Table of Contents
 Editor's Page
 Faunal Remains from the Palmer...
 Zoological Indicators of Habitat...
 Archaeobotanical Research at Shell...
 Heineken Hammock, 8CR231: A Late...
 Comment: Status of Florida's Isolated...
 Book Reviews
 About the Authors

Group Title: Florida anthropologist
Title: The Florida anthropologist
Full Citation
Permanent Link: http://ufdc.ufl.edu/UF00027829/00177
 Material Information
Title: The Florida anthropologist
Abbreviated Title: Fla. anthropol.
Physical Description: v. : ill. ; 24 cm.
Language: English
Creator: Florida Anthropological Society
Conference: Conference on Historic Site Archaeology
Publisher: Florida Anthropological Society.
Place of Publication: Gainesville
Frequency: quarterly[]
two no. a year[ former 1948-]
Subject: Indians of North America -- Antiquities -- Periodicals -- Florida   ( lcsh )
Antiquities -- Periodicals -- Florida   ( lcsh )
Genre: periodical   ( marcgt )
Summary: Contains papers of the Annual Conference on Historic Site Archeