space and function to a Mississippian settlement in
southern Illinois, a few speculations can be advanced. The
model subdivides small settlements into four major activity
areas. They include: 1) a communal front area reserved for
community, civic, or ceremonial activities, 2) a family front
area for domestic tasks and the entertainment of guests, 3) a
family back area used for messy and space-consuming domestic
household activities, and 4) a communal back area used for
messy and space-consuming community activities (Oetelaar
1993). Oetelaar modified Portnoy's model to reflect spatial
patterning that radiates outward from the center of a site, the
communal front area.
Activities in each of these areas would produce differing
types and quantities of artifacts (1993:Figs. 1 and 8). Oetelaar
(1993:665) states that "the communal front area, as defined by
Portnoy (1981), consists of a central open space, free of
physical obstructions, which is easily accessible to all members
of the resident households. This activity region serves as a
space for feasts, ceremonies, assemblies, intra- or intervillage
competitions, and various economic activities." Communal
front areas are usually kept fairly clean and free of large
amounts of refuse.
Figure 1 illustrates that there are areas lacking shell
midden deposits located around the midden mound that could
have been communal front areas, given Portnoy and Oetelaar's
definition. Although there have been no excavations to
investigate the potential for a communal front area, midden-
free areas are likely candidates. Given the central location of
the midden mound on the site, the types and quantities of
artifacts deposited in it, and the potential for a communal front
area located adjacent to the mound, it is entirely possible that
the mound served as a central place for communal gatherings.
There are numerous descriptions of communal or ceremonial
and ritual feasts among historic aboriginal Southeastern groups
(cf. Adair 1973; Bartram 1955; Hudson 1976; Swanton 1911,
1922, 1928, 1946), and mounding of materials was associated
with the busk and Green Corn ceremonialism.
Another intriguing bit of indirect evidence suggests a
possible ceremonial aspect for at least some of the pottery
deposited in the midden mound. Even though 8WL38 does not
appear to be a mortuary site, there are a few hints that the
partial vessels and large amount of sherds, many of which
articulate, may be indicative of ritual breaking of pottery.
William Lazarus (n.d.) reported what he described as virtually
solid layers of sherds in the upper layers of his excavations at
the following three Fort Walton cemeteries: the Johnson site
(Alaqua Bayou, 8WL30), Point Washington (8WL33), and

Hogtown Bayou (8WL50). In describing these pottery sherd
concentrations, Lazarus (n.d.) writes: "The first six inches
produced a compact mass of sherds. Large fragments were
stacked as many as six deep and overlapping to form a
complete sherd cap on the mound. Many of these sherds were
found to fit one another. There is also evidence for the ritual
killing of pots."
A potential analogy for explaining the "sherd caps" and
the nature of the pottery in the midden mound is the
observation of rituals that involve the breaking of pottery
during the busk or Green Corn (Adair 1973; Swanton 1922,
1928, 1946). In fact, Adair (1973) and Swanton (1922, 1928)
observed that ceremonies commonly took place on "elevated
ground" and that one common ritual was the breaking of old
pottery or ceremonial pots at the close of the busk or ritual
Domestic aspects of the village are evident in the form of
household refuse piles and activity areas represented by the
shell middens that form an arc around the center of the site and
the creek along the Bay shore (Figure 1). Following Oetelaar
(1993), these deposits represent "family front, family back,
and communal back areas". At this point in the investigation
of 8WL38, I would modify Oetelaar's settlement model further
to include several household or family front and back areas
scattered around the communal front area, rather than actually
encircling it. Communal back areas would probably shift
seasonally, depending on prevailing winds (Portnoy 1981).
The radiocarbon dates obtained indicate that significant
depositional episodes occurred between about A.D. 1200 and
A.D. 1550. What is not clear is whether the site was
continuously occupied during that 350 year span. The
overlapping range for the radiocarbon dates (see Appendix A,
Table 2) suggests that 8WL38 was continuously occupied. I
am not suggesting that the site's resident population was either
large or permanent. It is possible that a fairly small, resident
population permanently occupied the site and that the number
of people on site intermittently increased during certain seasons
or communal gatherings.

Discussion and Summary

Who were the people that created what is now called
8WL38? As for the prehistoric inhabitants, we can only say
that they were people belonging to a variant of Mississippian
culture known as the Fort Walton archaeological culture and
we can describe their material remains. As for the
protohistoric inhabitants, unfortunately, we can not do much
better since the Spanish were not in the Choctawhatchee Bay
area enough to record much about them (see Scarry 1990b).
Based on Hann's (1988c) interpretation of Spanish documents,
Scarry (1990b:99) suggests that the ancestors of the historic
Pansacola and Chine may have been the sixteenth-century

inhabitants of the Choctawhatchee. The Chine of the late
1600s had extensive knowledge of the Gulf Coast west of the
Apalachee Province and they served as pilots for voyages from
Apalachee to Pensacola (Hann 1988c:85). Unfortunately, this
bit of evidence is the best there is for drawing a link with
historically documented people.
Spanish accounts (Cabeza de Vaca's account of the
Narvaez expedition and the account of Tristan de Luna)
suggest that by the sixteenth century at least, the inhabitants of
the Gulf Coast were not farmers like the Apalachee to the east.
In fact, Narvaez traded beads, bells, and maize for fish and
water after leaving the Apalachee Province and sailing west
along the coast (Hann 1988a:17). Maize may have been a
trade commodity on the coast for some time prior to Narvaez's
account, and perhaps, the maize recovered from 8WL38 was a
trade item. Luna, however, described sites with small maize
fields around Pensacola Bay (Hann 1988c) and the same could
have been true of 8WL38. To underscore the importance of
fish to coastal populations, Vaca reported finding dried mullet
and roe at villages along the coast (Hann 1988c:61).
We know that the inhabitants of 8WL38 and
Choctawhatchee area in general, had a relatively simple society
compared to other Mississippian cultures. Even though what
little we know about their social and political organization
suggests that they did not develop archaeologically visible,
complex chiefdoms such as those witnessed at sites such as
Lake Jackson and other mound centers in the Tallahassee Hills
and Apalachicola Valley (Jones 1982, 1994; Scarry 1984,
1990a), mounds such as the one at Fort Walton suggest that
there probably were chiefdoms present on the Choctawhatchee.
Scarry (1990b:100) suggests such a possibility based on
settlement patterns and the distribution of known mound and
cemetery sites. There is also the potential of some status
differentiation in burial treatment at the known mounds and
cemeteries excavated by Moore (1901, 1918), at least for the
Protohistoric period. Although the burials in each site were
probably not contemporaneous, only a few of the burials
contain European (Spanish) artifacts, a pattern repeated at three
of the four Choctawhatchee Bay cemeteries, including the
nearby Hogtown Bayou cemetery (Moore 1918:537-538). The
problem is that Moore does not tell us specifically what got
buried with who. All we can say is that there are no
spectacular mound sites or elaborate burials of chiefs on the
No matter how one chooses to view Choctawhatchee Fort
Walton social and political organization, there are a few things
of which we can be reasonably sure. The people had leaders, a
functioning social order, economic, social, and political ties
with other societies, and the customs and ritual that reinforced
their way of life. We see a glimpse of that life at 8WL38 in
the form of domestic refuse, remains of ritual feasting and
ceremony, and items of personal adornment found at the site.



None of this work would have been possible without the interest, cooperation, and toil of the property owners, the students,
and the volunteers who participated. Thanks go out to the property owners: Don and Betty Hefner, David Forstrom, Dr. and Mrs.
Roger Beldt and Tom Patton of Sandestin Resorts, Inc.; to my assistant, Jennifer Lazowski; Okalossa-Walton Community College;
and to the others who helped get the work done and into print. The students participating in the archaeology field school were:
Joseph Hank, Amy Timms, Loren Boyer, Anna Ogle, Trisha Dall, Shawna Boland, Karen White, Michelle Buckellew, Andy
Burke, Jessica Duncan, Danielle Babski, Kim Bultez, James Craig, Bob Nagel, Dave Morris and Bill Reger. They all worked hard
and had fun, I hope. I also hope to be able to continue the work at 8WL38 before the realities of the twentieth century (bulldozers
and backhoes) hit hard. If this happens, there will be more to tell about days gone by on the Choctawhatchee.

Appendix A
Table 1. Summary of Excavation Units (TUs) and Excavation Blocks

TO 1/2
Unit Size: 1 by 2 m.
Location: southeastern slope of Feature 1.
Stratigraphy (Figure 2): 3-5 cm of hums; a 19-38 ca thick shell ridden (Stratum 2); a 9-22 ca thick black sand
Ridden (Strata 3); two layers of culturally sterile gray to brownish-yellow sand lie below Stratum 3
(Stratum 4/Stratu 5).
Notes: Stratum 2 contains primarily whole oyster (87%), dense vertebrate faunal remains, and a moderate amount of
pottery in organically rich black sand; Stratu 3 contains scattered oyster shell and small amounts of
vertebrate faunal remains and pottery; with the exception of Feature 7, cultural deposits truncate at base
of Stratum 3.
Other Features: an ash lens (Feature 7) was encountered at the base of Strata 3.
Radiocarbon Dates: A.D. 1442-1527, charcoal from Feature 7 (Appendix A, Table 2).

Excavation Block 1 (TUs 3-6/33-36)
Block Size: 2 by 4 m.
Location: Feature 1 (Hidden Hound).
Stratigraphy (Figure 3): 3-5 ca of hums; an 8-12 ca layer of crushed and whole oyster shell (Stratum 2); 12-16 ca
thick layer of whole oyster shell (Stratum 3); an 8-23 ca thick layer of oyster and scallop (Stratum 4);
a 14-36 cm thick layer of oyster shell (Stratm 5); an 11-17 cm thick sandy shell ridden (Stratum 6); two
layers of culturally sterile gray to brownish-yellow sand that lie below Stratu 6 (Stratum 7/Stratun 8).
These deposits extend to 83 ca below the sound summit as it exists today and the mound stands 75 to 85 cm
above the surrounding ground surface.
Notes: Stratus 2 contains primarily whole oyster (62t), very dense deposits of vertebrate faunal remains, pottery,
charred plant remains, and several other types of artifacts in organically rich black sand, much of the
crushed shell is burned; Stratum 3 contains whole oyster shell (91%), some of which is burned, very dense
deposits of vertebrate faunal remains, a large amount of pottery and charred plant remains, and a few other
types of artifacts organically rich black sand; Stratum 4 contains a moderate amount of scattered oyster
(79%) and scallop (16) shell, scattered vertebrate faunal remains, and a large amount of pottery;
Stratum 5 contains whole oyster (94%) with very few artifacts and very little organic material, including
soil: Stratum 6 contains whole oyster (82%) and other shellfish remains, some of which is burned, a moderate
amount of vertebrate faunal remains and pottery, and a few other artifacts, but like Stratum 5, cultural
materials in Stratum 6 do not approach the density of materials in Stratum 2, 3, and 4. Cultural deposits
abruptly disappear at base of Stratm 6.
Other Features: a intrusive fire pit or hearth (Feature 2) was encountered in Unit 35: Stratum 2 and Stratum 3.
Radiocarbon Dates: A.D. 1468-1552, charcoal frost Feature 2; A.D. 1278-1367, charcoal from base of Feature 1 (Hidden
Mound), Stratm 6 (Appendix A, Table 2).

TO 7/8
Unit Size: 1 by 2 a.
Location: an eroding shell Ridden along the southwestern margin of the site.
Stratigraphy: 2-5 cm of humus; a 20-26 cm thick shell Ridden (Stratum 2) containing mixed shellfish remains
dominated by oyster (881), sparse vertebrate faunal remains and a moderate amount of pottery; two layers
of culturally sterile gray to brownish-yellow sand lie below Stratum 2 (Stratus 3/Stratum 4).
Notes: the upper 10 c of this ridden appears to have been disturbed; a few pieces of shell and pottery were
recovered in Stratum 3, below the actual Ridden deposit.


Table 1 (cont.)

TU 9/10
Unit Size: 1 by 2 m.
Location: an eroding shell midden located along the shoreline toward the central portion of the site.
Stratigraphy: 5 cm humus; a 28 to 30 cn thick black sandy shell midden containing a dense deposit of mixed shellfish
remains dominated by oyster shell (78%), sparse vertebrate faunal remains and a moderate amount of broken
pottery (Stratum 2). Eighty-one percent of the oyster shell is whole and much of the
fragmented shell is burned. The shell midden in these units overlies a layer of light gray sand which
extends to 52 cm below the surface (Stratum 3), where it grades into a light brownish-yellow sand
(Stratum 4).
Notes: TU 9/10 was placed in a series of overlapping midden deposits measuring over 70 m along the shoreline and
extending up to 30 m in from the bay front; cultural materials were confined to the shell midden.

Excavation Block 2 (TUs 11-16)
Block Size: 2 by 3 m.
Location: a shell midden located on a notable bend in the creek near where it empties into Choctawhatchee Bay
Stratigraphy (Figure 4): 3-4 cm of humus; an 18 to 26 ca thick shell midden containing primarily whole oyster (87%),
along with scallop, conch, and burned and crushed shell and moderate deposits of vertebrate faunal remains,
and a fairly large amount of broken pottery (Stratum 2); a 5 to 20 cm thick dark grayish-brown sand midden
deposit (Stratum 3) below the shell midden containing scattered oyster shell, a small amount of vertebrate
faunal material, and a few pieces of broken pottery. The midden layers overlie two layers of sterile sand
(Stratum 4 and Stratum 5).
Notes: This midden slopes eastward toward the creek where it is visibly eroding along the creek bank.

Excavation Block 3 (TUs 17-22)
Block Size: 2 by 3 m.
Location: an extensive shell hidden located north of the creek that divides the site.
Stratigraphy (Figure 5): 3-4 cm of humus; a 21 to 36 cm thick sandy, dark grayish-brown shell midden deposits
(Stratum 2a and 2b) containing primarily whole oyster (47%) and scallop (25%), as well as conch and burned
and crushed shell (22%) associated with dense deposits of vertebrate faunal remains, and a large amount of
broken pottery, including a few large sherds. Within Stratum 2, a circular, lens-like and very dense shell
midden deposit (Stratum 2b) was encountered; the middens overlie two layers of sterile sand.
Notes: This midden area is the largest undisturbed shell midden remaining on the site north of the creek, measuring
approximately 60 by 20 meters: the shell midden has a slightly raised, ridge-like central feature that is
up to 25 cm higher than the surrounding midden area; Stratum 2b may be the remains of a refuse pit, but
other than the density of the shell it contained, it could not be discerned from the surrounding hidden.
Radiocarbon Dates: A.D. 1303-1386, charcoal taken from Stratum 2b (Appendix A, Table 2).

Excavation Block 4 (TUs 23-28)
Block Size: 2 by 3 a.
Location: the eastern margin of the midden sound (Feature 1), six meters east of Block 1.
Stratigraphy (Figures 6 and 7): 3-4 cm of humus; Stratum 2 is 8 to 12 cm thick in the eastern part of Block 4 and
increases in thickness of 10 to 21 ca in the western end of the block; this midden is an organically rich,
black sandy deposit containing a moderate amount of scattered oyster (85%) and scallop shell (8%),
vertebrate faunal remains, a large amount of broken pottery, and a few other artifacts, including an iron
band fragment and a hand-wrought nail. Stratum 3 is a uniform, 18 to 20 ci thick layer of black, sandy shell
midden rich in charred plant and vertebrate faunal remains; shell in Stratum 3 is primarily whole oyster
(79%) with some conch and burned oyster shell fragments and contains a moderate amount of broken pottery
along with a few other types of artifacts. Two layers of gray and yellowish-brown sand lay beneath the
midden deposits and features in Block 4.


Table 1 (cont.)

Excavation Block 4 continued
Notes: The upper layer of midden (Stratum 2) appears to be mound slope matrix similar to Stratum 4 in Block 1. The
lower midden (Stratum 3) appears to be part of the same midden identified in Block 1 as Stratum 6.
Other Features: Figures 6 and 7 illustrate that cultural features were encountered at the base of the
midden; these features consist of a fire pit/trash pit (Feature 3) and four postmolds. The profile of one
of the postmolds (Feature 6) is illustrated in Figure 6 and a profile of the fire pit is illustrated in
Figure 7 (see detailed description below). With the exception of a few scattered pieces of shell, cultural
materials abruptly disappear at the base of Stratum 3 or are confined to features.
Radiocarbon Dates: A.D. 1189-1330, charcoal from Feature 3 (Appendix A, Table 2); this date overlaps the earliest
date from the midden mound (A.D. 1278-1367).

Excavation Block 5 (TUs 29-32)
Block Size: 2 .
Location: a slightly raised midden area located immediately southwest of the midden mound (Feature 1).
Stratigraphy: 3 cm of humus; an 18 to 24 cm thick layer of black sandy shell midden (Stratum 2), containing a dense
deposit of oyster shell (78%), sparse vertebrate faunal remains, and a moderate amount of broken pottery;
a majority of the oyster shell in Stratum 2 is whole and much of the fragmented shell is burned.
An 8 to 10 cm thick black sand midden (Stratum 3) located below Stratum 2 contained a small amount pottery
compared the shell midden above it; the middens overlie a layer of light gray sand extending to 48 cm below
the surface (Stratum 4), where it grades into a light brownish-yellow sand (Stratum 5).
Notes: The majority of cultural materials are confined to (Stratum 2); the midden is only slightly elevated to a
height of 13 cm above the surrounding ground surface. Although this slightly raised midden is probably the
surface feature that caused Nidy to describe the midden sound as being a crescent-shaped feature (1974:5),
there is no evidence to suggest that it is part of the mound structure. In fact, there is a distinct break
in midden matrix between the two features, with no firm evidence of disturbance suggesting soil/shell
removal or other ground surface alterations.

Excavation Block 6 (TUs 37-40)
Block Size: 2 a2.
Location: a large, slightly raised shell midden located about 50 m southwest of the midden mound (Feature 1).
Stratigraphy: 3-5 cm of humus and grass; Stratum 2 appears to be the disturbed upper 8-10 cm portion of the much
more dense midden labelled as Stratum 3; Stratum 2 consists of fragmented shell and small pottery sherds
while Stratum 3 is a 10 to 16 cm thick, very dense deposit containing primarily whole oyster (67%) and
scallop (17%), along with conch and burned and crushed shell (15%), sparse deposits of vertebrate faunal
remains, and a moderate amount of broken pottery. Below Stratum 3, another intact, but less dense deposit
was found (Stratum 4) that is 20 to 22 ca thick, producing relatively few artifacts from an oyster (79%)
midden. Although some shell was recovered from a thin layer of dark gray to brown sand situated below
Stratum 4, only a single piece of pottery accompanied it. The midden deposits overlie two layers of sterile
sand (Stratum 5/Stratum 6).
Notes: This midden, like all of the cultural deposits located in the partially cleared lot (see Figure 1), has been
disturbed at the ground surface. As excavations in Block 6 demonstrated, however, the raised midden area
does contain undisturbed deposits below 8 to 10 cm.


Table 1 (cont.)

TU 41/42
Unit Size: 1 by 2 m.
Location: east of the midden mound (see Figure 1).
Stratigraphy: 3-5 ca of humus; a 19 ca thick layer of very compact, coarse-grained dark grayish-brown sand
containing numerous small sherds (Stratum 2); a 21 ca thick light gray layer of sand with a few sherds in
its upper 5 to 8 cm (Stratum 3); a layer of yellowish-brown, culturally sterile sand (Stratum 4).
Notes: This unit was excavated in an area with no shell midden deposits where shovel testing recovered an unusually
large number of pottery sherds. Although no features were encountered in this unit, the compacted coarse
sand containing many small sherds in Stratum 2 suggests that it was an area that was subjected to a great
deal of foot traffic and trampling.

TU 43/44
Unit Size: 1 by 2 m.
Location: the southern-most shell midden located on 8WL38 (see Figure 1).
Stratigraphy: 3-5 cm of humus; a single 22-26 cm thick shell midden consisting of whole (77%) and fragmented or
burned oyster shell that contains dense deposits of vertebrate faunal remains, and numerous large pottery
sherds. The shell midden (stratum 2) lies on top of two sterile layers of sand, a gray to light gray layer
and a brownish-yellow layer.
Notes: This shell midden is actively eroding and only a small area is left intact. The midden contains an unusually
high percentage of Pensacola Series pottery relative to other domestic refuse aiddens on the site.
Radiocarbon Dates: A.D. 1378-1466, charcoal from midden (Appendix A, Table 2); this radiocarbon date is consistent
with the later dates from the hidden mound.

Table 2. 8WL38 Radiocarbon Sample Analysis Results

PrnuP~n 1 O (PnntPvt

Excavation Block 1
TU35/L 3: Feature 2
Midden Mound

TUl/L3-4: Feature 7
below midden

TU44/L 3: base of midden

Excavation Block 3
TU21/L 4: base of midden

Excavation Block 1
TU34/L 8: base of Fea. 1
(Midden Mound)

Excavation Block 4
TU24/L 3: Feature 3
below midden








Calibrated Date*
(Ranae 1 siama) Sample#

A.D. 1496 63962
(A.D.1468 to A.D.1552)

A.D. 1478 64280
(A.D.1442 to A.D.1527)

A.D. 1418
(A.D.1378 to A.D.1466)

A.D. 1340
(A.D.1303 to A.D.1386)



A.D. 1283 64277
(A.D.1278 to A.D.1367)

A.D. 1238 64276
(A.D.1189 to A.D.1330)

*Calibration by CALIB (Stuiver and Becker 1986)

Provenigne-T rnnt-vt

. ........ ---- -


Table 3. Summary of TU 36, Stratum 3 Flotation Sample (100 L)

Invertebrate Fauna I wt. Vertebrate Fauna I wt. Botanical # vt.
virginia oyster 234 1973g fish (Osteichthyes) 10745 644g wood charcoal 72g
various whelk/conch 6 821g freshwater turtle 121 189g hickory nut shell 9 24g
bay scallop 56 211g white-tailed deer 5 179g maize cob/cupule/kernel 58 8g
quahog 4 178g other maaal 7 34g persimmon seed 2 2g
turbans/periwinkles 113 99g bird 1 7g grape seed 8 <2g
barnacle 82 94g sumac seed 3 <2g
other/unident. shellfish 28 23g maypop seed 1 blue crab (claw frags.) 11 18g
Totals 534 3417g 10879 1053g 81 110g

*count/weight includes whole elements and fragments

Table 4. Summary of Feature 2 Flotation Sample (37L)

Pottery by type/variety I Invertebrate/Vertebrate Fauna I wt. Botanical I vt.
Lake Jackson Plain white-tailed deer 11 216q wood charcoal 87g
var. unspecified 3 freshwater turtle 33 123g hickory nut shell 6 14g
var. Apalachicola 2 fish (Osteichthyes) 621 lllg maize cob/cupules 112 llg
var. OCattaboochee 1 fish (Chondrichthyes) 4 4g grape seed 11 <2q
var. Ingram 1 bird 1 39 persiuon seed 1 <1g
var. Jefferson 3 virginia oyster 87 211g squash seed 1 Jefferson Ware 1
Bell Plain totals 757 658g 131 115g
var. unspecified 2
Mississippi Plain
var. unspecified 4
var. Pine Log 1
Residual plain 30
Fort Walton Incised
var. unspecified 2
var. Fort Walton 3
Lake Jackson Incised
var. Blounstown 2
var. Walter George 1
Point Washington Incised
var. unspecified 4
var. Chambliss 1
var. Rogtown Bayou 1
D'Olive Incised
var. D'Olive 1
Pensacola Incised
var. unspecified 1
var. Bear Point 1
total 65

*count/weight includes whole elements and fragments

Table 5. Summary of Feature 3 Flotation Sample (35 L)

Pottery by type/variety i Invertebrate/Vertebrate Fauna I wt. Botanical I Vt.
Lake Jackson Plain freshwater turtle 18 1769 wood charcoal 64g
var. unspecified 1 fish (Osteichthyes) 53 105g hickory nut shell 4 16g
Bell Plain white-tailed deer 5 102g maize cupules 31 <2g
var. unspecified 1 alligator (dermal scute) 1 8g grape seed 4 <2g
Residual plain 10 virginia oyster 67 411g
Point Washington Incised bay scallop 34 29g
var. unspecified 2 unidentified shell fragments 87 121g
Carrabelle Punctated
var. unspecified 1
totals 15 265 952g 39 83g
*Count/weight includes whole elements and fragments


Table 6. Ceramics Recovered by Excavation Block, Test Unit, Arbitrary Level, and Feature*

Type/Classification T01/2
Lavels 1 2 3 4
Plain Wares
Lake Jackson Plain
var. unspecified 3 4 4 2
var. Apalachicola 1 -
var. Chattahoochee' I 1
var. Ingram -
var. Tallahassee -
var. Jefferson 1 -
Jefferson Ware -
Residual Plain
sand tempered 7 6 14 5
grit tempered 42 21 49 19
Incised Wares
Cool Branch Incised
var. unspecified -
var. Cool Branch -
Fort Walton Incised
var. unspecified 1 1 2
var. Fort Walton -
var. Cayson -
var. Choctawhatchee -
var. Sneads -
Lake Jackson Incised
var. unspecified 1 -
var. Blounstown 5 2
var. Walter George -
Lamar Bold Incised
Marsh Island Incised
var. Marsh Island -
var. Columbia -- 1
var. Carrabelle -
Ocaulgee Fields Incised
var. Ocaulgee -
var. Aucilla -
Point Washington Incised
var. unspecified 1
var. Point Washington -
var. Chambliss -
var. Hogtown Bayou 1 -
var. Griffith -
unidentified incised 2 -
Miscellaneous Wares
Carrabelle Punctated
var. Meginnis 1-
Fort Walton Beaded
var. Walton -
Lamar Complicated Stamped
var. early -
Leon Check Stamped
var. Leon -
unspec. fabric impressed -
ceramic disc -

1 2 3 4 5 6 7 8 9 F2





31 14 25 24 11 4 2 4 3 12
72 41 69 88 22 14 10 8 1 18

T17/B Tu9/10
1 2 3 4-5 1 2 3 4

1 1 -

2 1 -

6 15 10 5
16 28 38 11

- 1 4
1 -
3 2 -
1 -

21 10 6
31 26 18

1Bloc 2
plrrmi-i Total.~l

TU11-16 Total
1 2 3 4-5

- 1 1

- 1 1

2 -
1 -

29 13 5
66 28 17

1 2 -

- 1 2 --- 2
7 4 1 5 2 2 -
1 -1 -

- 1 1 1-- 1
1 -
2 1 1 2 1 1

- -
1 .

S 1

2 1 1 -

S 1 -

2 1 1----

- 1 2 1 ----1-

-1 -----

- --11

- 1--


- 1


- - 4

1--- 1

Plain Wares
Bell Plain
var. unspecified 2 9 10 1 5 3 -
var. Hale 2
Mississippi Plain
var. unspecified 1 4 5 3 12 4 7 11 4
var. Pine Log 1 1 -
var. Warrior ---- 2 2 -
Incised Wares
D'Olive Incised
var. unspecified 1 -
var. D'Olive 2 -
var. Arnica 2
var. Dominic 1 2 -
Moundville Incised
var. Moundville --- 1 -
var. Bottle Creek -
var. Douglas
Mound Place Incised
var. Waltons Camp 1 1 -
Pensacola Incised
var. unspecified 5 1 5 3 1
var. Pensacola -- 1 2 7-
var. Bear Point -
var. Perdido Bay --- -
Carthage Incised 2 1 1
unidentified incised -
ceramic disc -
TOTAL 58 49 87 36 6 188 107 169 206 48
*Ceramics recovered from features are not presented by level
*Count includes partial or whole vassel(s) where each specimen 1

1 1 2

- -1
3 8 4 4
- -- 1

1 3 1

32- 65

32 36 3 23 65

3 5 3 -
- 1 1 -

- 1 1-

6 2 1 -
2 -

7 6 3 -

--1 --1-

- -

- 1 -7

39 65 72

3 5
2 1
- 1

92 73

17 6 77
5 5 16

33 27 7 2 181
4 9
5 8 24

S 1 2

1 4 -- 7

4 9

LO 5 48
3 1 17
-1 7
-- 1

57 193 55 28 2125



Table 6 (cont.)

Plain Wares
Lake Jackson Plain
var. unspecified
var. Apalachicola
var. Chattahoochee
var. Ingram
var. Tallahassee
var. Jefferson
Jefferson Ware
Residual Plain
sand tempered
grit tempered
Incised Wares
Cool Branch Incised
var. Cool Branch
var. unspecified
Fort Walton Incised
var. unspecified
var. Fort Walton
var. Cayson
var. Choctawhatchee
var. Sneads
Lake Jackson Incised
var. unspecified
var. Blounstown
var. Walter George
Lamar Bold Incised
Marsh Island Incised
var. Marsh Island
var. Columbia
var. Carrabelle
Ocmulgee Fields Incised
var. Ocaulgee
var. Aucilla
Point Washington Incised
var. unspecified
var. Point Washington
var. Chambliss
var. Hogtown Bayou
var. Griffith
unidentified incised
Miscellaneous Wares
Carrabelle Punctated
var. unspecified
var. Meginnis
Fort Walton Beaded
var. Walton
Lamar Complicated Stamped
var. early
Leon Check Stamped
var. Leon
ceramic disc

Block 3
1 2 3 4

Block 4
1 2 3 4 F3 F4 F5 P6

Block 5 Block 6
TU29-32 T037-40
1 2 3 4 1 2 3 4

- 1 2

2 -
2 1 1

1 -

TU41/42 TU43/44
1 2 3 4 1 2 3 4

1 1 -
2 1 -

2 1--


- 1 1
- 1 -

1 2 1
1 -
1 -

28 19 7 3 73 52 28 5 1 4 20 13 2 24 32 9 3 7 23 3 1 13 9 16 7 402
52 6 22 8 92 84 68 22 9 1 2 2 19 41 35 9 41 82 21 11 22 42 15 3 33 28 31 12 871

1 -1 -- -

8 2 2 28

- 1 7
- 2
- 4

2 1 1 8

29 11 3
1 1 -
6 4 1
1 -
2 3

4 2
4 -

- 1- - - -

5 2-

- 1 -
-1 -
1 -

3 -
3 1-

1 2 1 -

1 1 1 -

5 1 1
- 1 -

2 -
- 1 -

- 3
3 1 -
2 -
1 -

- 1- 1 -- --- 2 -
31--- --------------------------------- -- 31-
- 1 2--------------------- ------1 -
3 1- -- --- -- -

1 -

19 14 3
1 -

1 -
- 1 -
1 -

9 3 -

1 -
- 1
--2 -

1 4 -

- 3 -

6 3 1
2 -
4 -
3 -

1 ~~- - -- -- --- 1-

- 5

- 1-- --- 2 1-

- ---------------------------

- 2 -
- - - - -1 -

-1-- 1

Plain Wares
Bell Plain
var. unspecified
var. Hale
Mississippi Plain
var. unspecified
var. Pine Log
var. Warrior
Incised Wares
D'Olive Incised
var. unspecified
var. D'Olive
var. Arnica
var. Dominic
var. Mary Ann
Moundville Incised
var. Moundville
var. Bottle Creek
var. Douglas
Mound Place Incised
var. Waltons Camp
Pensacola Incised
var. unspecified
var. Pensacola
var. Bear Point
var. Perdido Bay
var. Gasque
Carthage Incised
unidentified incised
ceramic disc

3 2 1 26 21 8 4 1 6 13 7 4 3 9 2 1 16 9 3 1 140
- 1 12 4 1 3 4 2 1 3- 3 4 1 39

23 19 3 38 36 11 19 13 25 18 2 15 18 5 6 4 22 9 4 3 293
1 1 9 3 1 2 3 4 24
4 1 9 7 4 3 5 7 6 3 7 14 11 81

- 1 -

- 1

3 -

1 -

5 4 2 13 11 2 3
1 4 5 -1

1 1 3

146 132 46 14 374 321 167 70

2 5 2 -
2 -

1 -

65 137 106 21

5 3-
1 -
1 -

1 1 -
S 1
118 185 53 19

- 2 -
S8 20

42 99 20 4

1 -

1 -

2 1 -

11 21 8 1
6 B 1-

3 1 1-

153 143 85 28

*Ceramics recovered from features are not presented by level
*Count includes partial or whole vessels) where each specimen = 1

- 2 -


Table 7. Proportional Summary of 8WL38 Pottery Types/Varieties

Type/Variety Count

Plain Wares
Lake Jackson Plain
var. unspecified 89
var. Apalachicola 24
var. Chattahoochee 36
var. Ingram 77
var. Tallahassee 12
var. Jefferson 25
Jefferson Ware 6
Residual Plain
sand tempered 710
grit tempered 1722
Incised Wares
Cool Branch Incised
var. unspecified 6
var. Cool Branch 4
Fort Walton Incised
var. unspecified 258
var. Fort Walton 25
var. Cayson 30
var. Choctawhatchee 18
var. Sneads 14
Lake Jackson Incised
var. unspecified 32
var. Blounstown 80
var. Walter George 4
Lamar Bold Incised 5
Marsh Island Incised
var. Marsh Island 6
var. Columbia 18
var. Carrabelle 5
Ocmulgee Fields Incised
var. Ocmulgee 2
var. Aucilla 2
Point Washington Incised
var. unspecified 185
var. Point Washington 27
var. Chambliss 21
var. Hogtown Bayou 5
var. Griffith 7
unidentified incised 10
Miscellaneous Wares
Carrabelle Punctated
var. unspecified 2
var. Meginnis 4
Fort Walton Beaded
var. Walton 12
Lamar Complicated Stamped
var. early 9
Leon Check Stamped
var. Leon 6
unspec. fabric impressed 1
Total 3499

% Total













% Decorated Type/Variety
Plain Wares
Bell Plain
var. unspecified
var. Hale
Mississippi Plain
var. unspecified
var. Pine Log
var. Warrior
Incised Wares
D'Olive Incised
var. unspecified
var. D'Olive
var. Arnica
var. Dominic
.5 Moundville Incised
.4 var. Moundville
var. Bottle Creel
23.9 var. Douglas
2.4 Mound Place Incisec
2.8 var. Waltons Camn
1.7 Pensacola Incised
1.3 var. unspecified
var. Pensacola
2.9 var. Bear Point
7.4 var. Perdido Bay
.4 var. Gasque
.5 Carthage Incised
unidentified incisw
.5 Total







Count % Total % Decorated
(n=4666) (n-1081)






2 10

Table 8. Identified Vertebrate Remains from 8WL38

Class Nammalia:

OdocoiQlus virginianus (deer)
Didelphis virginiana (opossum)
Procyon lotor (racoon)
Sciurus spp. (squirrel)
Rodentia (unspecified rodent)
Unspecified mammal

Class Aves:
Meleaaris gallopavo (turkey)
Phalacrocorax spp. (cormorant)
Unspecified bird

Class Reptilia:
Alligator miss. (alligator)
Pseudemys spp. (cooter/slider)
Terrapene carolina (box turtle)
Testudinidae (unspecified turtle)
Serpentes (unspecified snake)

Class Osteichthyes:

Lepisosteus spp. (gar)
Amia calva (bowfin)
Elops saurus ladyfishh)
Ariidae (saltwater catfishes)
Caranx spp. (jack)
Sciaenidae (drum fish)
Cynoscion spp. (seatrout)
Pogonius cromis (black drum)
Sciaenops ocellata (red drum)
Archosargus probatocephalus
Mugilidae (mullet)
Hugil cephalis (mullet)
Paralichthys spp. (flounder)
Diodontidae (burrfish/puffer)
Unidentified fish

Class Chondricthyes:
Raa spp. (skates)
Dasvatis americana (stingray)

_ ____ __ ____ ._ ____ ___ I__ _I

_ ____ __ ~~

_ ____


Table 9. Identified Floral Remainsfrom 8WL38

Plant Food Remains Wood Charcoal
Zea mays (maize) Ouercus spp. (oak)
Cucurbita spp. (squash seed) Carya spp. (hickory)
Vitus vitus (grape seed) Pinus spp. (pine)
Dryopsyros spp. (persimmon seed) Acer spp. (maple)
Passiflora spp. maypopp seed) Cedrus spp. (cedar)
Rhus spp. (sumac seed) Vitus vitus (grape vine)
Carya spp. (hickory nut shell)

Table 10. Artifacts* Recovered from the Upper 50 cm of Midden Mound (Feature 1, Block 1, Strata 1-4)

Partial Vessels
Feature 1: Lake Jackson Incised, variety Blounstown (1); Lake Jackson Plain, variety Ingral (2);
Leon Check Stamped, variety Leon (1); Moundville Incised, variety Moundville (1);
Point Washington Incised, variety unspecified (1)

Individual Sherds (see Appendix A, Table 1, Block 1, Levels 1-5 for varieties)
Feature 1: Bell Plain (21), Carrabelle Punctated (1), Carthage Incised (4), Cool Branch Incised (7),
D'Olive Incised (7), Fort Walton Incised (50), Jefferson Ware (1), Lake Jackson
Incised (22), Lake Jackson Plain (63), Lamar Bold Incised (1), Lamar Complicated
Stamped (1), Marsh Island Incised (6), Mound Place Incised (2), Mississippi Plain (45),
Ocmulgee Fields Incised (2), Pensacola Incised (31), Point Washington Incised (48),
Residual Plain (397), unidentified incised (1)

Feature 2 (fire pit): Bell Plain (2), D'Olive Incised (1), Fort Walton Incised (5), Jefferson
Ware (1), Lake Jackson Incised (3), Lake Jackson Plain (10), Mississippi Plain (5),
Pensacola Incised (2), Point Washington Incised (6), Residual Plain (30),

Ceramic Discs (4): one engraved with equal-armed or Christian cross motif

Ceramic Adorno (1): part of headdress or mask?

Lithic Artifacts: chert arrow point (1), bipointed micro drill/engraving tool (1), mano-like object (1),
possible Chunkey stone (1), celt spall (1)

Bone and Shell Artifacts: shell beads (4), shell pendants (4), shell cup (1), shell spatula (1), shell
awl/potter's tool (1), bone awl (1), gar-scale arrow/dart points (2), altered alligator
scutes (1), turtle shell gorget fragment (1), stingray spines (22), sharks teeth (7)

Potential Ceremonial Objects: medicine bundle/ceremonial cache: stingray spines, engraved ceramic disc,
turtle shell gorget fragment, Olive shell beads (2), alligator scute, brass buckle; shell
(Busycon) drinking cup, whelk columella spatula, ceramic adorno, celt, pottery used in

*previously noted faunal and botanical subsistence remains are not included here


References Cited

Adair, James
1973 Adair's History of the American Indians. Promontory
Press, New York.

Bartram, William
1955 Travels of William Bartram, edited by Mark Van Doren.
Dover, New York.

Brose, David S.
1984 Mississippian Period Cultures in Northwest Florida. In
Perspectives on Gulf Coast Prehistory, edited by Dave D.
Davis, pp. 165-197. Ripley P. Bullen Monographs in
Anthropology and History 5. University of Florida Press.

Cambron, James W. and David C. Hulse
1975 Handbook of Alabama Archaeology: Part 1, Point
Types, edited by David L. DeJarnette. Archaeological
Research Association of Alabama, Birmingham.

Fairbanks, Charles H.
1958 Notes on sites located on Four Mile Peninsula in South
Walton County, Florida. Ms. on file, Florida Master
Site File, Florida Bureau of Archaeological Research,

Ford, Pamela J.
1992 Interpreting the Grain Size Distribution of
Archaeological Shell. In Deciphering a Shell Midden,
edited by Julie K. Stein, pp. 283-326. Academic Press.

Fuller, Richard S. and Noel R. Stowe
1982 A Proposed Typology for Late Shell Tempered Ceramics
in the Mobile Bay/Mobile Tensaw Delta Region. In
Archaeology in Southwestern Alabama: A Collections of
Papers, edited by Cailup Curren, pp. 45-93. Alabama
Tombigbee Commission.

Hann, John H.
1988a Translation of the Florida Section of the Alvar Nunez
Cabeza de Vaca Accounts of the 1528 Trek from South
Florida Led by Panfilo de Narvaez. Ms. on file, Florida
Bureau of Archaeological Research, Tallahassee.

1988b Translation of The Florida of the Inca: History of the
Adelantado, Hernando de Soto, Governor and Captain
General of the Kingdom of Florida. Ms. on file, Florida
Bureau of Archaeological Research, Tallahassee.

1988c Florida's Terra Incognita: West Florida's Natives in the
Sixteenth and Seventeenth Centuries. The Florida
Anthropologist 46:61-107.

Harrisse, Henry
1892 The Discovery of North America. A Critical,
Documentary, and Historic Investigation. H.Stevens and
Son, London.

Hopkins, Stephen H.
1956 Notes on Boring Sponges in Gulf Coast Estuaries and
Their Relation to Salinity. The Bulletin of Maritime
Science 6:44- 58.

1962 Distribution of Species of Cleona on the eastern Shore of
Virginia in Relation to Salinity. Chesapeake Science

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

Jones, Calvin B.
1973 A Semi-Subterranean Structure at Mission San Joseph de
Ocuya, Jefferson County, Florida. Bureau of Historical
Sites and Properties Bulletin 3:1-50.

1982 Southern Cult Manifestations at the Lake Jackson Site:
Salvage Excavation at Mound 3. Midcontinental Journal
of Archaeology 7(1):3-44.

1994 The Lake Jackson Mound Complex (8Lel): Stability and
Change in Fort Walton Culture. The Florida
Anthropologist 47: 120-146.

Kent, Bretton W.
1992 Making Dead Oysters Talk, Techniques for Analyzing
Oysters from Archaeological Sites. Maryland Historical
and Cultural Publications, Crownsville.

Lazarus, William
n.d. Notes on the following sites: 8WL30, 8WL33,
8WL38/39, and 8WL50). Ms. on file, Florida Master
Site File, Florida Bureau of Archaeological Research,

Lee, Chung Ho and D. Joy
1989 Archaeological Investigations at Bayside Properties,
Santa Rosa County, Florida. University of West Florida
Office of Cultural and Archaeological Research Report of
Investigations 24. Pensacola.

Mikell, Gregory A.
1990 The Sheephead Bayou Site (8BY150): A Single
Component Fort Walton Hamlet in Northwest Florida.
The Florida Anthropologist 43:198-208.


1992 The Fort Walton Mississippian Variant on the Northwest
Florida Gulf Coast. Southeastern Archaeology 11:51-65.

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

Moore, Clarence B.
1901 Certain Aboriginal Remains of the Northwest Florida
Gulf Coast, Part 1. Journal of the Academy of Natural
Sciences of Philadelphia 11.

1918 The Northwest Florida Coast Revisited. Journal of the
Academy of Natural Sciences of Philadelphia 16.

Nidy, Scott L.
1974 An Archaeological and Historical Survey of the San
Destin Property in Walton County. Florida Department
of State, Division of Archives, History, and Records
Management. Miscellaneous Project Report Series 13.

Oetelaar, Gerald A.
1993 Identifying Site Structure in the Archaeological Record:
An Illinois Mississippian Example. American Antiquity

Perino, Gregory
1968 Guide to the Identification
Points. Bulletin No. 3,

1971 Guide to the Identification
Points. Bulletin No. 4,

of Certain Indian Projectile
Oklahoma Anthropological

of Certain Indian Projectile
Oklahoma Anthropological

Phillips, Phillip
1970 Archaeological Survey in the Lower Yazoo Basin,
Mississippi, 1949-1955. Papers of the Peabody Museum
of Archaeology and Ethnology 60. Harvard Peabody
Museum, Cambridge.

Powell, John
1990 Points and Blades of the Coastal Plain: A Guide to the
Classification of Native American Hafted Implements in
the Southeastern Coastal Plain Region. American
Systems of the Carolinas, Inc. West Columbia, S.C.

Portnoy, Andrea
1981 A Microarchaeological View of Human Settlement Space
and Function. In Modern Material Culture: The
Archaeology of Us, edited by R.A. Gould and M.B.
Schiffer, pp. 213-224. Academic Press, New York.

Scarry, John F.
1984 The Development of Mississippian Chiefdoms in
Northwest Florida: Fort Walton in the Upper
Apalachicola Valley. Ph.D. dissertation, Case Western
Reserve University, Cleveland.

1985 Proposed Revision of the Fort Walton Ceramic
Typology: A Type-Variety System. The Florida
Anthropologist 38(3):199

1990a Mississippian Emergence in the Fort Walton Area: The
Evolution of the Cayson and Lake Jackson Phases. In
The Mississippian Emergence, edited by Bruce D. Smith,
pp.227- 250. Smithsonian Institution Press, Washington.

1990b Beyond Apalachee Province: Assessing the Evidence
for Early European-Indian Contact in West Florida. In
Columbian Consequences, Volume 2, Archaeological and
Historical Perspectives on the Spanish Borderlands East,
edited by David Hurst Thomas, pp. 93-105. Smithsonian
Institution Press, Washington.

Smith, Hale G.
1951 A Spanish Mission in Jefferson County, Florida. In Here
They Once Stood: The Tragic End of the Apalachee
Missions, by Mark F. Boyd, Hale Smith, and John W.
Griffin, pp. 107-136. University of Florida Press,

Stuiver, M. and B. Becker
1986 The CALIB Program for Calibration of C-14 Dates and
Its Implications. Radiocarbon 28:863-910.

Swanton, John R.
1911 Indian Tribes of the Lower Mississippi Valley and
Adjacent Coast of the Gulf of Mexico. Bureau of
American Ethnology Bulletin 43. Washington, D.C.

1922 Early History of the Creek Indians and their Neighbors.
Bureau of American Ethnology Bulletin 43. Washington,

1928 Social Organization and Social Usages of the Indians of
the Creek Confederacy. 42nd Annual Report of the
Bureau of American Ethnology, pp.23-472. Washington,

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

Thomas, Prentice
1989 Summary of Investigations at the Monday Midden Site
(8WL99). New World Research, Inc. Report of
Investigations 185.

Thomas, Prentice and Jan Campbell (editors)
1990 Cultural Resources Investigations at Eglin Air Force
Base, Florida: Technical Synthesis (Draft). New World
Research, Inc. Report of Investigations No. 192.

Waselkov, Gregory A.
1982 Shellfish Gathering and Shell Midden Archaeology.
Ph.D. dissertation, Department of Anthropology,
University of North Carolina, Chapel Hill.

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

Gregory A. Mikell
2572 Barron Court
Shalimar, Florida 32579



William Gray Johnson

Florida archaeologists have long recognized the
application of aerial photography for detecting earthwork sites
in the Lake Okeechobee Basin. Ross Allen (1948) utilized this
technique to portray Tony's Mound (also known as the Big
Circle Mounds). John Goggin applied the technique to locate
numerous sites in the area (Florida Museum of Natural History
site files) and noted how useful it was to view the large
earthworks from the air (Goggin and Sturtevant 1964:196).
Robert Carr's (1975) survey around Lake Okeechobee also
utilized this technique. However, William Sears' (1982)
publication of an aerial photograph of the Fort Center site and
Carr's (1985) article on circular earthworks are probably the
best known examples of utilizing aerial photographs to
document sites in this area. Based on these works, the West
Okeechobee Basin Project was developed to conduct a
comprehensive overview of the aerial photographs on the west
side of Lake Okeechobee, ground- truth accessible locations,
and conduct informant interviews to document archaeological
sites in the area. This article provides some of the results of
that project with a focus on the Kissimmee River Basin.

Aerial Photography: Early versus Late

Aerial photographs were taken in south Florida as early
as the late 1930s (University of Florida, Marston Science
Library files). The purpose was to characterize the landscape
for agricultural use. The earliest of these show few cultural
modifications whereas the more recent indicate major
alterations to the environment from agricultural activities. By
far, the most prevalent modifications the increased number of
drainage ditches and roads and the resulting changes in
vegetation patterns. Such changes have obscured many large
prehistoric earthworks and, in some cases, have obliterated
them. Thus, the early aerial photographs provide the best
record (and in some cases, the only record) of these sites. On
the other hand, the later photographs can be extremely useful
to researchers in determining modern damage to sites as well as
for providing a basis for deciding what elements remain.

The West Okeechobee Basin Project

The West Okeechobee Basin Project was designed to

review aerial photographs of the west side of the Lake
Okeechobee Basin (Figure 1) to locate prehistoric earthworks.
The review was followed by a pedestrian survey to ground-
truth possible site locations. During fieldwork, informants
provided information on sites that were not observed on aerials
but could be discerned on them after securing locational data.
A total of 38 previously unrecorded sites and additional
information on nine recorded sites was obtained from this
project (Johnson 1990). Of these, the earthworks located
along the Kissimmee River proved to be the most interesting
and puzzling. On the Okeechobee County side of the river,
they are the Fort Kissimmee Earthworks (80B28), Fulford
Earthworks (80B29), Clemens Square and Mound (80B30),
Kissimmee Circle (80B31), Whitehurst Mound (80B32), and
Underhill Sawgrass Pond site (80B33); and on the Highlands
County side, Daugherty (8HG3) and Highlands Parallel Ridges
(8HG636). In addition, an effigy borrow called Oxer Borrow
(8GL79) is noted here because of its uniqueness. The
following narratives provide details on each.

Fort Kissimmee Earthworks

The Fort Kissimmee Earthworks consist of a large
rectangular- shaped ditch with rounded covers and measures
approximately 450 meters north-south by 200 meters east-west
(Figure 2). The northwest corner of the rectangle opened to a
bend in the Kissimmee River such that water could flow from
the river into the ditch. Two embankments measuring
approximately 110 meters each were located within the interior
of the rectangular earthwork and may have terminated at
mounds. Though none are visible on Figure 2, many such
embankments at other prehistoric sites in the area, such as the
Big Circle Mounds, terminate at mounds (see Allen
1948:Figures 5-7).
This site was not visited during the field survey. Access
is restricted by the Kissimmee River on one side and the
Kissimmee Canal on the other. Aerial photographs reveal that
construction of the Kissimmee Canal resulted in destruction of
about two-thirds of the site, leaving the northwest corer of the
earthwork intact (see large arrows on Figures 2 and 3). The
embankments and most of the ditch were dug through during
excavation of the canal (Figure 3).


Vol. 47 No. 3


---- -






---- ----I



g ln -
1 *- -






-- -------

Approximate Scale


50 Miles

50 K-lomelers
60 KIlornelers

BMJ 93

Figure 1. Area of the West Okeechobee Basin Project unshadedd) and sites featured in this article.





Figure 2. Fort Kissimmee Earthworks (from aerial CYW-4H-
178 dated 1-26-53).

Fulford Earthworks

The Fulford Earthworks consist of a large rectangular-
shaped ditch with rounded covers measuring approximately
270 meters north-south by 225 meters east-west (Figure 4). In
addition, a linear ditch extends from the southwest corner of
the earthwork approximately 225 meters west, which connects
the earthwork to the Kissimmee River. Similar to the Fort
Kissimmee Earthworks, an embankment measuring
approximately 100 meters is located within the interior of the
earthwork and may terminate at a mound.
Most of this site was covered by fill from the dredging of
the Kissimmee Canal and was bisected by a modern ditch that
forms the boundary of the canal fill (Figure 5). As such, much
of the site is no longer visible. Only the northeast corner and
eastern side of the earthwork remain undisturbed. In this area,
the embankment extends from the east side to the interior of
the earthwork and is cut off by the modern ditch at about 20
meters. It cannot be discerned beyond the ditch because of the


fill from the Kissimmee Canal but is thought to be intact. One
50-cm diameter shovel test placed in approximately the center
of the earthwork (where the mound--if there ever was one--
might still be located) revealed only that the earthwork must
have been flooded many times prior to the dredging of the
Kissimmee Canal. Obviously stratified grey and tan sands lie
under 20 cm of grey sand that contains small shell fragments,
or what appears to be fill from dredging. The fill covers the
rest of the site such that one can only discern its outline by the
presence of cabbage palms that are growing on the earthwork's
The edges were created by digging ditches and piling the
dirt to form a berm on the inside of the earthwork (much like
Sears [1982:175] description of the construction of the Great
Circle at Fort Center). Three 50-cm diameter shovel tests and
seven bucket auger tests failed to yield any artifacts nor were
any found on the surface of the site. Soil profiles suggest
considerable time has elapsed since construction of the berm,

Clemens Square and Mound

The Clemens Square and Mound site has a large square-
shaped earthwork with squared corners measuring
approximately 165 meters on a side (Figure 6). Data on
construction of the earthwork are limited to the observation
that the berm is located on the outside of the earthwork
(opposite of the construction of the Fulford Earthworks
[above] and the Great Circle at Fort Center). The mound part
of the site is located approximately 300 meters northeast of the
square and measures approximately 36 meters north-south by
32 meters east-west at a maximum elevation of 1/2 meter
above the surrounding terrain. Between these lies a small
unnamed pond that is possibly the source of fill for the mound.
The mound and earthwork are presented as associated
parts of the site because one photograph observed during
research showed the pond to be a circular-shaped borrow
(though this is not apparent in Figure 6). In addition, the
earthwork has a slight curvature just right of the center of the
berm adjacent to the wooded vegetation. Examination of this
curvature in the field revealed a slight drainage from the pond
that connects the pond and earthwork (albeit a tenuous
connection because the drainage appears to be natural rather
than cultural).
One 50-cm diameter shovel test was dug in the mound
and yielded ceramics and one lithic flake. A summary of the
materials retrieved from the shovel test is presented in Table 1.
The presence of the sherds and flake suggest an
interpretation of the mound as a habitation locus, but its
proximity to the earthwork and lack of animal bone (ubiquitous
in the habitation mounds in this area) are indicators that it may
have served some other purposess. Sherds and flakes are also
found in burial mounds in the area and the results from this
one shovel test can only confirm that the mound is from Belle


'Wscale 1:9125
Iscale 1:9125

Figure 3. Fort Kissimmee Earthworks after construction of the Kissimmee Canal (from aerial 12055 174-126 dated 2-23- 74).

Figure 4. Fulford Earthworks (from aerial CYW-3C-23 dated 2-27-44).


S cabbage pa

Figure 5. Fulford Earthworks after construction of the Kissimmee Canal (from aerial CYW-1MM-84 dated 12-27-70).

Glade cultural activities. It should be noted here that the pH
from the first and second levels indicate almost neutral soils
(readings for both levels are 6.7), thus suggesting that soil
acidity is not a factor in the absence or lack of preservation of
animal bone. As such, the mound is considered an integral
part of the site, but not necessarily a focus of habitation

Kissimmee Circle

The Kissimmee Circle is a semicircular arc with a linear
embankment attached along the northern portion of it (Figure
7). A road (now replaced by Highway 98) was built on top of
the linear embankment so that it practically cut the site in half.
Field inspection of the area revealed nothing left of the
embankment or road nor could any part of the arc be
A comparison of several aerial photographs of the site
over a couple of decades reveals the old road (and thus, the
linear embankment) was located to the east of present-day
Highway 98. The arc is probably still in existence but is
buried under fill from the Kissimmee Canal. Unfortunately,
the linear embankment (or what was left of it after the road
was built on top of it) may have been used for fill in the
construction of Highway 98.
Based on similar sites and how they appear on the aerial
photographs, the arc probably was composed of a crescent-

shaped embankment. Its appearance is very similar to 8GL26,
the Lakeport Earthworks (Carr 1975).

Whitehurst Mound

The Whitehurst Mound consists of a linear embankment
approximately 200 meters long (northwest-southeast) by 25
meters wide and at least 1 meter above the surrounding terrain
(Figure 8). Two 50-cm diameter shovel tests yielded no
cultural materials nor were any found on the surface. This
absence strongly suggests that the site was not utilized for
habitation purposes. Rather, this mound and two like it
(8HG13 and 8HG636) must have had some other function.
Their proximity to the Kissimmee River suggests some sort of
riverine orientation.

Underhill Sawgrass Pond Site

The Underhill Sawgrass Pond site is a habitation mound
that measures 31 meters north-south by 40 meters east-west
and has a maximum elevation of approximately 1 meter above
the surrounding terrain. Initial inspection of the site resulted
from a conversation with an informant, Mr. Jeff Clemens, and
subsequent inspection of the aerial photographs indicates a
linear embankment is associated with this site (Figure 9). One
50-cm diameter shovel test in the mound yielded a large
quantity of animal bones but no artifacts (at least none in the


Figure 6. Clemens Square (as seen on aerial CYW-5C-7 dated 4-13- 44).

Figure 7 Kissimmee Circle before construction of the Kissimmee Canal (aerial CYW-2C-15 dated 2-27-44).


1 J '
0 6 -scale 1:8000
C~'k~ %

Figure 8. Whitehurst Mound (aerial CYW-5C-10 dated 4-13-44).

[scale 1:8000

Figure 9. Underhill Sawgrass Pond (aerial DSL-2CC-62 dated 1-31- 62).

Figure 10. Circular borrow at the Daugherty site (aerial CYW-2C-40 dated 2-27-44).

" 1 "

Figure 11. Highlands Parallel Ridges (aerial CYW-2C-13 dated 2-27- 44).


1/4" fraction). The weights of recovered animal bones are
presented in Table 2.
The large amount of animal bones suggests that this site
functioned as a habitation site. Species represented in the
faunal assemblage include, but are not limited to, members of
the order Testudines (turtle), Squamata: Serpentes (snakes),
members of the family Rodentia (rodents), and members of the
Class Aves (birds). Additional tests at the site would likely
yield Belle Glade Plain ceramics as other sites like this one

Daugherty Site

The Daugherty site is well documented in the Florida
Master Site Files with a description of several mounds and
some of the artifacts that have been observed there. The
review of aerial photographs coupled with information
obtained from James Marshall (personal communication 1989),
a surveyor interested in prehistoric earthwork geometry, has
revealed a feature of the site that has never been documented.
It is a circular borrow located due south of the site across
Highway 98 (Figure 10). This sort of geometric- shaped
borrow has been mentioned only casually at other sites in the
Okeechobee Basin (cf. Sears 1982:175; Carr 1985:300).
However, due to the efforts of this project, we now know that
similar borrows were found at the Fort Center site (8GL13),
Clemens Square and Mound (80B30), the Lakeport Circle
Ditch (8GL50), and Maple Mound (8HN5).

Highlands Parallel Ridges

The Highlands Parallel Ridges appear on the aerial
photographs as two parallel ridges measuring approximately
400 meters long (Figure 11). A possible crescent-shaped
earthwork can be seen on the same aerial about 350 meters
northwest of the ridges. This site was not visited because
permission to enter the property was restricted. However, its
similarity to the Highlands Linear Ridge site (8HG13), located
less than a mile to the north, and the Whitehurst Mound
suggests it is not a habitation site. Further research is needed
to develop an interpretive model.

Oxer Borrow

The Oxer Borrow appears on the aerial photographs as an
anthropomorphic-shaped borrow complete with a head, body,
and arms (Figure 12). Subsequent field inspection revealed a
barely discernible linear depression (the body portion of the
borrow) located in a pasture. This portion is visible by the
darker color of the grass that grows within it. An informant,
Mr. Brad Oxer, states that it tends to hold standing water
during the rainy season. The rest of the borrow is no longer
discernible as the head and arms have been obliterated by
modern agricultural practices. Mr. Oxer revealed that a sand

feature in the woods (located just northwest of where the head
used to be and in alignment with the body) could be explained
by the borrow. It is approximately 30 meters in diameter and
not even .25 meter high. According to Mr. Oxer, this sand
feature has always puzzled him, his father, and his grandfather
ever since his grandfather discovered it and used it for a
garden. A single 50-cm diameter shovel test was dug in the
sand feature, but no cultural materials were found. Subsequent
testing by Buchner (1992) around the borrow and sand feature
found no cultural materials. However, he believes that closer
examination of the soils may yield significant data on borrows
and the Belle Glade culture (Buchner 1992:62).
The Oxer Borrow remains an enigma. Only one other
site is known that is similar, the Pestle Earthworks (8GL43).
Like the Oxer Borrow, the Pestle Earthworks is (or was when
it was recorded) barely discernible as it is only several inches
lower than the surrounding terrain. However, it shows up
clearly on some of the early aerial photographs (see
documentation on this site in Florida Master Site Files) but not
others. This suggests that the ability to discern the site
depends on wet conditions (that is, when wet conditions
prevail, the site is discernible).


Early aerial photographs enabled identification of seven
new sites along the Kissimmee River and new information was
gathered for a recorded site. In addition, an effigy borrow was
identified. These sites represent an expanded definition of the
Belle Glade culture, the prehistoric culture of the Lake
Okeechobee Basin. The square and rectangles, the
anthropomorphic and circular borrows, and the linear
embankments indicate greater variation in Belle Glade
earthworks than has been previously described.
The square and rectangles are perhaps the most intriguing
of these newly identified earthworks. Their limited
distribution along the Kissimmee River coupled with a lack of
diagnostic artifacts may lead some readers to question their
association with the Belle Glade culture. It could be argued
that such shapes may be associated with much later Seminole
or historic Euroamerican activities. I argue for a Belle Glade
origin based on the berm and ditch construction similar to
other Belle Glade earthworks in the region, the presence of
linear embankments at the two rectangular- shaped earthworks,
and the apparent association of the mound and earthwork at the
Clemens site. Moreover, square-shaped earthworks are known
from other prehistoric sites in North America. Squier and
Davis (1848) documented numerous examples at Hopewell
sites in Ohio. Thus, the evidence supports a Belle Glade
origin for the square- and rectangular-shaped earthworks along
the Kissimmee River.
Mounds, ditches, borrows, and embankments are the
basic elements of all Belle Glade earthworks. Generally, they
are found in combination but sometimes occur in isolation.


Figure 12. Oxer Borrow (aerial BUO-1D-41 dated 12-5-48).

Table 1. Materials Retrieved From Shovel Test at Clemens Mound.

Level* Materials

1 10 Belle Glade Plain sherds
2 1 tertiary flake; 3 Belle Glade Plain sherds
3 2 Belle Glade Plain sherds

20-cm increments

Table 2. Material Retrievedfrom Shovel Test at Underhill Sawgrass Pond.

Level* Materials

1 1,190 g of animal bones (260 g in 1/4" fraction;
930 g in 1/16" fraction)
2 1,167 g of animal bones (362 g in 1/4" fraction;
805** g in 1/16" fraction)

20-cm increments
** weight includes concretions


Variations in size, shape, and associations probably represent
functional differences as well as chronological distinctions.
These features are discernible on aerial photographs. The
benefits of utilizing early aerial photography to identify their
locations is stressed because modem land-use practices have
altered and, in some cases, completely obliterated many of the
sites. Thus, it is important for researchers working in this part
of the state to use them. Faint traces of the earthworks may be
visible on recent photographs, but the early photographs
provide the best means of identifying and locating the sites.


The West Okeechobee Basin Project was funded by a
grant from the Florida Department of State, Division of
Historical Resources, Historic Preservation Grants-in-Aid

References Cited

Allen, Ross
1948 The Big Circle Mounds. The Florida Anthropologist

Buchner, C. Andrew
1992 Phase I Cultural Resources Survey at the Proposed
Glades County Landfill Site, Florida. Garrow and
Associates, Memphis.

Carr, Robert S.
1975 An Archaeological and Historical Survey of Lake
Okeechobee. Miscellaneous Project Report Series No.
22, Bureau of Historic Sites and Properties, Division of
Archives History and Records Management, Tallahassee.

1985 Prehistoric Circular Earthworks in South Florida. The
Florida Anthropologist 38:288-301.

Goggin, John M., and William C. Sturtevant
1964 The Calusa: A Stratified, Non-agricultural Society (with
Notes on Sibling Marriage). In Explorations in Cultural
Anthropology: Essays in Honor of George Peter
Murdock, edited by Ward H. Goodenough, pp. 179-219.
McGraw-Hill, New York.

Johnson, William Gray
1990 A Report of Investigations on the West Okeechobee Basin
Archaeological Survey. Florida Museum of Natural
History, Gainesville. Florida Department of State,
Division of Historical Resources, Tallahassee.

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

Squier, E. George, and E. H. Davis
1848 Ancient Monuments of the Mississippi Valley.
Smithsonian Contributions to Knowledge 1. Smithsonian
Institution, Washington, D.C.

William Gray Johnson
Desert Research Institute
755 East Flamingo Road
Las Vegas, Nevada 89119



Clark Hardman, Jr. and Marjorie H. Hardman

Large points are always a striking find and finding 55 in a
cache less than one meter in diameter is something of an
exciting find. Some time ago, Calhoun W. Hendrix, a
neighbor, discovered a few (two or three) large points on the
surface. Using his bare hands, he "excavated" further and
soon uncovered a total of fifty-five points. We borrowed the
artifacts, photographed them, and obtained information of the
circumstances of their discovery.
The artifacts were found on a coastal ridge (Figure 1) in
southwestern Taylor County. The immediate question was is
there anything associated with the artifacts? Mr. Hendrix had
marked the spot. We returned to the site with Mr. Hendrix to
check the area. The ground surface (Figure 2) was white sand.
Top soil had been removed to a loss depthwise of 30 to 60
centimeters sometimes more, as a result of mechanically
scraping the larger area within where the cache was found.
While we did find a few broken biface tips, nothing was
associated directly with the points at the immediate surface or
in his "excavations," despite our close search of the area.
Roughly, the points were found north of the sink. Sherds
and flakes were found, however, on the east and west sides of
the sink, perhaps 100 meters distant from the cache location.
Our surface collection included 52 chert flakes, one well-
patinated chert scraper-knife, and 500 sherds. Three hundred
and fifty of the sherds measured less than 2.5 cm. The
remaining 150 sherds are larger, generally 4-5 cm with the
largest only about 8 cm long. Mean thickness of 130 larger
sherds was 8.2 mm with range of 6 to 12 mm. Three sherds
have repair holes. One rim sherd has a pronounced out-flare -
2 to 3 cm of flare, while three rims (one with a repair hole)
have flat-edged straight sides. One small sherd, is estimated to
be from a five to eight-cm deep, five-mm thick, well-fired
bowl. One lump may be from a vessel foot or ear. A portion
of a handle was also found. Most sherds are plain, either from
undecorated vessels or from the undecorated portions of
vessels with restricted decorated areas. Decorated sherds
include one Tucker Ridge-pinched sherd, one slightly out-
flared rim with three incised lines, one unique incised or
engraved sherd, two punctated sherds with an incised line (not
random), four random punctated Lochloosa Punctated variety
Lochloosa (Worth 1992:194), and 24 roughened Fig Spring
Roughened, generally variety Santa Fe (Worth 1992:194).
Sherd paste was sandy and the pottery generally not well fired.
With six tentative exceptions, sherds were too small for
estimates of diameters. One sherd appears to be from a very
small "pot" with a diameter of about 2.5 cm. Five other
sherds appear to be from larger vessels ranging from 15 to 30
cm in diameter.

The Cache

The 55 large points are a fascinating assemblage. Figure
3 shows all the points. We considered the points to have been
made from Florida or Georgia chert. The workmanship can be
considered excellent to only fair. Measurements done by Mr.
Hendrix are length (Figure 4), width (Figure 5), and maximum
thickness (Figure 6). The colors are off-white (not grey) with
reddish streaks and areas. The points did not show patination
or wear.
Classifying these implements is a problem. The
expanding tang, corner notched type fits the Lafayette (Bullen
1975:26) description. The straight tanged, corner notched
specimens best fit the Culbreath (Bullen 1975:28; Warren et al.
1967) type. Many of the implements we classify as Florida
Archaic Stemmed (Bullen 1975:32) type.
Using the photographs, we classified twenty-eight of the
points as Florida Archaic Stemmed, fourteen as Culbreath and
thirteen as Lafayette. In some instances, the criteria tended to
become hazy and the classification a forced fit. Something of
the nature of it's one of these, which? At another time we
did not get the same result. Our estimates were based
primarily on the basal configuration which we consider to be
Bullen (1975:26,28,32) gives the data estimates for these
types: Lafayette, "radiocarbon dated at circa 960 B.C.";
Culbreath points, "belong in the very late Archaic period, circa
3000-1000 B.C."; Florida Archaic Stemmed were "made over
several thousand years, 5000-1000 B.C." Regardless of
typology, the point assemblage hangs together as being very
The type problem is illustrated in Figures 7 and 8.
Figure 7 is an excellent example of a Lafayette (Bullen
1975:26) point. Figure 8, which we also classify as a
Lafayette type considering basal treatment, is roughly one-
third longer than the type maximum and the overall shape is
different from the Bullen description. Similar problems crop
up with the Culbreath and Florida Archaic Stemmed. The two
types tend to merge. Overall the points in the Taylor County
cache tend to be longer and wider than the descriptions given
by Bullen. However, some of the points may be unfinished
final stage reforms, as this large size would be expected since
Bullen's typology is based on finished specimens, generally
reduced in size through repeated blade edge maintenance. The
cache, then, offers a unique opportunity to better understand
lithic reduction sequences. It further reminds us that
established point typologies are heuristic devices created to
satisfy our own need to place artifacts in some form,,of


Vol. 47 No. 3



-R7 \.' R8E E IOE-- X tjlI.
.R5E 6E Perr "- --" \ "

.Hamoton \ 'FenholIowa -

Sponge Point-

& Long Crass Point
Rock Poin t ehae ~ '
oleyo 'C M '"May

Bowles Point '- --
e sn asDrue Pol st.nt -
peona a Point c e a

]- -

Fimare 1. Site location. The approximate location of the find is indicated by the black circle.

appear to have sought to achieve an ideal of form to satisfy a elsewhere.
desired function, they were limited by their skill and quality of
the lithic material being workBowleed. Discussionin
Some of the points in the cache seem to be final stagent
approximate Sithe final point form, but have not receivedion of their or not associated with other artifacts or features, and puzzling.le.

final flaking needed sought to achprovideve finished blade edges and forms. Recently, Tesar and Watson (1993:68-81) report on where.
Thdesired final flaking liklction, they wouere limited by thin a greater uniformality ofin cache of "eight large prehistoric chert bifaces" found in Bay
theform. Theic material behind worked.manship suggest that the points County, Florida. They briefly describe other biface cacheson
werefo made by a single flintknapper. The absence of quarry with as many aches artwentyone unusual, rather rTesare, generally iWatsolated


conclude with a request for information about other finds and
comment that "more research needs to be done to better
understand the phenomenon of caching artifacts." We agree
and add that some sort of theoretical construct or model would
be an aid in testing the meaning and function. This seems to
involve psychological considerations and as such is rough
country for prehistoric study.
Also, Elaine Holzapfel (1993:30-37) recently did a study
of prehistoric flint caches in the Ohio area. Her study included
"open-ground caches only." The largest cache she described
contained 342 Adena type blades. The "largest cache ever
discovered" was 8300 flint discs from Mound 2 at Hoerwell
(Moorehead 1922:95). She briefly describes 46 caches in the
Ohio area, locates them by county and classifies them as:
Plano caches, Red Ocher caches, Adena caches, Hopewell
caches, and caches of rough blanks. Holzapfel, like Tesar and
Watson ends with a plea of report and record. The Holzapfel
report suggests the possibility of such studies for other areas,
even Florida.
We call the artifacts of the Taylor County cache points
and implements. Bullen in his description of the Florida
Archaic Stemmed type suggests "they were hafted to spears
and propelled by atlatls or spear throwers...asymetric forms
are knives (Bullen 1975:32)."
The points in this cache range from essentially perfect
symmetry to knife edge curves on one side and straight on the
other. A knife has a multitude of uses including cutting,
scraping, prying, digging, sawing and other related uses. The
spear point needs to be symmetrical and balanced. Certain
items from this cache could and probably did have a variety of
It is reasonable to assume that the points were made by
one individual or workshop. This suggests a possible
deliberate variation in purpose and methods in hating.
Perhaps, it also suggest caution in classification.
Surface material collected from the surrounding area does
not add information about the points. The surface collections
suggest a possible late date for the surface artifacts with the
exception of the knife-scraper. The Tucker Ridge-pinched
pottery (Willey 1949:428) is considered a Weeden Island type.
The roughened sherds fit the Fig Springs Roughened type
Weisman (1992:195) finds at the mission site. The other types
random punctated and undifferentiated punctate tend to be
Weeden Island or later. The surface artifacts do not add any
data that can be associated with the cache.
The sink is a good source of fresh water. A very small
stream runs from it. Small means it can be stepped across.
The source of fresh water, surface collections, and these points
suggest the site may have been occupied at least intermittently
over a long period of time. The reason for the abandonment of
the cache remains unknown and the subject of speculation.

Figure 2. Find area. The black circle indicates the location of
the cache.


Figure 3. All 55 points. Scale is 5 cm long.

13.0- 14.0- 15.0- CM
13.9 14.9 15.9

Figure 4. Distribution of point lengths.

7.5 8.0- CM
7.9 8A

Figure 5. Distribution of point widths.

13 15 17 19 21 23 MM
14 16 18 20 22 24

Figure 6. Distribution of maximum point thickness.


N 15
R 5

N 15
R 5




Figure 7. Lafayette point.

7T 7TTT7T T IVrTT TI T 1 1 1111111111111

Figure 8. Lafayette-like point. The darker streaks and areas
are shades of red.


We wish to thank Mr. Calhoun W. Hendrix for his
courtesy in letting us photograph the cache. We also
appreciate the suggestions of Brent R. Weisman.

References Cited

Bullen, Ripley P.
1975 A Guide to the Identification of Florida Projectile
Points. Kendall Books, Gainesville, Florida.

Holzapfel, Elaine
1993 A Study of Prehistoric Flint Caches in the Ohio Area.
Ohio Archaeologist 43(3):30-37.

Moorehead, Warren K.
1922 The Hopewell Mound Group of Ohio. Field Museum of
Natural History Publication 211, VI(5):95.

Tesar, Louis D. and Thomas C. Watson
1993 The Adkins Biface Cache Site (8BY701), Bay County,
Florida. The Florida Anthropologist 46:69-81.

Warren, Lyman O., William Thompson, and Ripley P. Bullen
1967 The Culbreath Bayou Site, Hillsborough County,
Florida. The Florida Anthropologist 20:146-163.

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

Worth, John E.
1992 Revised Aboriginal Ceramic Typology for the Timucua
Mission Province. Appendix D in Excavations on the
Franciscan Frontier: Archaeology of the Fig Springs
Mission, by Brent R. Weisman, pp. 188-205. University
of Florida Press, Gainesville.

Clark Hardman, Jr., and Marjorie H. Hardman
P.O. Box 667
Cross City, Florida 32628-0667



Louis D. Tesar

Over the years we all have asked the question, "what do
you call this point?" We have obtained the answer through a
variety of means, including relying on the opinion of someone
else who should know the answer or by obtaining a point
identification book and comparing the specimens in question
with the "identified types" described until we find a "best fit"
which may closely or only vaguely match.
The results of this deductive process have been both
positive and negative. It is positive when the identified types
are so well described and illustrated that several individuals
taking the same group of artifacts could use the same
publication and reach the same conclusions. The result has
been negative when the identified types are so generally
described and minimally illustrated that many of the artifacts
being identified will be classified differently by different
The reasons for this are the subject of this article. A
recently found unusual stemmed artifact, flaked on one side
and unifacial on the other, is described to illustrate that
established typologies do not cover the full range of
Ceramic, lithic and other typologies are heuristic devices.
Their purpose is to give order to what we observe. However,
it must be understood that the attributes used to establish
typologies are culturally biased. Indeed, the precepts upon
which they are based may be wrong--that is, based on
erroneous assumptions and deductions.
The most common error is the presumption that form
equals function. An example among lithic artifacts concerns
the bifacial lithic items commonly referred to as "spear
points," "dart points," and "arrowheads." Until recently it
was assumed that the time and effort invested in their
manufacture would only be for special-use purposes, since any
sharp flake could be used to cut hides and butcher animals.
The sequence assumed was evolutionary, from hand-held
thrusting and jabbing weapons like spears; to hand-thrown
weapons like darts, often propelled with the use of an atlatl; to
tips for arrows, with decreasing artifact size being the principal
identifying factor in its presumed function. Large points
cannot have been used to tip arrows because of their weight,
whereas smaller points almost certainly were used for this
purpose, so the reasoning goes.
The flaw in this reasoning, of course, was that it failed to
consider use wear and maintenance reduction. We now know

that these and other factors must also be considered in point
typologies. Indeed, most of the points found have been
reflaked during repeated repair and maintenance events until
they are broken or no longer useful. Further, use wear and
breakage analysis has shown that although some of the
identified artifacts functioned as projectile points, many others
functioned as knives, while some served as both projectile
points and knives, and still others were used as scrapers, drills,
perforators, and so forth. Indeed, an artifact's function likely
changed through different stages in its maintenance and use.
For that reason, the compound term "projectile point/knife"--
abbreviated as ppk--is now commonly used.
During the past 25 years, I have assembled a number of
publications dealing with projectile point/knife typologies. In
addition to distinct point types, nearly all of these publications
contain descriptions of what appear to be the same point type,
but with different names. Likewise, when the range of variety
within a point type is described and illustrated, it frequently
overlaps with examples in other point types. Furthermore,
almost none of the point typology publications adequately
consider point maintenance and repair reduction sequences.
The result has been increasing confusion.
The process of manufacturing and maintaining a stone
tool is a reduction process. Material is removed at each stage,
as well as through use. This process also results in changing
the shape of the object being used. The process may be
gradual if the tool remains unbroken and is simply maintained.
Or, the changes in form may be rapid if the object is broken
and reworked rather than discarded.
There are two primary classes of tools, expedient and
curated. Expedient tools are those prepared and used for a
single task as needs arise. They generally are flake tools, such
as knives or scrapers, and receive little or no edge maintenance
during the task at hand before discard. Curated tools are those
created for repeated use in specific tasks and generally are
maintained until exhaustion or breakage renders them useless.
They include both unifacial and bifacial tools and are often
hafted. The term "exhausted" is applied to curated lithic
artifacts that have been reworked until they can no longer serve
their intended function. For instance, the edge angle on a
knife may have become too steep to permit further
resharpening or perhaps further resharpening was not possible
without removing its hafting wrapping, at which time it likely
was discarded. Most of the whole ppks found by collectors,


Vol. 47 No. 3



especially Wacissa or Bolen Beveled ones, are exhausted ppks
(Figure 1).
The process of manufacturing bifacial ppks generally
begins at the quarry where suitable flakes are detatched for use
as tool blanks and where cores are formed to provide a
portable lithic material source for later flake detachment.
Flakes selected for later knapping into a specific artifact type
are referred to as blanks. The blanks are then knapped to a
desired shape from which the final artifact will be formed.
These intermediate artifacts are called reforms and may be
cached (placed in a safe location for storage) at or near to
quarry site or at some other site to which their maker plans to
return, or may be directly processed to finished products.
The final shape of reforms intended for use in making
ppks generally is ovoid or relatively flat-based with excurvate
sides. All sides of such reforms usually are characterized by
thin edges gently sloping toward a thicker central ridge.
Portions of the gradually thickened basal edges of such final
stage reforms are knapped to create recognizable lanceolate,
side-notched, corer-notched, basally-notched, stemmed, or
other forms. Preforms with unremovable lumps or inclusions
likely will be discarded without further reduction shaping,
although they may be utilized as expedient knives or scrapers
before final discard.
The Suwannee/Simpson ppk manufacture sequence
deduced by Daniel, Wisenbaker, and Ballo (1986:29-30) from
Harney Flats site material illustrates the ppk manufacture
process. Replication studies also have contributed markedly to
our understanding of this process. Figure 2 is used as a visual
dictionary for ppk terms used in this article. The other
illustrations are derived from a report recently completed by
the author (Tesar 1994), which served as the stimulus for this
Once a ppk is hafted, the hafted area (the basal or
proximal end) remains unchanged while it is hafted, except for
accidental breakage. However, repeated use and resharpening
of the blade (the unhafted portion of the ppk) reduces blade
size and changes its form, through the bifacial or unifacial
removal of flakes through use wear and especially through
pressure or percussion flaking to maintain a sharp edge (Figure
3). This process is also affected by breakage of the blade tip,
midsection and base or hafting area. Reworking the near-haft
blade shoulder area when the incurvate angle of that area
becomes too steep or the resulting barb-like projecting
shoulder (the basal corer of blades on a stemmed ppk) is
accidentally or intentionally broken during tool use and
maintenance also causes a change in form.
If the ppk tip (distal end) breaks, the repaired knife tip
may be reworked to a rounded cutting edge to conserve
material, or worked into a more truncated triangular point if a
sharp projectile point tip is required. If the break is diagonal,
it may be reworked to an asymmetrical knife cutting edge.
Furthermore, if the ppk tip is reworked to meet expedient

drilling and graving needs and then subsequently reworked to
meet its primary function, a truncated form similar to that of a
repaired broken tipped ppk may be produced.
A similar, more exaggerated process occurs with
midsection breaks, if the ppk user or scavenger (in this sense, a
person who searches older sites in lithic-poor areas for
reuseable material) determines that useable material remains.
Furthermore, some of the reworked ppks begin to take on new
names as a result of this process. For instance, unifacially
retouched, diagonally fractured Big Sandy, Bolen and
Hardaway Dalton side-notched ppks are often identified as
Edgefield Scrapers. Straight midsection breaks may be
bifacially or unifacially reworked to produce pointed tips or
unifacially reworked as hafted end scrapers (Figure 4). While
such reworked specimens may be expediently used and
discarded, they generally are assumed to have been made part
of a curated tool kit.
When the blade breaks near the base, both the remaining
basal and blade fragments may be unifacially reworked as end
scrapers, if sufficient material remains (Figure 4).
Alternatively, the broken blade might be reworked near the
basal area to create a new hafting area (Figures 5 and 7B). If
the blade is reworked close to the time of the breakage, it
likely will be restemmed or tanged in the same or similar style
as in the original. However, if the broken blade is later
scavenged and reworked, the stem or tang likely will be of a
different style--similar to one commonly used by the later
Reworked blades are generally recognized by the contrast
in the knapping style on the remaining blade versus the larger
percussion flakes in the restemmed or retanged area. They
may also be recognized by the relatively steep angle in basal
blade thickness, being somewhat stubby and thicker, tapering
more sharply from a thin edge to a thicker midsection than in
the original (Figure 6). However, whether original or
reworked, the form of the hafting area is the most stable
element in a ppk and is thus the most diagnostic element, since
the blade form constantly changes through use and
Nevertheless, basal forms do change as a result of
accidental breakage or in response to point maintenance needs.
Accidental breakage may be ignored by the ppk user if it does
not affect tool function and use. However, if tool function and
use are affected, an effort to repair the tool will be made,
unless the flawed tool is discarded. The repaired base of the
blade or hafting stem or tang may be asymmetrical if only the
damaged is reworked (Figure 7A, C); however, it will be
symmetrical if the knapper determines that a balanced form is
important and reworks both the broken damaged area and its
opposite undamaged counterpart.
A ppk with distinctive blade shoulders may go through a
blade maintenance reduction sequence, and then have the
shoulders removed (Figure 8). Such reworking of diagnostic


Figure 1. Sketches of exhausted Bolen ppks from Johnson Sand Pit site (8LE73).

Figure 2. Common terms used to identify various portions or areas of ppks
Sketch based on earlier reduction stage of Figure 1 (left) ppk.

discussed in this article.



stem or tang


Tip (distal end)


Figure 3. Three similar Johnson Sand Pit site (8LE73) Kirk Stemmed Serrated subtype 2 ppks
illustrating maintenance reduction sequence.

Figure 4. Examples of Johnson Sand Pit site (8LE73) reworked ppks. (A-C, F) Stemmed ppk basal
sections reworked as end scrapers. (D-E) Big Sandy/Bolen (?) beveled blades reworked as end scrapers.


Figure 5. Retanged Kirk Serrated (?) ppk blades.

Figure 6. Big Sandy variety Bolen subtype 2 ppks. (A, D) Original basalforms. (B-C) Reworked basal

j ,.I

Figure 7. Examples of reworked Hamilton-like ppk blade shoulders and tangs.

I a



Figure 8. Hamilton-like ppk examples of blade reduction.

ppk attributes can result in the original ppk type, such as a
Hamilton, being identified as another ppk type, such as a
concave-based Wacissa subtype 1 or Arredondo-like ppk.
Finally, the initial ppk form is governed by the skill of
the knapper and the shape and workability of the preform
material being knapped. Thus, a desired end product may or
may not be achieved, depending on such abilities and
limitations. The artifact described below is an unusual
example illustrating this point (Figure 9).
For the above reasons, ppk identification should be
approached cautiously, and the qualifier "-like" appended as
appropriate to named specimens. Furthermore, although
caches of pristine initial stage ppks or unmodified isolated ppks
are rarely found, and lost or broken partially reduced ppks are
more frequently found, it must be remembered that most ppks
found at sites, especially at base camps, represent manufacture
failures, exhausted ppks, and ppks discarded during
maintenance because of unacceptable flaws.
Ppks discarded by their user as flawed or exhausted (not
useable) comprise the vast majority of the ppks collected and
studied. Much value lies in the deductive conclusions that may
be reached concerning manufacture and maintenance
sequences, tool use, and the parameters of form considered
acceptable or desirable by their maker. Also the extent of
scavenging and reworking of broken artifacts also provides
data on lithic resource availability. Finally, caches of artifacts
in preform or finished, essentially unused form provide a
baseline on the range of attributes considered acceptable in the
initial manufacturing stage, the beginning of the maintenance
reduction continuum.

(Left) shoulders removed.

Any ppk analysis must be undertaken with an
understanding of the preceding limitations and considerations.
While such analysis will always involve a consideration of
reduction processes, they need not be typological dilemmas if a
conservative identification approach is used.
Nonetheless, some artifacts are anomalies. One such
unique artifact was recently found at site 8LE148, near
Tallahassee, Florida. This artifact was surface collected from
a portion of the site scheduled to become a multi-acre
stormwater runoff pond dug as part of middle school
construction. The Bureau of Archaeological Research,
Division of Historical Resources, because of the importance of
the site, sought permission to conduct emergency excavations.
The salvage excavation field work was supervised by B. Calvin
Jones with volunteer help from March through June, 1994, and
resulted in the recovery of significant data on the late Deptford
through early Weeden Island cultural transition (ca. 1 B.C.-
A.D. 500), as well as limited information on Late Paleoindian
through Middle Archaic occupations. Jones and I are
collaborating on the analysis and report preparation. An
equally large area containing even more significant elements of
the site than those lost has been set aside for permanent
protection through the cooperation of the former property
owner and the County School Board.
From one side the unique 8LE148 artifact appears to be a
stemmed bifacial ppk, while from the other it is unifacial
(Figure 9). It has been accessioned (Number 94A-68) in the
Bureau of Archaeological Research's research and type
collections. It is made from a fossiliferous chert that has
patinated to a pale grayish-white color, suggesting that the


ii ,
I :1"~
J. I ~II I

BIYLtlll I HV3~VFL~flT:!'I, ';


Figure 9. Unique unifacial ppk-like artifact from site 8LE148 near Tallahassee, Florida. (Left) Bifacial side, (Right)
unifacial side. (Photograph prepared by Roy Lett, Bureau of Archaeological Research, Division of Historical Resources.)

original material likely was grayish translucent. It has
dimensions of 83 mm long, 38 mm wide at the shoulders, and
has a 13 mm long stem that is 21.5 mm wide near its base.
The thickest portion of the specimen is 9 mm at the interface of
the stem with the shoulders at the blade base. The object was
shaped through percussion flaking. All but four flakes were
detached from one side. There is no hafting edge dulling.
However, there is use wear dulling from the tip up 37 mm
along the excurvate blade edge. The dulling appears to be the
result of use as a knife and shave on a medium hard to hard
surface (such as wood or bone).
Because of its unfinished condition this artifact is not
assignable to any specific ppk type. Indeed, it lends itself to
speculation as to its origin. One possible explanation is that it
was created by an individual, perhaps a child, learning to flint
knap. It is speculated that the item then had a handle affixed
and that the split handle was so wrapped with thongs about the
tang and the upper blade as to leave only the lower blade area
available for use. The minimal use along the lower edge
indicates that it was discarded or lost shortly afterwards.
In conclusion, the study of ppks can provide and

understanding on changes in lithic technology, the availability
of lithic resources, resource exchange networks, subsistence
patterns, and cultural norms, at least as they relate to ppk
styles. Recognition of lithic tool production and maintenance
and repair sequences is important, as are use wear analyses.
Care should in recording the context from which ppks are
found in excavations or elsewhere. Point identification should
be approached conservatively, using the qualifier "-like" or
designation unknown where appropriate. Working together we
can achieve a better understanding of Florida's (and other
state's) prehistoric and early historic native peoples.


This paper benefitted from the comments of several
reviewers. Of particular note are Brent Weisman, George
Ballo, Mike Wisenbaker, and Marion Smith. Roy Lett of the
Bureau of Archaeological Research prepared the two
photographs used in Figure 9. The author prepared the
illustrations used in Figures 1-8.


References Cited

Daniel, I. Randolph, Jr., Michael Wisenbaker and George
1986 The Organization of a Suwannee Technology: The
View from Harney Flats. The Florida Anthropologist

Tesar, Louis D.
1994 Johnson Sand Pit (8LE73): An Analysis and
Comparative Review of a Paleoindian through Early
Deptford Base Camp in Leon County, Florida. Florida
Archaeological Reports 32, Tallahassee.

Louis D. Tesar
Bureau of Archaeological Research
Division of Historical Resources
500 South Bronough Street
Tallahassee, Florida 32399-0250

... yes, but what was the

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

Back issues of The Floyida Anthivpologist -- 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



Louis D. Tesar

Distinctive ground stone objects, commonly referred to as
clubheads, bola stones, eggstones, or dimple stones, have been
found occasionally in Florida, south Georgia, and elsewhere in
the Southeastern United States. The distinguishing attribute of
these groundstone objects is the concave depression--the
dimple--found on one end of the generally egg-shaped, ovoid-
shaped or teardrop-shaped artifacts. The function of these
objects has been the subject of debate for nearly 50 years. Are
they bola stones or clubheads, or did they serve some other
This article provides a very brief review of the literature
on Florida and south Georgia specimens, and describes three
specimens from three Leon County, Florida sites. At least one
of the Leon County specimens has noticeable dimple area
hafting polish, suggesting it functioned as a club head.
Another is made of soft material similar to pipestone with red
ochre inclusions and would likely have broken upon impact
with a hard surface if used as either a clubhead or bola stone.
The third could have served as a bola weight, and is the
smallest of the three. In addition to describing the three Leon
County specimens, the purpose of this article is to stimulate
further research on this distinctive class of artifacts and their

Background Information

A very brief review of available information is offered to
provide the reader with a sense of what is generally known and
believed concerning these artifacts. A more detailed review
would be warranted in a more comprehensive study.
Simpson (1948:14) was among the first to describe this
class of artifacts:
Another object which occurs at the Folsom-like localities
in Florida, and is not known from established later horizons, is
an unusual type of clubhead. These clubheads, of sandstone,
limestone, or quartz, are about the size and shape of a hen's
egg, and have a shallow indentation in the smaller end,
presumably for fitting a handle.
As Neill (1971:61) notes, "Simpson's term, 'clubhead',
may conveniently be retained for these artifacts although their
function is uncertain. No one has specifically refuted his
contention that the clubheads are of Suwannee, and, therefore,

of Florida Paleo-Indian age." He further observes, "it is
significant that the clubheads have not been reported in Florida
from any of the numerous excavated sites that post-date
Suwannee points" (Neill 1971:62).
Neill (1971:63) describes a clubhead from the Cavern
site, located beneath the waters of Silver Springs in Marion
County, Florida. Neill's specimen is among the largest
reported, measuring about 125 mm in length (about the size of
a small grapefruit) and "made of a coarsely crystalline
material, apparently a silicate with much calcite impregnation"
(Neill 1971:63). Furthermore, in an Editor's Note at the end
of Neill's article, Bullen (1971:70) writes:
For another example of a Florida clubhead or bola stone
see Waller Figure 2 last item, on page 39 of The Florida
Anthropologist, vol. 20 (sic [22]), nos. 1-4, 1969. Neill's
article, published above, was written before Waller's. Readers
will also be interested to note that smaller (about 3 in. [75
mm]) but otherwise duplicate implements have been found in
Massachusetts surface collections and several examples are in
the Simpson Collection at the Florida State Museum [since
renamed the Florida Museum of Natural History] while over
30 others have been found in the Santa Fe-Suwannee rivers in
recent years. The ones referred to by Neill on p. 61 are those
in the Simpson Collection.
Bullen seems to have introduced the alternative
interpretation that the subject ground stone objects may have
functioned as bolaa stones." Since then, they have almost
exclusively been termed bola stones, although the functional
issue is again raised in this article. Purdy observes that
"although Simpson believes they were clubheads, they closely
resemble specimens that are tied with thongs, knotted at the
indentation, and thrown to ensnare the legs of running animals
being hunted by South American Indians" (1981:30).
In an unpublished manuscript, Philip Gerrell (n.d.)
analyzed and provided measurement data on 59 specimens of
Florida bolaa stones," although he did not list the site
provenience for artifacts, coding them instead by collectors.
Undoubtedly many of these are among the more than 30
Florida specimens noted by Bullen (1971:70). It is hoped that
Gerrell will eventually complete and publish the results of his
study; the use of the draft for citation in this article is



Vol. 47 No. 3


While most specimens seen by the author have rounded or
flattened-rounded distal ends (the end opposite the dimple), in
Gerrell's larger sample only 37 percent fit this form, while 42
percent had distal projections and the remainder had other
forms (Gerrell n.d.:12). Gerrell (n.d.:3) reports that 54 (93
percent) of the specimens that he studied came from present-
day riverine systems, while the remaining five came from
upland locations near water sources. Forty two (72 percent)
came from the Suwannee, Santa Fe and Itchetucknee Rivers
(Gerrell n.d.:3-4), which have also produced the most
Paleoindian projectile point or knife find sites in Florida (see
Dunbar 1991; Dunbar and Waller 1983; Waller and Dunbar
1977), although many of them remain unrecorded (May 31,
1994, review of Florida Site File data) and a distinct search
and discovery bias is noted--disproportionate collecting efforts
in karst river beds versus other settings.
In south Georgia, ground-stone objects similar to those in
Florida are called "eggstones." The Georgia specimens "have
also been recovered adjacent to present day river or pond type
environments. Cultural materials surface collected at these
sites indicated an assemblage of Early through Late Archaic
artifacts (Snow 1976, Whatley 1986)" (Gerrell n.d.:4). One
such "eggstone," found in Early County, Georgia, is described
by Ron Hunt (1975:72) as being made of white quartzite
ground to a very smooth surface and measuring 57 mm long,
48 mm wide, and 15 mm across the concavity. It was surface
collected from a context that Mr. Hunt believed to be Early
Archaic. The site also yielded, "along with archaic stemmed
points, one perfect Dalton point" (Hunt 1975:72). B. Calvin
Jones (1994, personal communication) reports that some years
ago Ron Hunt showed him three "beautifully polished quartzite
'dimple stones' found at Bolen period upland sites along
Spring Creek in southwestern Georgia," one of which may be
the object depicted in Hunt's (1975) report. Further, Ripley
Bullen reported on the occurrence of "three Suwannee-like
points from the Blakely region of Georgia" (1975:52).
All of the specimens in the Florida sample evaluated by
Gerrell (n.d.) are reportedly made from naturally occurring
Florida rock: hematite (6.8 percent), limestone (17 percent),
quartzite (20.4 percent), sandstone (40.7 percent) and
Suwannee River chert (8.4 percent); 7 percent are of
undetermined origin. Purdy (1981:30) reports that of the
"more than one hundred examples, many recovered from the
Santa Fe River," seen by her many were made of non-Florida
stone. Waller (1983:34) reports the finding one made of
"carved manatee bone."
In his discussion of Paleoindian tool kits, Milanich
(1994:51) writes:
Another item in the tool kit is oval ground stone
weights the shape and size of eggs with one end
flattened (Neill 1971; Purdy 1981:30). Most likely
these are bolas; the stone weights attached by thongs
and thrown to bring down water birds and other game.

One such stone was recovered from the Page/Ladson
site from a stratum dated 12,330 years ago (Dunbar,
Webb, and Cring 1989:479).
Dunbar, Webb, and Cring (1989:484) write: "as far as
we are aware, this is the first 'bolo (sic) stone' to be recovered
in direct association with Pleistocene megafauna." However,
Dunbar (1994, personal communication) indicates that the
specimen was actually found near the top of marls within
which Pleistocene remains are also found and not amongst or
near those remains. Nonetheless, the specimen is capped by a
10,000 year old stratigraphic level.
While this literature review is not comprehensive, the
context of the discoveries suggests that these ground stone,
dimpled, egg-shaped objects appear to have originated in the
Paleoindian, or at least Late Paleoindian/Early Archaic period,
although their function remains an enigma. All but one have
been found in disturbed settings, primarily in eroded river
locations. Those made of softer stone often are weathered such
that use wear analysis becomes impossible, or at least very
difficult, to conduct. The term dimple stone is used here
because of its non-functional implications.

The Site 8LE148 Dimple Stone

Site 8LE148 is a major Late Paleoindian/Early Archaic
through Ft. Walton village-ceremonial-mortuary site located in
an area stretching from the north shore of Lake Lafayette
northward to the upland ridges. B. Calvin Jones recently
conducted emergency archaeological salvage excavation of a
portion of the site in advance of school construction activities.
Although they were expected to be present, no Paleoindian or
Early Archaic artifacts were reported from the site (Tesar
1994:113) prior to Jones' work. In addition to more than 80
pit features and a single structure associated with the Swift
Creek component, the excavation project has produced Late
Paleoindian and Early Archaic materials, including the dimple
stone described below. Unfortunately the dimple stone was a
surface find in the exposed plow zone cleared to expedite
salvage of intact features.
The site 8LE148 dimple stone (BAR accession number
94A-68) is made of a locally available brown colored (reddish-
brown colored where the patinated surface is scraped) hematitic
sandstone with a thin band of limonite (yellow ochre), eroded
or scraped out on one side. The surface of the object is
irregular, owing to the general lack of smoothing of the pecked
manufacture marks (Figure 1). It is somewhat egg-shaped,
measuring 50 mm long from the dimple base to its rounded
end. The maximum width of the object is 32.5-36 mm,
occurring 33 mm from its apex. The variation in width is the
result of its more flattened sides conforming to its sedimentary
grain, thus giving it an ovoid cross-section. The width across
the dimple is 10 mm and the depth of the dimple is very
shallow, less than 0.5 mm. The specimen weighs 79.3 grams


and displaces a volume of 70 ml.
This specimen could have functioned as either a small
clubhead (although inspection of the dimple area failed to
detect any hafting polish on the weathered, pecked concave
surface [Figure 2]) or as a bola stone. Of the three Leon
County specimens described in this article, this one is the best
candidate to have functioned as a bola stone, since the dimple
is too shallow to have been easily glued to a handle to prevent
movement, and the leather covering unlikely would have been
sufficient to have prevented movement resulting in use wear
polish if it had functioned as a clubhead. Furthermore, its
unfinished (pecked not polished smooth) quality suggests that
it was not likely used for ceremonial or other non-utilitarian
purposes, since such items are generally well made and

The Site 8LE651 Dimple Stone

Site 8LE651 is located on the end of a ridge projection
overlooking Long Pond in Section 35, T2N-R1E, northeast of
Tallahassee in Leon County, Florida. The site was recorded
by B. Calvin Jones and Frederick Gaske (8/25/82 Florida Site
File form). Jones and Gaske identified it as a more than one
acre, severely disturbed Fort Walton residential midden based
on the surface collection of Lake Jackson Plain and Incised
sherds in the area following road construction activities
heralding a housing development project. Several weeks later,
an unidentified City of Tallahassee employee brought the
ground stone object described below to the Florida Bureau of
Archaeological Research for identification, and following
conversations with B. Calvin Jones left the specimen for
inclusion in BAR's artifact collections (Jones 1994, personal
communication). The object had been recovered during
monitoring of the road construction, which occurred prior to
the Jones and Gaske visit. Through interviewing the donor,
Jones determined that the object "flipped up out of the sandy
red clay more than a foot below ground surface when
encountered by the road grader blade" (a recent nick mark is
present on the specimen's side) and that nothing else was
observed at the time of the isolated discovery; from which
information Jones concluded that it predated the Fort Walton
component which was restricted to disturbed surface levels
(1994, personal communication).
The ground stone object from site 8LE651 is egg-shaped
with a pecked depression at the smaller end (Figure 3) and is
located in the Bureau of Archaeological Research type
collections (accession number 94-87). The surface color is a
dark yellowish brown (becoming gradually darker toward the
dimple) with a slightly reddish brown heat induced
discoloration--likely the result of the friction heat generated
during the manufacture grinding and polishing process in
reaction with the specimen's high iron content. The object
weighs 268.5 grams and displaces a volume of 80 ml. Close
examination of the artifact's surface reveals sedimentary

banding in a very fine grained quartzitic material. The stone
used in the object's manufacture appears to have originated as
a hematitic siltstone cobble, of a type that may be found in the
upper Apalachicola-Chattahoochee-Flint River drainage.
Alternatively, it may have originated in a quarry, as B. Calvin
Jones (1994, personal communication) suggests that the artifact
is "composed of good quality hematite, probably from the
Troy, Alabama area, where iron is mined today--which is
probably the nearest good source for this material."
From the dimpled base to the slightly flattened, rounded
end measures 54 mm, while the maximum width measures 52
mm about 30 mm from the dimpled base. The edges of the
dimpled base are rounded, measuring 23.5 mm across the
outside and 16mm across the inside, and the depression
measures approximately 2 mm to the central point of the
concavity. The depression retains the pecked impact points
remaining from its manufacture and clearly shows hafting
polish, although the polished surface appears only as a
smoothing of the raised portions of the manufacture pecks in
Figure 4 and may be seen as a smoothed patch in the lower
right dimple area in Figure 5. Furthermore, the dimpled area
has what appears to be a dark stain--a dark brownish black
with a slightly silverish sheen on the hafting polish surfaces,
although it might simply be the result of the parent material
grading into a band of naturally occurring darker material.
The exterior surface of the specimen beyond the dimple retains
very faint evidence of the pecking used to shape the object, as
well as striations at random directions from the grinding and
polishing used to attain its final form and finish (Figure 5).
It is noted that the length (54 mm), width (52 mm) and
dimple width (16 mm) of the 8LE651 specimen is close to the
measurements described for the specimen reported by Hunt
(1975): 57 mm long, 48 mm wide, and 15 mm across the
dimple, although the photograph in the Hunt (1975:72) article
is slightly out of focus such that an assessment of hafting
polish cannot be made and also gives his specimen a more
rounded appearance than the 8LE651 specimen. The site
8LE651 dimple stone appears to have functioned as a

The Site 8LE1949 Dimple Stone

Site 8LE1949 is located on the west side of Lake Jackson
in northern Leon County, extending from the hill top to the
lower slope near the lake edge. It was recorded by B. Calvin
Jones (03/12/92 Florida Site File form), who identified it as a
"very large (25+ acre) intact Fort Walton village; few, if any,
sites of this size [are] left in [the] area," and it was
recommended that "further evaluation and salvage [excavation
occur] before development." Unfortunately, development
began without any professional archaeological investigation,
and quantities of ceramics, Middle Archaic and earlier lithics,
as well as the dimple stone object described below, reportedly
have been found by construction workers and others.

Figure 1. Side views of the dimple stone found at site
8LE148 in Leon County, Florida. (Photographs
prepared by Roy Lett in the Bureau of Archaeological
Research, Division of Historical Resources.)

Figure 2. Dimple end view of the dimple stone found
at site 8LE148 in Leon County, Florida. (Photograph
prepared by Roy Lett in the Bureau of Archaeological
Research, Division of Historical Resources.)

Figure 3. Side and oblique views of the dimple stone found at site 8LE651 in Leon County, Florida.
(Photographs prepared by Roy Lett in the Bureau of Archaeological Research, Division of Historical



Figure 4. Closeup view of usewear on the dimpled end of the dimple stone found at site 8LE651 in Leon
County, Florida. (Photograph prepared by Roy Lett in the Bureau of Archaeological Research, Division
of Historical Resources.)

Figure 5. Closeup view of manufacture grinding striations and evidence of manufacture pecking on the
dimple stone found at site 8LE651 in Leon County, Florida. (Photograph prepared by Roy Lett in the
Bureau of Archaeological Research, Division of Historical Resources.)

The site 8LE1949 dimple stone is unusual in shape and in
the material from which it is made. It was found by Ted
Avellone (1994, personal communication) in the bottom (about
one meter deep) of a stormwater culvert trench cut, which had
been filled in when Mr. Avellone returned to the site. There
were no other artifacts with the dimple stone, and while found
in the trench bottom, Mr. Avellone was uncertain as to
whether it was found in place or had been displaced from
shallower levels of the trench sides. When first found,
conditions were so moist, that the red ochre inclusions in the
material from which the object is made easily marked the hand
of their discoverer--who at first thought that it might be a rusty
metal object, such as a gear shift knob or joint, until the soil
was cleaned from the dimple area revealing that it was not a
drilled hole and that the object was not metal. The stone,
probably a low grade of hematite, is of the hardness of steatite
or pipestone and has a very silty or clayey texture. The shape
is somewhat like that of an apple, being larger toward the
dimple end, rather than toward the apex as in most such
objects (Figure 6). The object appears to have been
manufactured through grinding only, as pecking was not
required because of the softness of the material and likely
would have caused stress fractures and manufacture breakage.
The sides are irregularly smoothed, while the dimple is
exceptionally well-polished and deeper than the norm. There
is, however, surface leaching of material causing some
cracking, which may also be the result of water absorption
expansion and evaporation shrinkage cycles. The specimen
measures 44 mm long, from 38-39 mm wide at its widest part
(being slightly ovoid in cross-section), 16.5 mm across the
dimple, and the dimple measures 5 mm deep. The author was
called to a meeting and Mr. Avellone had to leave before
weight or volumetric measurements could be made.
The material from which the object is made appears such
that it likely would break under the types of impact expected
for either clubheads or bola stones. The highly polished
dimple surface suggests that it might be the result of use, such
as a socketed knob used to apply pressure to the shaft of a
bow-drill; however, if so the head of the shaft was so smooth
that it left no circular striations from rotation activities.
Alternatively, the object may have been mounted on a handle
with an adhesive, such as pine sap, for some non-utilitarian
purpose similar to that of a septre. The function of this object
remains an enigma.

Functional Attributes

Interpretation of the possible function of an object is
guided by its form, the material from which it is manufactured,
wear and breakage patterns, and the context in which it is
found. The quality of the finished product depends on the
limitations of the material being worked and the ability and
intent of the craftsperson manufacturing the object, as well as

the circumstances under which the item is being manufactured-
-all of which likely would involve speculative interpretations,
at best.
The form of the dimple stones is generally egg-shaped,
ovoid-shaped or teardrop-shaped, although other forms have
been observed. The dimple is the diagnostic attribute
separating these objects from those with circumferencial
grooves around the near base area or perforations through the
basal area. The latter two categories of objects, generally
smaller than dimple stones, are usually classified as pendants
and generally found in Deptford or later settings (ca. A.D. 1-
Since any rounded stone could be wrapped in leather to
serve as a bolaa stone," why would someone go to the effort
required to make the pecked, ground and polished artifacts
discussed in this article? Also, if they functioned as bola
stones, what is the purpose of the cup-shaped basal depression?
While Purdy notes that South American specimens have a
depression at the end where the thong is tied, she only cites
grooved specimens from Paleoindian Southwestern United
States contexts (Purdy 1981:30). Gerrell (n.d.:ll) reports that
his analysis of the Florida specimens "did not indicate any use
wear or hafting polish in the area of the dimple" (the
depression on the small end), such as would be expected on at
least some specimens if they served as clubheads, unless they
were firmly secured to prevent movement. It is possible that
the erosive weathering conditions associated with the inundated
settings, from which 93 percent of the specimens inspected by
Gerrell came, removed pertinent evidence of hafting polish.
However, the 8LE651 specimen does evidence hafting polish
and possible staining in the hafting area possibly attributable to
application of an adhesive (if the discoloration is not a
naturally occurring darker band in the stone). Furthermore, the
8LE1949 dimple stone does not appear to have functioned as
either a clubhead or a bola stone. Finally, the size of the
8LE651 specimens and those of similar dimensions, not to
mention the huge (grapefruit-sized) one found at Silver
Springs, suggests that they were too large to have served as
bola stones.
Dimple stones are made from a wide range of materials
and these may be an indication of intended function; however,
the author currently has no data with which to determine
whether such is the case. Furthermore, there is little data on
breakage or damage to reported specimens. Such information
would also indicate function.
The context in which dimple stones have been found has
consistently been as isolated finds in settings suggesting
Paleoindian or Early Archaic origins. While bolas generally
have weights at the end of at least three connected cords or
thongs, breakage and loss would likely involve individual
weights, and only rarely the entire compound artifact. Thus,
the individual occurrence of dimple stones does not favor their
interpretation as clubheads over that of bola stones. It will be


Figure 6. Side and oblique views of the dimple stone found at site 8LE1949 in Leon County, Florida.
(Photographs prepared by Roy Lett in the Bureau of Archaeological Research, Division of Historical

Figure 7. Example sketch of dimple stone using method described in text.


through the excavation of an accidentally burned dwelling in
which personal items have been abandoned in place or through
burial contexts that an accurate interpretation may be made.
With the exception of Gerrell's'preliminary effort, little
comparative data has been collected on dimple stones. Such
data needs to be assembled before questions concerning the
function or functions of these objects can be furthered.


Obviously, the function of these artifacts remains an
issue, in spite of the functions implied by the names
"clubhead" and bolaa stone." The term "dimple stone" has
been used above because it lacks functional implications.
Gerrell (n.d.:15) sums up the interpretive problem this way:
Do these stones represent a distinctive hunting
implement or are they thrown by hand or twirled about
the head? Were they hafted on a stick or encased in
leather? Do they represent an artifact used for ascetic
(sic, [aesthetic]) values rather than as a procurement
means? Or do they represent a presently unknown
usage that has no archaeological or historical
While clubs are widely reported among Native Americans
in the Southeastern United States, bolas are not. Indeed, while
bolas may have been used, the author found no reference to
such use. It might be argued that bolas are most effectively
used in an open savanna setting, such as that generally
envisioned for the region during the late Pleistocene, and that
they were gradually abandoned as woodlands emerged as a
result of climatic changes and other associated factors.
However, the settings in which most of these distinctive
ground stone objects generally are found would have been
wooded or at a minimum dominated by at least a fringe of
dense brush owing to their association with surface waters.
Indeed, the setting favors the use of clubs over bolas and the
object's form, as Simpson (1948) early noted, suggests it
functioned as a clubhead or at least was fashioned for
attachment at the end of a rigid handle. However, it is again
noted that survey efforts have been disproportunately
concentrated in entrenched karst drainages, thus biasing known
distribution patterns for the dimple stones. The three Leon
County specimens reported here and the three found by Hunt
in south Georgia plowed fields demonstrate they occur in other
Nonetheless, while some specimens appear likely to have
functioned as bola stones and others appear more likely to have
served as clubheads, at least some may have had another
function--such as ceremonial objects or symbols of rank or
authority. Perhaps the silted environment of Florida's
sinkholes or peaty muck ponds will someday yield a specimen
preserved together with its normally deteriorated organic

components. Or, perhaps specimens will be found in mortuary
contexts or in some other setting, such as an accidentally
burned structure, where lost items were abandoned in place.
However, while a cluster of three such stones together in a
mortuary context might suggest a bola, it would take more
than one such cluster to be certain. Further, a single stone
might suggest a clubhead or simply be a single stone found and
retained for some non-utilitarian personal purpose. Further
research on dimple stones is clearly needed.


The author would appreciate obtaining information on
dimple stones from the Florida, Georgia and Alabama area.
Please send photographs and other information, including
sketches with key measurements, weight, type of material, any
wear or breakage noted, the location of the find (marked on a
map), where and how found, whether a surface find or
recovered from some depth below surface (how deep?),
associated artifacts, and so forth.
If you do not have a sliding caliper, measurements may
be made by using smooth-sided wood blocks with right angle
(squared) edges. Place one block on a sheet of paper and draw
a line along its edge. Next place the dimpled end flat against
the block with a notecard between it and the block, then mark
the outer edges of the stone's base where it touches the paper
to record its width; transfer those marks to the first marked
line. Remove the paper or note card and place a second block
at the end away from the dimple (the apex) and parallel to the
first block, then draw a line along the inside edge of the block
to obtain a measure of the object's length; mark where the apex
of the specimen touches the block. Next, being careful not to
move the object, move the two blocks to place them along the
sides of the specimen and at right angles (perpendicular) to the
first lines. Then draw lines along the inside edges of the two
blocks to obtain a measure of the specimen's maximun width;
also mark on the lines the points at which the specimen's sides
touch the blocks. From these marks, being careful not to
obscure them, complete a sketch of the side view of the
specimen. Next place a square wooden matchstick across the
dimple so that one edge crosses the center of the dimpled area.
Mark where the dimple edges touch the match to obtain the
inside width of the dimple. Then take another match, place its
end in the center of the depression, holding it upright to mark
where it touches the bottom edge of the match lying across the
dimpled area. This will provide a measure of the depth of the
dimple. Transfer both measurements to the sketch paper (see
Figure 7). The author will measure the various marks to
obtain the comparative measurements to be used in the study.
Make any other notes and observations on the sketch sheet and
send it along with any photographs, maps, and other material
to the address listed at the end of this article.


References Cited

Bullen, Ripley P.
1971 Editor's Note [at end of Wilfred T. Neill's article, "A
Florida Paleo-Indian Implement of Ground Stone"]. The
Florida Anthropologist 24:70.

1975 Suwannee-like Points from Southwestern Georgia.
The Florida Anthropologist 28:52.

Dunbar, James S.
1991 The Resource Orientation of Clovis and Suwannee
Age Paleoindian Sites in Florida. In Clovis: Origins and
Adaptations, edited by Rob Bonnichsen and Karen
Turmeyer. Center for the Study of the First Americans,
Oregon State University.

Dunbar, James S., and Ben Waller
1983 A Distribution Analysis of the Clovis/Suwannee
Paleoindian Sites of Florida -- A Geographic Approach.
The Florida Anthropologist 36:18-30.

Dunbar, James S., S. David Webb and Dan Cring
1989 Culturally and Naturally Modified Bones from a
Paleoindian Site in the Aucilla River, North Florida. In
Bone Modification, edited by Rob Bonnichsen and M.
Sorge, pp. 473-497. Center for the First Americans,
University of Maine, Orono.

Gerrell, Philip R.
n.d. Untitled manuscript report on Florida bolaa stones."
Copy on file with author courtesy of Gerrell.

Hunt, Ron
1975 Problematical Stone Find. The Florida
Anthropologist 28:72.

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

Neill, Wilfred T.
1971 A Florida Paleo-Indian Implement of Ground Stone.
The Florida Anthropologist 24:61-70.

Purdy, Barbara A.
1981 Florida's Prehistoric Stone Technology. University
Presses of Florida, Gainesville.

Simpson, J. Clarence
1948 Folsom-like Points from Florida. The Florida
Anthropologist 1:11-15.

Tesar, Louis D.
1994 Johnson Sand Pit (8LE73): An Analysis and
Comparative Review of a Paleoindian through Early
Deptford Base Camp in Leon County, Florida. Division
of Historical Resources, Bureau of Archaeological
Research Florida Archaeological Reports 32, Tallahassee.

Waller, Ben I.
1983 Florida Anthropologist Interview With Ben Waller.
The Florida Anthropologist 36:31-39.

Waller, Ben I. and James Dunbar
1977 Distribution of Paleo-Indian Projectiles in Florida.
The Florida Anthropologist 30:79-80.

Louis D. Tesar
Bureau of Archaeological Research
Division of Historical Resources
R.A. Gray Building, Room 312
500 South Bronough Street
Tallahassee, Florida 32399-0250
FAX (904) 488-3353



James Dunbar

At 79 years of age, Hubert L. Chason, Sr. passed away in
mid-January, 1994 and will be missed by the people who were
honored to have known him. Hub was a self-made man, a
collector of artifacts, and the author of the book, Treasures of
the Chipola River Valley (1987). Because he was locally
recognized as an authority, Hub (Figure 1) was once shown an
Indian mound on an old man's property and asked to dig it.
Hub refused saying that he did not think it was proper. In the
forward to his book he wrote, "I am not an archaeologist. I'm
not even sure I qualify as an amateur archaeologist, though I
have in the last twenty years spent several thousand hours of
my leisure time on the bottom of Florida's Chipola River
searching for artifacts, or, after the dive, cleaning, mounting
and labeling the best of my finds, while always studying books
in an effort to acquaint myself better with these treasures."
Hub characterized his purpose for publishing as a cautious but
common sense attempt to smoke out the thousands of
interesting stories being harbored by his fellow collectors, to
further persuade archaeologists to take advantage of the large
number of extensive river collections before they become
widely dispersed, and to find a new point type or subtype to be
named for the Chipola River.
In this article Hub will be remembered with some of his
own words both written and recorded. A recorded interview
took place on December 14, 1993 (videotape on file, Bureau of
Archaeological Research, Tallahassee) and on other visits a
quick note or two was jotted down on paper. Hub also
provided information in other interviews conducted in 1985
and 1986 and in a number of unpublished papers he liked to
share with good company. And, of course, there is Hub's
1987 book. Hub's son, Hub Jr., his daughter, Lisa and his
wife Lillian are all to be thanked for their hospitality, support,
and help in putting this article together.

Historic Perspective of the Chipola River Valley

In 1827 John Lee Willams described the Chipola (in his
spelling, Chapola) River in his book, A View of West Florida:

The Chapola is a western branch of the
Appalachicola. It rises in several very large springs, on
both sides of the north line of the territory, in Jackson
county; after running twenty miles, and receiving

considerable accessions from both sides, it divides, and
both branches sink into the earth; the eastern branch
continues under ground several hundred yards; the
western branch but a few rods: these streams unite again
about half a mile below. To this place the Chapola is
navigable. About half a mile west of the natural bridge,
a large stream bursts from the base of a gentle hill, and
joins the Chapola a mile below; it is large enough for
boat navigation. Five miles below the natural bridge,
the Big spring of the Chapola rises, three miles east of
the river; this spring is, also, navigable to its source.
Thirty miles below this spring, an arm of the
Appalachicola has lately burst into the Chapola, and
formed a lake [now known as the Dead Lakes] twenty
miles in length, and seven wide, in which the forest are
yet standing. This river enters the Appalachicola nine
miles above Colinton, or Fort Gadsden. The
adjoining lands are among the best in the territory, and
generally settled with able planters.

It was in this same Chipola Valley that Hub was raised,
spent most of his days, explored the beauty of the unspoiled
land, and discovered many artifacts left by prehistoric peoples.

Hub Chason--The Man

Hub Chason lived in a small community near the Chipola
river in the panhandle of Florida. Unlike many areas which
have been urbanized by the state's rapid population growth, the
Chipola River courses through farming country. Hub was
born into this setting in 1914 and developed into an athlete
who eventually played on the University of Florida basketball
team. In November 1941 Hub traveled to Alabama to meet the
famous boxer Gene Tunney, who was recruiting people to
become athletic (fitness) trainers for the Navy. Hub was
selected by Tunney but was told the wait might be many
months due to a long line of men from around the country who
had enlisted before him. However, after the December 7th
Japanese attack on Pearl Harbor the call to serve his country
came quickly. Hub became an athletic trainer and eventually
served on board a navy aircraft carrier in the Pacific theater.
At places like Okinawa the Japanese Kamikaze attacks made
things interesting. Hub says a close call came when one of the


Vol. 47 No. 3



Figure 1. Hub Chason, Sr. (courtesy of the Chason family)

planes slammed into his ship, seriously denting the hull,
without exploding. After the war Hub returned home and a
couple of decades passed before he began diving and finding
artifacts in the Chipola river in 1967.
The following are Hub's words about his experiences and
time spent exploring the fields, river banks and channel of the
Chipola River (Figure 2). It is also an expression of his
feeling about the natural environment and his hobby of surface

collecting arrow and spear points. Permission to quote this
lengthy passage from an unpublished manuscript of Hub's was
graciously granted by his family.

Why I Collect Artifacts

I collect artifacts because the hunt takes me
outdoors. There I can watch nature's show put on by


Figure 2. The Chipola River. (courtesy of the Chason family)

the clouds in the sky or by nature's wild creatures while
I search for an artifact I will treasure. An artifact
created by an ancient artist who worked hard enough to
hone his skills to the point that he could weave such
beauty into a piece of stone. His best work carries this
message down through the ages the love of
craftsmanship and beauty is not limited to the age of
modern man.
I like to start before sunrise or continue until after
sunset so that I can watch the clouds change colors. As
I walk through the forest and see a rabbit rush through
the trees, I freeze hoping to see the bobcat, fox, hawk,
or owl that may be in hot pursuit. If I hear two nesting
jays screaming indignation and rage at some invader, I
hurry to the root of the nest tree to see if an oak snake
or hawk is threatening the eggs or baby birds.
When I search a ravine I like to watch for such
things as a recently molted coral snake who seems to
flaunt his brilliant rings as he slowly burrows under a
pile of leaves. When I crawl through thick brush, I
sometimes stop in the hope of seeing something like the

deep red of a cardinal who fails to see me before
pitching briefly on a twig just three feet away. When I
search the river banks I watch for the tiny flower that
may be so well hidden in a small crevice that I can walk
away knowing that mine may be the only human eyes to
have seen that small but wonderful treasure.
When I hunt plowed fields I like to become a boy
again by taking off my shoes so the soft earth squeezes
up between my toes as I tune my legs for a very long
walk. However, my favorite area to hunt it is on a
section of river bottom when the water is clear. There I
must be alert for the sharp rocks, edged by the river, so
they do not rip my arms or legs as I search through a
swift water shoal. When I meet a bull gator or large
cottonmouth while [hand-] fanning the bottom of a deep
eddy, my right arm goes to the handle of my knife. As
I continue fanning it is a pleasure to recognize that both
know and usually obey the basic law of the river bottom
- whoever is bigger has the right-of-way. Both
recognize me as a strange looking fellow creature and
seldom show the slightest fear or animosity.

I~i a~9


When I am fortunate to find a well-made spear
point in the river, I always pause to examine, meditate,
and wonder if the spirit of the river truly has
appreciation for beauty. The river uses its swift water
to smooth the flake scars as it slowly applies its
minerals to the surface and forms a patina emphasizing
the skill of the knapper. The artifact has been
transformed by this river bottom process into a glossy
specimen of beauty a treasure of the Chipola. All
these great natural and manmade things are legitimate,
life-enriching rewards for those of us who love the
experience and the worthwhile time spent. It is to
honor the talents of the unknown artists of the distant
past that we treasure these beautiful stone artifacts.
Unlike most divers, Hub's experience is limited
geographically to the Chipola River (see Figure 2) between
Look and Tremble Shoals and the Magnolia Bridge. Hub
found no need to venture outside the Chipola Valley in order
to dive other limestone rivers in Florida or south Georgia.
From a generalized perspective, the Chipola is similar to other
limestone karstt) rivers because it is shaped by acid-charged
water slowly dissolving its exposed limestone channel.
However, a closer look at the local geologic setting reveals an
ever-changing, sculpted, rock channel of distinctive character.
Hub's experiences made him a strong believer in the
importance of water safety. Most swift-water rivers in Florida
still have an occasional cottonmouth moccasin or large bull
gator that can pose danger if encountered under some
circumstances. But by far the greatest hazard for any swimmer
or diver is panic. Once a swimmer panics, he or she too often
turns into the current which can soon exhaust strength and
magnify other problems as long as the swimmer attempts to
fight flowing water, one of nature's more powerful forces.
The current can just as easily sweep the swimmer downstream
to shallow water and safety. For divers and swimmers,
sometimes there is important meaning to that old adage go with
the flow! To illustrate the danger of panic, Hub offered the
following example, recounted here from the 1993 interview.

I once observed a large bull alligator panic for a
few seconds, and though no serious harm came to him
because he was in his own domain, he gave a vivid
demonstration of what could happen to any swimmer
who gives way to panic while in the water. I was going
up the Chipola on a windy November afternoon just as
a cold front was moving in. I was on a small john boat
propelled by a motor. Just as I started to go around a
large willow tree partially submerged near the bank the
gator surfaced slightly to my right about fifteen feet
ahead of the boat. He had apparently been after some
prey under the willow and had not heard the motor or
seen the boat until he surfaced. He was obviously one

of the older gators who had, at some time in the past,
had an encounter with a boat propeller, or had been shot
at by someone riding in such a boat. He panicked at
first sight of the boat.
I cut the motor instantly to watch the show that he
was to put on at close range. Like a frightened deer
trying to run across the road in front of a speeding car,
he, without submerging, rushed across the path of the
rapidly slowing boat to reach the middle of the river.
Once there, he submerged about three feet and started
swimming directly into the swift current with all the
panicked force of his powerful tail.
Because of his large girth, his almost unbelievable
speed and the force of the opposing current, he was
literally splitting that river apart leaving a trough fully
two feet deep and four feet wide behind him. Even
with his great strength in his panicked state of mind, he
seemed to realize that he was exhausting himself by
swimming against the current. After about fifty feet, he
surfaced again, turned hard left still swimming at a
speed that sent water spraying in front of his nose until
he neared the opposite bank, where he turned hard right
again, stopped briefly with hardly a ripple and was seen
no more. No human swimmer could have better
demonstrated the effects of panic.

Hub started out diving with a mask and flippers. This
equipment, when properly used, can become one of the
world's better physical conditioners. Staying underwater,
holding one's breath long enough to hand-fan a pot hole,
crevice, or spot in a gravel bed in order to see what it holds,
leads to top conditioning of the muscles and breathing process.
This approach seemed to work for Hub until he reached holes
deeper than ten feet deep. It took too much time on the surface
to catch sufficient breath, swim to the bottom and then return
to a particular spot.
Hub's solution to the bottom time problem was the use of
a small air compressor powered by a three horsepower gas
engine which floated on a small platform supported by a large
truck inner tube. This type of diving apparatus is known as an
air buoy or hooka-rig and feeds air to the diver through a hose
fit with a mouth-piece breathing regulator. The air buoy
solved some of the other problems of diving. The hose on
Hub's hooka-rig was only 34 feet long and prevented the diver
from going to depths where decompression might be needed.
The bends potentially could be a problem if the diver did not
decompress properly. When two divers used the hooka-rig, it
caused the divers to have a built-in "buddy system" because the
air hose prevented the divers from being more than 68 feet
apart. Hub's hooka-rig ran four hours before needing to be
refueled, a serious amount of time spent on one dive in the
swift water of the Chipola River.


Chipola River Environment

Hub found the best time to dive the Chipola to be during
the fall months of September, October, and November when
the weather is generally dry, the river low, and the water
transparent blue. Another dry time of the year sometimes
occurs in the late spring after winter cold front activity
subsides and before the thunderstorm activity of summer
begins. Hub has also observed that when a dry spell lasted too
long in duration the resulting exceptional low water stages and
reduced flow cause the water to become murky. Murky water
conditions during a drought are greater up river (towards the
Magnolia Bridge) because there is less spring water feeding the
river's base flow. In contrast, the river becomes muddy,
carrying a suspended sediment load and dark with tannic stain
during times of flooding and high water. Therefore the best
time to dive the Chipola is during dry times, just after the
water clears.
Springs in and along the Chipola's course sustain its base
flow during dry periods because the surface of the limestone
aquifer (the Floridan Aquifer) is elevated above the channel
bottom. Because the spring water discharging into the river is
clear, so is the river when there is no other water sources
feeding the river. There have been a number of occasions
when Hub found hidden springs in the Chipola. Most of the
mid-river springs are nothing more than small seeps and
represent the least obvious source of artesian water discharging
into the channel. Hub found mid-channel springs by first
noticing the water temperature change. Since the average
summertime temperature of the river water is warmer than
water discharging from a spring, Hub simply followed the cold
water to its source. The smaller spring openings are
particularly difficult to spot, but, with patience and by
following the cold water, not hard to find.
During a normal year's transition through the summer
months, green algae begins to show up and becomes most
abundant during the warm months of July through September.
Hub felt the growth of algae in the Chipola may be a relatively
recent occurrence. Some years are worse than others for algae,
a factor which probably varies with water clarity. When the
river is tannic-stained or muddy, sun light is partly blocked
and the algae is not able to grow as quickly as when the water
is clear and the sun light penetrates with more intensity. At its
worst, balls of this algae move with the river's flow like giant
tumbleweeds in a wind storm. If the diver is not paying
attention he can be temporally consumed by its mass of hair-
like strands.
Hub noted that the typical sand and gravel sediment shifts
with the current as part of the continuum of this river's
dynamics. There are many gravel beds (mixed with sand)
scattered along the channel bottom. In the spring of the year
quartz and other pebbles are washed off and show an

ar.-rtment of red, yellow, blue and other colors. In the spring
the bottom looks new, fresh and unpolluted. By late summer
to early fall, during the dry time, the bottom collects a certain
amount of fine silt over the deeper gravel beds. While the
visibility in the water is generally good because the water is
clear, without hand-fanning, the bottom looks soiled and
polluted. During extended droughts silt deposits can become
thick and completely blanket objects laying on the bottom
except in the shoal areas. During one such drought, Hub
found a point because its outline was highlighted by the
covering blanket of silt. Most often, however, one must hand-
fan into the silt and gravel bar sediments.
He believed evidence of the river's ultimate force was
accomplished by two flash floods which occurred since he
started observing the river as a boy. Flash floods are definitely
not common events and can be characterized by a rapid river
rise, severe current, and a massive sediment load suspended in
the water. The sediment load was presumably derived from
both the adjacent land and channel bottom. Hub remembered
that the floodwater subsided within 24 to 48 hours. During the
flood's decline the sediment load dropped out of suspension
and covered the shoal areas to elevations above the level of the
subsided river stage. The sediment deposited on the shoals
formed temporary earthen dams which were then breached by
small rivulets which eventually increased in number and size
until the flood deposits were swept from the shoals.
Other than the action of a flash flood, the shoals in the
Chipola are exposed to the relentless action of water being
forced over their shallow races. Shoals are nevertheless
environmentally interesting and good places to find artifacts.
On the swift-water shoals which have pothole catch areas in the
limerock the movement of sediment and artifacts can be
dramatic and, as Hub found out, disappointing. On a
summer's day, when the water was particularly clear, Hub
spotted a point partially hidden by sediment in such a pothole.
He was not equipped to dive that fateful day and came back
prepared to handle the task the next day. He lowered himself
down the shoal by rope only to find the pothole had been
cleaned of its contents by the current. The artifact was gone!
Yet, with all this evidence of erosion at work on the
shoals, Hub found evidence to the contrary. For example, he
found one particularly large mammoth tooth which once
weighed many pounds, however, the dentine part of the tooth
(which accounts for much of the weight and bonds the enamel
plates together) had rotted away leaving the unattached enamel
plates in anatomical, undisturbed position. Had river current
effected the tooth, it would have been scattered and broken.
Also some of the shallows that are slightly deeper and exposed
to less current than the shoals are covered by aquatic grass.
This vegetation helps hold the sediments resting above bedrock
in place, at least until erosion displaces the grass. The grass-
covered shallows are also good places for aquatic creatures
such as snakes, crawfish and turtles to lurk.

Artifacts and Sediments of the Chipola River

In 1986, Hub told fellow river diver Phil Gerrel about 20
favorite diving sites in the Chipola. Typical locations where
points are found are usually associated with gravel beds located
in spots here and there along the channel bottom. Sometimes
sites are located in straight channel sections, sometimes on the
outside of river bends, and sometimes on or near shoals, but
always where erosion causes sediment movement and exposes
gravel beds in blowouts.
Crevices and potholes in the limestone bottom are good
traps and places to look for points. Points may be found
anywhere in a crevice or pothole including resting on the rock
bottom. There are some depressions where Hub could not
reach the bottom and who knows what they contain. Hub used
size sorting along the river bottom as a general rule of thumb
to find points. Early points and larger points can be found in
larger gravel and smaller points in smaller gravel. This rule
does not always hold true, but it is good to keep in mind
because it has proved to have frequent success.
There are isolated places along the river bottom where
artifact-bearing sediments have become cemented. Hub was
impressed by the formation of sandstone around artifacts.
Such conglomerate deposits can be broken apart with moderate
effort by hand. These hardpan-like deposits can sometimes be
found in isolated areas of the channel bottom, in potholes, and
in crevices. Artifacts that have been found in the conglomerate
range in age from Paleoindian to Archaic. Diagnostic artifacts
include Redstone, other Paleoindian forms, and a variety of
stemmed Archaic points. Just how such sandstone deposits
were formed and under what environmental conditions is a
legitimate research question which is now being investigated.

Late Pleistocene Fossil Bones in the Chipola River

Hub admitted that he maintained a collecting bias when
diving the Chipola River and mainly concentrated on finding
points. He did not look for fossils and, although he
occasionally picked up a stray tooth or bone, he most often
ignored them. He believed his evaluation of the fossil remains
in the Chipola is likewise biased, but offered a few opinions.
Hub remembered seeing mammoth, mastodon, sloth,
camel, and horse fossils in the river. Horse teeth were
apparently the most common fossil seen. He also remembered
seeing frequent amounts of "giblet-size" pieces of broken
mastodon tooth enamel. There was more evidence of
mastodon than mammoth although many of the mammoth teeth
were more or less intact.

Evidence of the Paleoindian Occupation in the Chipola

Hub found the majority of his Paleoindian points and
virtually all of his Simpson points at sites adjacent to or within

50 yards of a tributary entering the river. The tributary near a
Paleoindian site may be small but seems to be an important
factor of the site distribution along the river channel.
Paleoindian artifact concentrations were found in the Chipola,
not in the adjacent tributary.
Hub was from the same school of thought as Ben Waller,
and believes the Chipola was a flowing river during the late
Pleistocene when mammoths, mastodons and other now extinct
ice-age beasts roamed Florida. He did not think the river was
mostly dry with ponds in deeper stretches that served as water
holes and occasional Paleoindian hunting localities. Rather,
he, like Ben, thinks the shallow areas of the rivers provided
game-trail crossings and that flowing water provided a useful
impediment that facilitated Paleoindian hunting.
A variety of Paleoindian point types have come from the
Chipola including the typical Florida forms--Clovis, Suwannee
and Simpson--as well as other less common forms such as
Redstone, Dalton, and Quad. It must be mentioned that
Redstone points are rare in Florida although it is a northern
Alabama type. Perhaps it should not be surprising to find a
occasional Redstone in northwestern Florida because of the
geographic proximity to Alabama.
Three of Hub's large Simpson points were not found in
the river but came from land on an adjacent plowed field. The
so-called Chason Cache found by Hub's cousin about 25 years
ago, deserves special mention because it was an unusual find.
Three very large (6 to 7 inches long) and very wide (2 1/2 to 3
inches) Simpson points were found after a spring rain in a
washout of a freshly plowed field. The specimens, which were
given to Hub, included one complete and two broken points.
Apparently the plowing activity resulted in the breakage of two
of the specimens. The Chason Cache is important in three
First, the cache represents a set of points and/or knives
which had just been completed by the knapper as a finished
product. They display little or no evidence of being
resharpened and, therefore, show the first stage of the finished
product. In other words, the points display the evidence of
how they were produced, not how they were subsequently
reworked for sharpening. They display the traits of a point
type requiring its method of manufacture to be conceived in the
mind of the knapper but brought to fruition only after years of
acquired skill. Most Simpsons display evidence of post-
production resharpening and their width and, to a lesser extent,
their length are reduced in a way which conceals their original
production. Most likely, these new made points were not
random losses or discarded inventory.
Second, this rare glance at a set of new made Simpsons
compares well with the Central and South American type
generically known as fishtail points. In the book Clovis
Origins and Adaptations, Bonnichsen (1991: 316-317)
indicates that the lithic reduction sequence for fishtail points
differs from the other Central America Paleoindian forms--
Clovis and Clovis waisted (a Suwannee-like point). It should

Figure 3. Simpson point from the Chason Cache. Left, computer-scanned image produced by Shawn Kelly; right, line drawing by
Lloyd Schroder (used with permission).


be mentioned that the distribution of classic Clovis points
extends from North America into Central America but not into
South America.
Hub's Simpson points are similar to the Central and
South America fishtail points in that they were produced from
flake blanks only slightly thicker than the finished point.
During the Simpson reduction sequence large thinning flakes
were struck off by impacting carefully prepared platforms
along the lateral sides of the preform. The scars of successful
thinning flake extractions are broader toward their distal ends
and overlapped similar detachment scars extracted from the
opposite blade edge. Upon their completion, both Simpson
and fishtail points are marked by characteristically large and
wide flake scars which more or less meet in the center of the
blade. In cross section both Simpson and fishtail points also
display flattened profiles. In contrast, Clovis points were
produced from thicker flake blanks (or cores) which were
knapped to a much narrower blade width and were bi-convex
to plano-convex in cross section (Bonnichsen 1991). Clovis
points also have large, thinning flake scars but their
detachment from one lateral edge generally runs across the
width of the blade face to the opposite side. Classic Clovis's
are lanceolate, straight sided points. Simpson points are wide,
thin lanceolate points that are so recurvate that their outline
looks like a fish. The hafting area of a Simpson is fishtail-
shaped (Figures 3,4). Therefore in terms of their production
sequence, shape and appearance, the Chason Cache of Simpson
points closely resembling the Paleoindian fishtail points of
Central and South America.
Third, although Paleoindian cache sites are rare, they are
known to be geographically scattered throughout North
America. Perhaps the most famous or, at least, most
publicized cache is from the East Wenatchee site in
Washington State. Similar to the Chason Cache, the East
Wenatchee site had a number of very large points, all Clovis.
The Chason Cache is Florida's only documented Paleoindian
cache site.
The points of the Chason Cache should be considered the
type specimens for the Florida Simpson. Other Simpson points
like the example Hub found at Johnny Boy Landing (Figure 5)
represent reworked specimens with reduced lateral margins.
Reworked points are important for comparative purposes
because they show the resharpening, reduction sequence for the
type. For reasons that will not be explored here, the Florida
fishtailed or Simpson point should not be confused with
snapped-base, reworked Suwannee points or other excurvate
blade Paleoindian types.
Perhaps the Chason Cache was left as an offering,
perhaps it is all that remains of a burial, perhaps the cache was
stored with the intent of eventual retrieval, or for other reasons
known only known to its original owner.


The Chipola River is as individual as this river diver who
first explored it, Hubert Chason, Sr. He came to know the
Chipola River valley like few others. His part of Florida is in

Figure 4. Second Simpson point from the Chason Cache,
as drawn by Lloyd Schroder (used with permission).

them thrilling because they recall similar experiences in their
own searches" (personal communication with J. Dunbar,
January 1994). Hub represents the voice of many collectors:
people who are impassioned by archaeology and who represent
the roots of an endeavor others of us now carry on as a
profession. Hub wanted me to tell you that!

References Cited

Bonnichsen, Robson
1991 Clovis Origins. In Clovis Origins and Adaptations,
edited by R. Bonnichsen and Karen L. Turnmire, pp.
309-329 Center for the Study of the First Americans,
Oregon State University, Corvallis, Oregon

Chason, Hubert. L.
1987 Treasures of the Chipola River Valley, Father & Son
Publishing, Inc., Tallahassee, Florida

Williams, John L.
1827 A View of West Florida. Published for H.S. Tanner
and John Lee Williams by L.B. Bailey, Philadelphia.

James Dunbar
Florida Bureau of Archaeological Research
500 S. Bronough St.
Tallahassee, Florida 32399-0250

Figure 5. Reworked Simpson point from the Chason
collection, from Johnny Boy Landing.

a rural area located away from university settings or easy
access to archaeological information. Nevertheless his passion
for the subject led him to occasionally seek the knowledge of
archaeologists like Ripley Bullen and to explore libraries for
published information. He had an enthusiasm for archaeology
and field experiences both envied and admired by others. In a
time when there seems to be a tendency for amateurs to be
secretive, Hub chose to speak out in a productive way. He did
not agree with the uncontrolled destruction of in situ sites.
However, he opposed any attempt to prevent people from the
loved hobby of surface collecting. Many of us admired Hub
Chason for his truthful yet strong opinions. To quote Hub,
"the search takes one out of today's run-dash-hurry world and
provides some joy that comes from a visit into one of nature's
wonderlands. Artifacts that are found in the wilderness along
the mountain sides and lake shores and on the river bottoms
have been placed there through many thousands of years by
various tribes that had cultures which we have a hard time
relating to. Many collectors have read my anecdotes and find




Randy V. Bellomo

In the March issue of The Florida Anthropologist
(Volume 47, Number 1), Wheeler (1994) included a
photograph and brief description of an engraved bone artifact
which had recently been recovered during archaeological
investigations at the Summer Haven site. Unfortunately,
Wheeler (1994) presented inaccurate information about the
artifact and its context, which I will correct in this brief article.
The engraved bone artifact which Wheeler presented in
his Figure 20 (Wheeler 1994:58), was recovered during the
1992 archaeological excavations at the Summer Haven site
(8SJ46), located in St. Johns County, Florida. The 1992
excavations were conducted by Janus Research, under my
direction, for the Florida Department of Transportation.
Wheeler's Figure 20 was duplicated from a color slide
showing one of three conjoining segments of a much larger
bone artifact. The original artifact appears to have been made
from a long bone of a white-tailed deer (Odocoileus
virginianus). The fragment depicted in Wheeler's Figure 20
was intricately engraved around nearly all of its circumference,
while the other two segments (which attach to the right portion
of the engraved fragment as it is shown in Wheeler's Figure
20) were undecorated. Based on an examination of the
reconstructed artifact, the original engraved design extended
beyond the left edge of the engraved fragment as it is depicted
in Wheeler's Figure 20.
Part of Wheeler's Figure 20 caption states that "In this
specimen the meander almost resembles dermatoglyphic
(fingerprint) patterns (Wheeler 1994:58)." Wheeler's
assessment of the patterning on this artifact was made on the
basis of examining the two-dimensional image of the artifact as
depicted in his Figure 20. If undecorated areas of the design
pattern are excluded, one small portion of the engraved design
depicted in Wheeler's Figure 20 vaguely resembles a whorl or
double loop fingerprint pattern; however, had Wheeler viewed
the actual bone artifact in three-dimensions, I believe he may
not have made this analogy. In three-dimensional context, the
so-called "dermatoglyphic patterns" are lost, since the overall
design consists of a repetitive series of finely-incised
interlocking parallel curvilinear lines (Figure 1). A broadly
similar design motif also appears on engraved bone artifacts

from Tick Island (see Jahn and Bullen 1978, Figures 34f and
Wheeler's short discussion of this engraved bone artifact from
Summer Haven erroneously appeared under the heading Post-
Archaic (2,500 B.P. to European Contact). I am puzzled by
the fact that this artifact was discussed in a section about the
post-Archaic. Although Wheeler did not consult me regarding
the chronological or depositional context of the artifact, he
recognized its affinity with the Orange period since part of his
Figure 20 caption states that "The nested meander of this
artifact is an Orange period motif (Wheeler 1994:58)." Had
Wheeler contacted me, he would have learned that five oyster
shell samples from the Summer Haven site were submitted to
Beta Analytic Inc. in July of 1993 for radiocarbon assay.
Those five samples were collected from spatially distributed
areas throughout the site, including upper and lower midden
layers and sub-midden features. All five oyster shell samples
yielded dates ranging between approximately 3,600+60 years
B.P. and 3,440+70 years B.P. Since the artifact was
recovered from within Orange 2 midden deposits dated to the
Late Archaic period (see Milanich 1994:85-88), it is unclear

Figure 1. Incised motif which appears on the engraved bone
artifact recovered from the Summer Haven site. This drawing
depicts the design in its entirety, showing the extent of
engraving around the circumference of the artifact. When
viewed from this three-dimensional perspective, the overall
design consists of a repetitive series of finely-incised
interlocking parallel curvilinear lines.


Vol. 47 No. 3



why it was included in a discussion about the post-Archaic.
Based on the context and association of the artifact, it is no
wonder that the ". .decorated bone pin fragment from the
Summer Haven site exhibits the unbounded design
characteristics so often found in Orange period carvings
(Wheeler 1994:57)." Wheeler's error in temporal placement
could have been avoided had he taken the interest to contact
The 1992 archaeological investigations at the Summer
Haven site by Janus Research have provided important new
data about early Orange period settlement, subsistence
practices, and lifeways. Once the final report of investigations
is completed, and written permission from the Florida
Department of Transportation is received, I intend to publish a
short article which focuses on the results of the 1992
investigations at the Summer Haven site.

References Cited

Jahn, Otto L., and Ripley P. Bullen
1978 The Tick Island Site, St. Johns River, Florida. Florida
Anthropological Society Publication Number 10.

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

Wheeler, Ryan J.
1994 Early Florida Decorated Bone Artifacts: Style and
Aesthetics From Paleo-Indian Through Archaic. The
Florida Anthropologist 47:47-60.

Randy V. Bellomo
Piper Archaeology/Janus Research
P.O. Box 919
St. Petersburg, Florida 33731



Archaeology Of Precolumbian Florida. Jerald T. Milanich.
(University Press of Florida, Gainesville, 1994. 498 pp. 109 b
& w photos, 32 maps, bib., index. $49.95 Cloth; $24.95

Reviewed by Gordon R. Willey, Harvard University

In the Preface to Archaeology of Precolumbian Florida,
Jerald T. Milanich states that it has been written "to provide an
introduction, albeit a thorough one, to the various
precolumbian groups." Milanich achieves this goal in
admirable fashion, providing a landmark stock-taking and
synthesis of Florida archaeology. It builds upon the earlier
synthesis, Florida Archaeology, which Milanich wrote with the
late Charles H. Fairbanks in 1980; however, this new book
significantly updates and adds to the former one. As its title
indicates, the present book is confined to the Precolumbian, or
the pre-European, past of the state, but it should be noted that
Milanich now has in preparation a book on Florida's Colonial
Period archaeology.
Milanich leads off with a chapter on the history of
archaeological research in Florida, one that goes back more
than a century. Between the 1860s and the 1920s, exploratory
and descriptive work was carried out by Wyman, Cushing,
Moore, and Fewkes, among others. Taxonomic formulations
of cultural complexes and their geographic and chronologic
ordering followed in the early 1940s. Such systematics
remained the prime goals for more than a decade following
World War II. By the 1960s, however, Florida archaeologists
were adding to such traditional objectives as they pursued
information on cultural-ecological adaptations and culture
change, particularly as these pertained to the reconstructions of
social, political, and ideological systems. It is also to be noted
that over these past 30 years or so archaeological investigations
in the state have been steadily increased through the growth of
local institutional support and the numbers of archaeologists
As a reviewer, let me confess that I address this book
with a certain sense of nostalgia--from the perspective of one
who was, a long time ago, involved for a brief period with
Florida archaeology. What was it like then? And how does
"then" compare with "now"? My comments will be selective.
They will also be essentially laudatory. I leave more critical
appraisals to those more conversant with the current Florida
archaeological scene than I. From my point of view, I am very

happy with the status of Florida archaeology. It seems to me it
has progressed very well indeed, proceeding through those
logical steps or stages outlined in the preceding paragraph.
While I believe this would be criticized by some as a
"Whiggish" outlook, I willingly confess to being of a
"Whiggish" turn of mind.
Milanich organizes his material chronologically, and I
shall begin by following along that way. In 1940, a
Paleoindian horizon for Florida was largely hypothetical.
Such data as there were pertained mostly to human skeletal
materials of debatable antiquity. True, there were a few other
clues in random finds of fluted projectivel points, but these
lacked good in situ contexts. Now, Milanich can devote
several pages to discoveries which demonstrate a substantial
late Pleistocene occupation of the state--one replete with
Clovis, Suwannee, and Dalton points--and he even goes as far
as talking about an "Oasis Model" of Paleoindian settlement, a
reference to ancient kill and camp sites centered around former
watering places.
Moving on to the Archaic, it is of interest to recall that
what were probably the first systematically recorded
archaeological excavations in Florida--Wyman's St. Johns shell
mound efforts--were concerned with Archaic cultures. Wyman
observed pre-pottery and early fiber-tempered pottery levels in
stratified deposits. Over the years, some subsequent attention
was given to fiber-tempered wares in the St. Johns region, and
in 1940 fiber-tempered sherds were found in the bottommost
levels of northwest Florida coastal shell middens. In the years
since, information on the Archaic---both preceramic and early
ceramic--has poured in from many quarters. Currently, Early
(7500-5000 B.C.), Middle (5000 3000 B.C.), and Late
(3000-500 B.C.) Archaic divisions are defined, all with
appropriate radiocarbon dates. Milanich is able to describe
and amply illustrate stone and bone artifact inventories from all
of them. Further, fiber-tempered pottery, which characterizes
the late Archaic, can now be typologically and chronologically
subdivided within the latter part of the Late Archaic and to a
time span of 2000 to 500 B.C.
Milanich uses the general term of "Regional cultures" to
refer to those which follow the Archaic, and these can now be
treated in some detail from virtually all sections of the state.
The Deptford culture or complex (ca. 500 B.C. 100 B.C.),
which we had known formerly from a few potsherds, is now
recognized from many sites in both panhandle and peninsular
Gulf Florida. Previously, we thought that all Deptford sites
were small village middens; but Milanich, drawing upon newer
research and interpretations, makes a case for burial mound


Vol. 47 No. 3


ceremonialism (the "Yent Complex," as originally defined by
W. H. Sears) beginning as early as Deptford and featuring
ritual paraphernalia (copper panpipes and ear ornaments and
various polished stone forms), suggesting Adena-Hopewellian
Burial mound ceremonialism was fully established in the
succeeding Swift Creek and Santa Rosa-Swift Creek cultures
(100 B.C. A.D. 200) of Gulf Florida. Milanich now has
information to allow him to revise our old Santa Rosa-Swift
Creek concept. Swift Creek Complicated Stamped and the
Marksvillian- Hopewellian incised and rocker-stamped pottery
are no longer lumped into a single Santa Rosa-Swift Creek
complex. Instead, a Santa Rosa culture is now defined for the
western Florida panhandle where the Santa Rosa types are
found in both village middens and burial mounds. In contrast,
there is a Swift Creek culture in the eastern panhandle, where
village middens show only Swift Creek pottery with the
relatively rare Santa Rosa types confined to the ceremonial
contexts of burial mounds. This more refined classification
embodies an improved understanding of the cultural processes
involved in this case, cultural borrowing with status
In my treatment of Weeden Island in Archeology of the
Florida Gulf Coast (1949), I conceived of it as a single culture
with two significant chronological divisions: Weeden Island I
(with Late Variety of Swift Creek Complicated Stamped and
Weeden Island Incised and Punctated pottery); and Weeden
Island II (with a disappearance of Swift Creek Complicated
Stamped, continuation of some Weeden Island Incised and
Punctated types, and a strong appearance of a type known as
Wakulla Check Stamped). As of today, this picture has been
greatly refined and expanded, with Milanich summarizing
seven Weeden Island variants.
In addition, the chronological division between Weeden
Island I and II has been given new importance, at least in the
panhandle region. This has come about by the discovery of
evidence for maize agriculture in Weeden Island II, or
"Wakulla Weeden Island." Back in the 1940s, I offered the
speculation that the Weeden Island culture (both Periods I and
II), and perhaps the Santa Rosa-Swift Creek culture which
preceded it, may have had maize farming. To me, at that time,
such a speculation seemed supported not only by Weeden
Island and Santa Rosa-Swift Creek cultural complexity, but by
the knowledge that small amounts of maize had been found in
northern Hopewellian sites. Today, there is general consensus
among most Eastern United States archaeologists, that the
significant growing of maize did not get underway until late
Middle Woodland times. Weeden Island I (ca. A.D. 200-750)
would have been too early for it; but Weeden Ilsnd II (ca. 750-
1000) has a time range in which maize might have been
present. And, apparently, it was, at least in the interior parts
of northwest Florida.
This shift over to agriculture is now believed to have
effected significant changes in the Weeden Island social and

political systems. The larger communities and burial mound
centers of earlier Weeden Island times were abandoned,
probably as a result of population dispersals associated with the
frequent shifting of farm lands. Eventually, after A.D. 1000,
the Mississippian- influenced Fort Walton culture of the region
would initiate a new settlement and social integration in which
numerous farming villages were drawn together by paramount
temple mound centers, but in the intervening Weeden Island II
time interval the old burial mound ceremonialism declined.
One aspect of this decline was the drop-off in the elaborately
modeled and ornamented burial pottery that had characterized
Santa Rosa-Swift Creek and Weeden Island I burial mounds.
As I reflect on all this, I am impressed again--as I was
fifty years ago--with the unusual flamboyance of Gulf Coast
Florida pottery on a middle Woodland time level.
Fantastically modeled ceramics are not a northern Adena-
Hopewellian trait; nor is Marksville pottery from the Lower
Mississippi Valley especially elaborate in this regard. Indeed,
for the time span of A.D. 200 to 750--or prior to the rise of
the substantially later Mississippian modeled wares--the most
imaginative and elaborate pottery in the Eastern United States
is found in Gulf Florida.
Florescences like this are never easy to explain, but,
inevitably, they arouse our curiosity. What was there in the
life of the Gulf Coastal estuaries and bays that offered the
stability and the incentives for such a climax? Was it a rich,
predictable, and place-stable economy, one focused upon
abundant marine resources but also substantially supplemented
by land hunting and collecting? Did such a subsistence base
support the rise of a number of competing "great men," each
with his ceremonial center in which his ancestors, himself, and
his descendants would be memorialized in a richly furnished
burial mound? Did the competition for prestige lead to the
creation of burial wealth in the form of ornate ceramics we
find in the Weeden Island burial mounds?
Would the introduction of a maize farming economy into
such a context been as profoundly disturbing as Milanich and
others have hypothesized? It is certainly a reasonable
hypothesis to be further explored. However, as we shall see
further along, the role of maize agriculture in Florida's
Precolumbian past may be a complex one.
I have stayed too long on Santa Rosa, Swift Creek, and
Weeden Island--my old special interests. The late Regional
cultures of the Gulf coast deserve some mention. In brief, the
Fort Walton culture of the panhandle, which succeeds Wakulla
Weeden Island, is strongly Mississippian-impressed, with
temple platform mounds, large centers, and Mississippian
innovations in ceramics. It is now divided into two parts, Ft.
Walton in the east and a Pensacola culture to the west. Some
of those who have worked on these cultures recently would
seem to see more continuity out of Weeden Island than I did.
One would presume that the Fort Walton people were the
Apalachees. Were the Weeden Islanders of the panhandle also
Apalachees? Milanich will undoubtedly face up to this


question and others like it in the book to come. Safety Harbor
culture also reflects Mississippian influences. Temple mound
ceremonialism is an example; however, Safety Harbor
ceramics show very obvious continuity out of Weeden Island.
There is one difference, though. A decline has set in. The old
zestful Weeden Island style is gone.
To move away from the Gulf coast, there are very full
and substantial discussions of Regional cultures from the other
parts of the state to which I cannot do justice here. Milanich is
able to discuss the St. Johns region--northeast quadrant of the
Florida peninsula--in a fine-grained chronological frame of
reference, much of which reflects the diligence of the late John
Goggin. St. Johns ceramics are in a "chalky ware" tradition
which persists from 500 B.C. up to historic times. Most of
this pottery, found in both village sites and burial mounds, is
undecorated. Contacts with the Gulf Coast regions are seen at
various points in the sequence: Weeden Island decorated
vessels turn up in burial mounds of the St. Johns Ib Period
(A.D. 500-750); Wakulla-like Check Stamped appears in the
St. Johns IIa Period (A.D. 750-1050); and trade sherds of the
Fort Walton and Safety Harbor styles are present in the St.
Johns IIb Period (A.D. 1050-1513). In numbers of
communities, sizes of village middens, and numbers and sizes
of associated burial mounds, the old St. Johns inhabitants were
undoubtedly the equal of the Gulf Coast groups in population
numbers and political organization--in spite of their preference
for plain pottery.
There are detailed treatments of other regions, including
those of the interior and southern parts of the state. I have a
particular interest in the Okeechobee region of south Florida as
I once had the task of reporting on the sites of Belle Glade and
Big Mound City from there. This was done from my
examination of Smithsonian collections which came from that
institution's Federal Relief excavations of the 1930s. I had
little in the way of field notes with which to work, and my
contribution was largely a descriptive one. I remember,
though, my excitement in examining the wooden artifacts from
the Belle Glade site--carvings of birds, animals, and humans,
as well as trays, pestles, and a four-legged wooden stool. I
was also astounded by the amazing earthworks at Big Mound
City. To my way of thinking then, all such features that I
knew about were from places like the Ohio Valley.
Fortunately, there is another Okeechobee earthwork site,
a place called Fort Center, which was excavated in detail and
with great care some years ago by Sears. In addition to Fort
Center's extraordinary earthworks, the most startling find from
there a site which ranges in time from the late Archaic up to
the historic is the presence of maize pollen in a number of
contexts, including those which would indicate that maize was
a part of the diet of the inhabitants. And, equally surprising,
this maize pollen is not late in the site's history, but it appears
by at least 500 B.C. If this can be accepted at face value--and
Milanich seems to feel that it can--it suggests a separate and
disconnected occurrence of maize agriculture in ancient Florida

from that recorded in the Weeden Island II and Fort Walton
contexts from the panhandle region. What are we to make of
it? Should it be taken to imply that maize agriculture was a
part of the food economy of other Florida prehistoric cultures,
such as Weeden Island and the others we have been discussing?
Or is it to be looked upon as an isolate in south Florida, the
result, perhaps, of a diffusion to that part of the peninsula from
Mesoamerica? Or is the Fort Center maize pollen some kind
of an error in the sampling, a more recent contamination? See
Milanich for the pros and cons of this.
There is an ample treatment in the book of the Glades
region, the southernmost in the state. The region is known for
the spectacular painted wooden objects which Cushing
recovered from the muck at Key Marco so many years ago.
But the region is now known in a much more encompassing
way as a result of the recent studies of Randolph Widmer on
the Calusa culture.
The format of Archaeology of Precolumbian Florida is
impressive. The type face is attractive. There are numerous
illustrations. All the drawings and tables are excellent and get
across their respective messages without any ambiguities, and
there are a generous number of photographic illustrations. I
think here I will register my only complaint. While these
photo reproductions are all right, they are not excellent. This,
I am afraid, is a criticism that could be levied against many
modem book illustrations. They tend to be duller and flatter
than the old fashioned half-tones (for example, the publications
of C. B. Moore, ca. 1900-1918).
Let me close by saying that with this book, Florida
archaeology has come of age. It is now evident that there is an
adequate formal-spatial-temporal structure that has been
reasonably well filled out. Archaeologists can now examine
past social and cultural behaviors within that structure and ask
and pursue questions concerning the whys and wherefores of
those behaviors.


Early Pottery in the Southeast: Tradition and Innovation in
Cooking Technology, Kenneth E. Sassaman. (University of
Alabama Press, Tuscaloosa, 1993. xvi, 304 pages, figures,
tables, references, index. $27.95 paper).

Reviewed by Nancy Marie White, University of South Florida

The earliest appearance of ceramics in the archaeological
record used to be thought of as a sign of major sociocultural
change, the origins of sedentism and early agriculture. Then
when we learned people could be well-settled without farming
(for example in rich coastal or lakeshore environments), or
could cultivate crops without being settled, or could develop
fired clay technology before either, opinions changed. The
first pottery was sometimes considered more important to the
archaeologist than to the prehistoric people who developed it,
since they would have had many other cooking technologies
and their lives would not have changed so much by simply
adding another type of container to their material inventory.
Other studies emphasizing functional approaches indicated
pottery indeed represented a vast improvement in technology
that did affect society. The oldest pottery in the Southeast is
fiber-tempered and dates from over 4500 years to some 3000
years ago. Sassaman's study demonstrates how it must be
understood in relation to variable social interaction and
exchange among Late Archaic hunter-gatherers.
The first part of the book summarizes data on early
southeastern pottery and traditional diffusionalist and
functionalist interpretations of its appearance and spread, as
contrasted with Sassaman's social interpretations. Using data
from the Savannah River Valley region of Georgia and South
Carolina, which has had extensive archaeological investigation,
he examines the contrasting distributions of fiber-tempered
ceramics and soapstone (steatite) from coastal to interior
subregions, and the chronology of different ceramic types and
assemblages. He then presents excellent technofunctional
analyses of pottery, baked clay objects ("clay balls"),
soapstone vessels and perforated soapstone slabs, using
ethnohistoric, experimental and archaeological data. Soapstone
slabs, for stone boiling, occur widely throughout the Coastal
Plain, where this material does not occur naturally. Baked
clay objects, thought to be alternatives to the soapstone, are
demonstrated to be ineffective as boiling stones. Mechanical
performance of different pottery styles helps to distinguish
direct-heat vessels from those used in indirect heat cooking
(stone boiling) and use wear such as soot on vessels
corroborates the classification. Direct-heat cooking pots were
used throughout the Late Archaic on the coast but indirect heat
cooking persisted in the interior long after pottery was

Sassaman's interpretive chapter discusses these alternative
cooking technologies in the context of production and
exchange. He argues that in the interior, individuals gaining
prestige from direct access to soapstone influenced people to
resist the innovation of ceramic vessels and maintain the stone
boiling with soapstone. He assumes the soapstone exchange
was controlled by men, and that resistance to innovation had
ritual importance. On the coast, adopting pottery presented
less of a threat to existing exchange relationships because of
different socioeconomic organization. Large coastal shell rings
suggest larger groups that stay together more. Sassaman sees
an increase in women's labor requirements as shellfish
collecting intensifies and thus a faster cooking method is
favored. He hypothesizes a possible ceremonial aspect to
constructing shell mounds and corresponding enhancement of
women's status, thus the impetus for women to develop more
efficient direct-heat cooking with ceramic vessels.
The Savannah River Valley case study is used to model
cultural processes involved in the diffusion of fiber-tempered
pottery elsewhere in the Southeast. Much more work is
needed to examine the broader implications of this study and
the relationship of the emergence of ceramics with the Poverty
Point interaction sphere. Sassaman's map showing major Late
Archaic pottery traditions (Figure 3) shows a huge gap for
most of Florida and all of the Gulf Coastal Plain, for example,
even though there is lately more and earlier evidence from
these areas. But the data and technological studies in this book
are superb and provide several hypotheses for testing in many
In addition, the study goes beyond chronology, function,
and ceramic attribute lists and makes a great attempt to talk
about people, individuals and society. The book is a revision
of Sassaman's dissertation done at the University of
Massachusetts at Amherst, a place known for postprocessual,
Marxist, and other theoretical innovations that seldom are
acknowledged in southeastern archaeology. Such an intellectual
atmosphere does result in phenomena such as difficult phrasing
and overuse of postprocessual buzzwords (coastal groups have
"constrained options for fissioning") as well as more positive
contributions such as willingness to discuss individuals in
prehistory, socioeconomic roles, and the divisions of and
control of labor.
I commend the author for his brave discussions of gender
in prehistory, though I could disagree with several aspects. It
is dangerous to assume that gender roles seen in the
ethnographic record had persisted through several millenia of
prehistory. If anything, rather than try futilely to determine
the gender of the prehistoric actors, we might make more
dependable statements about their age: children would have a
much easier (safer, more flexible,) time collecting shellfish
than they would fishing or hunting or even gathering other,


Vol. 47 No. 3



more widely dispersed resources. Is shellfish collecting really
more intensive a labor than fishing or hunting? Furthermore,
we now often interpret shellfish consumption as something
done as a sideline to obtaining fish and other much more
optimal coastal resources; an enormous pile of shells really
represents a far smaller amount of meat. As for intensive
dependence on shellfish accompanying increasing status for
women, it is assumed shellfishing (and large group sedentism)
took off during the Late Archaic, but it may have begun or
even intensified much earlier. Paleo-Indians probably ate
plenty of shellfish but we lack proof because their shell
middens were all drowned by rising sea levels since the
Sassaman does not try to explain the origins of
southeastern fiber-tempered pottery and its relationship to
fiber-tempered ceramics in northern South America, where
they are even earlier. (The newest evidence from the lower
Amazon indicates that other types of ceramics predate fiber-
tempered wares, going back 8000 years.) As we find with
many other material correlates of behavioral patterns, there
was probably wide variation in the types, contexts, and
frequencies of pottery use among the first ceramic-making
peoples. Pottery's early appearance on coasts could just as
easily be due to diffusion across water from South America and
along U.S. coasts. Additional case studies from interior,
riverine shell middens might determine if it is the shellfishing
itself that is important, not the coastal location.
This book is highly recommended for any southeastern
archaeologist. It is well written, with good photos and
readable graphs, maps, and tables. More important, it nicely
combines technofunctional, chronological, and environmental
studies with the theoretical advances for the nineties, so there
is something for everyone. While interpretations of the wider
archaeological implications may be debatable, the thorough
discussions of techniques, implements, and contexts for
cooking provide much food for thought.



Theodore Morris

I am creating a series of oil paintings depicting the
vanished tribal peoples of Florida. My purpose is to try to
bring them back to life. The paintings are confrontational in
the sense that the figures are looking directly out at the viewer.
Seven of my paintings were shown at the South Florida
Museum and Bishop Planetarium in Bradenton, Florida, in
June and July 1994. Laura Branstetter, Curator at the
museum, arranged the exhibition. Ten paintings will be on
view at the Port Royal Museum Gallery in Naples, Florida,
during August and September, 1994.
Most of the artifacts in the accompanying painting
(Figure 1) are based on an archaeological report written by
John W. Griffin and Hale C. Smith concerning the Goodnow
Mound in Highlands County, Florida, and published by the
~*cBSgyti-lf n. -a -

Florida Park Service in 1948. The mound is located
approximately six miles south of Sebring. The artifacts shown
are of European derivation, obtained through trade and from
shipwrecks, and were used in indigenous ways. The old man
wears a single strand of blue glass beads around his neck as
well as strands of glass seed beads which are mostly white with
a small quantity of blue, yellow, and green. He holds a silver
pendant in his left hand, and he wears blue glass seed beads on
his wrists. His ear decorations are made from fish bladders.
The child's necklace is made of multicolored glass seed beads
with one silver bead. I wish to thank George M. Luer for his
expert advice and critical eye on all my paintings. I would
also like to thank Jeffrey M. Mitchem, James S. Lord, and
Ryan J. Wheeler for their input concerning artifacts.

Figure 1. This painting, titled Highlands County Legacy by Sarasota artist Theodore Morris, depicts two members of the
postcontact Mayaimi tribe of south-central Florida. The artist can be contacted at 1211 34th Street, Sarasota, FL, 34234,813-351-


Vol. 47 No. 3





submitted by Keith Lawrence, District Ranger, Osceola National Forest

Entering this fully accessible trail in the Osceola National
Forest is like stepping back into a memory. It winds through
pines and hardwoods, eventually finding its way to a
boardwalk which crosses a Florida swamp, before ending on
the edge of a lake. You will step back 130 years in time,
discover what logging was like in its heyday, and see what the
oldest steam locomotive in Florida looked like.
Employees of the Osceola Ranger District cut and built
the railroad ties and reconstructed the tramway with original

track from that era. It was here on the shore of Ocean Pond
that John Russell and Isaac Eppinger built and operated a
sawmill in the late 1890s and early 1900s.
After reading the interpretive signs and imagining
hundreds of logs floating in front of you and the distant whine
of saws, you can take a dip in the lake, have a picnic, or go
boating and fishing. Ocean Pond offers a little something for
everyone. For additional information contact Osceola National
Forest at (904) 752-2577.

Laying railroad track. Osceola National Forest employees aligning track from the late 1890s. Left to right, Bruce Hagen, Keith
Lawrence, Robert Stalvey, Morris Morgan, and Kenneth Owen.


Vol. 47 No. 3




Text and photo submitted by Vera Zimmerman, Indian River Anthropological Society

The Roesch House was built about 1900 in the town of
Eau Gallie and is named for its former resident, the mayor of
the town. Eau Gallie was later annexed by Melbourne. The
five-foot deep cistern was used to hold drinking water gathered
by guttering and downspout from the eaves of the house. It
was probably in use from 1900 until 1948, when the house was
no longer occupied.
The Roesch House is being renovated by the Florida
Historical Society with upstairs offices to be used as their

headquarters. Original plans called for the cistern to be
covered by the handicapped entrance ramp. Associate Director
of the FHS, Emily Perry, says that plans have now been
changed to use the cistern as an educational exhibit. The many
canning jars, bottles, historic ceramic sherds, and other
artifacts found in the cistern will be cleaned and cataloged by
IRAS members and used by the FHS in an exhibit. For more
information, contact Emily Perry, Associate Director, Florida
Historical Society, (407) 259-0847.

Indian River Anthropological Society members Dean Zimmerman, Walt Hersing, and Jan Rattenbury help excavate the cistern at
8BR1650, the Roesch House, in Melbourne, Florida.


Vol. 47 No. 3



Join the Florida Anthropological Society (FAS)!
A non-profit organization founded in 1947, with chapters throughout Florida

Florida Indian
This Bird-man
Dancer is the
main illustration
of an attractive
and informative
poster depicting
the major tribes
that once in-
habited Florida.
Available for a
$6.50 donation
to FAS, this 18 by
36-inch poster is
printed maroon
and purple on a
heavy paper.

Anthropology is the study of people and their cultures. Join FAS and help
save and enjoy Florida's heritage! FAS holds an annual meeting and banquet
featuring renowned speakers. FAS members receive a newsletter and informa-
tive journal four times a year. The journal features interesting articles on
Florida archaeology, history, folklore, and preservation.

El YES! I want to join FAS!
I Membership is only $20 per year (individual) and is tax-deductible.
SOther rates: $20 institutional, $25 family, $35 or more, sustaining,
Patron $100, and life $500.
D0 YES, I would like to donate an additional $6.50, also tax-deductible,
and receive a poster by mail (allow 3-5 weeks).
I Address:
SCity: State: Zip:
STelephone:( )
SFAS Membership, c/o Terry Simpson, CGCAS, P.O. Box 82255,
STampa, FL 33682

Join the Florida Anthropological Society (FAS)!

A non-profit organization founded in 1947, with chapters throughout Florida

Florida Indian (
This Bird-man
Dancer is the
main illustration
of an attractive I
and informative
poster depicting
the major tribes
that once in-
habited Florida. '
Available for a
$6.50 donation
to FAS, this 18 by
36-inch poster is
printed maroon
and purple on a
heavy paper.

Anthropology is the study of people and their cultures. Join FAS and help
save and enjoy Florida's heritage! FAS holds an annual meeting and banquet
featuring renowned speakers. FAS members receive a newsletter and informa-
tive journal four times a year. The journal features interesting articles on
Florida archaeology, history, folklore, and preservation.

r m=== ======m ==m=m== ===----- -.- -- --
SO YES! I want to join FAS!
SMembership is only $20 per year (individual) and is tax-deductible.
SOther rates: $20 institutional, $25 family, $35 or more, sustaining,
patron $100, and life $500.
ID YES, I would like to donate an additional $6.50, also tax-deductible,
and receive a poster by mail (allow 3-5 weeks).

I Name:

City: State: Zip:

Telephone: ( )
SFAS Membership, c/o Terry Simpson, CGCAS, P.O. Box 82255,
I Timna FI TIAR9

Working on the




Published in the Spring of 1983

It and other back issues

of The Anthropologist

are available from the

481 South




FL 33004

(305) 9
FAX (305)


Sole Agents for Back Issues of
The Florida Anthropologist

J 1h





You and

Florida's Past

Florida's history is long: it goes back 10,000
years to people who hunted mammoth with
stone-tipped spears.
It is colorful: 7,000 years ago, Florida's Native
Americans wove cloth as fine as a T-shirt.
It is unique in the world: around 800 years
ago, some Floridians had a civilization so
complex that they built long canoe-canals and
huge pyramid-shaped mounds of shells and
You can be part of it! New pages of this story
are being written every week. Teams of
amateur and professional archaeologists
together are making fascinating discoveries in
the field and in the lab.
You can help save it! Florida's rapid develop-
ment puts many valuable sites in jeopardy.
Amateur and professional archaeologists,
elected officials and planners, and just plain
concerned citizens are working together to
save this history in the soil.
How do you put yourself into this picture? By
joining the Florida Anthropological Society
(FAS) or one of its chapters, or both, as many
interested citizens do!


Each spring, an FAS chapter hosts a state-
wide meeting attended by members of FAS
and its chapters, and the public. Both pro-
fessionals and amateurs deliver papers about
their activities and investigations. A banquet
features a guest speaker who is usually
nationally-known in the field of archaeology
or anthropology. FAS elected officers are
instated at a business session.
During the year, the FAS Executive Board
holds several meetings. FAS chapters have
monthly meetings, field trips, and other


1. FAS publishes a scientific journal, THE
year. Both professionals and amateurs con-
tribute articles about investigations in Florida
and nearby areas. These articles keep FAS
members up-to-date on many aspects of
Florida archaeology, history, folklore, and pre-
servation. Many libraries around the nation
and world subscribe to the journal.

2. FAS publishes a newsletter four times a
year which keeps FAS members abreast of
FAS chapter activities and of pertinent events
and news around the state and wider region.


FAS has chapters throughout Florida
which are open to the interested public. By
joining FAS and one of its chapters, citizens
can take an active part in helping to study and
preserve Florida's heritage. Activities include
meetings, field trips, and archaeological digs
supervised by professionals.

FAS Chapters
Write your area's chapter for membership informa-
tion today!
Archaeological Society of Southern Florida
2495 NW 35th Avenue, Miami, FL 33142
Broward County Archaeological Society
481 S. Federal Hwy., Dania, FL 33004
Central Florida Anthropological Society
810 East Rollins Street, Orlando, FL 32803
Central Gulf Coast Archaeological Society
P.O. Box 82255, Tampa, FL 33682
Indian River Anthropological Society
3705 S. Tropical Terrace, Merritt Island, FL 32952
Kissimmee Valley Arch. & Hist. Cons.
P.O. Box 970, Sebring, FL 33871
Northeast Florida Anthropological Society
10415 Skycrest Dr., Jacksonville, FL 32216
Pensacola Archaeological Society
P.O. Box 13251, Pensacola, FL 32591
St. Augustine Archaeological Association
P.O. Box 1987, St. Augustine, FL 32085
Southwest Florida Archaeological Society
P.O. Box 9965, Naples, FL 33941
Time Sifters Archaeology Society
P.O. Box 25642, Sarasota, FL 34277
Volusia Anthropological Society
P.O. Box 504, New Smyrna, FL 32170








-: -;--.





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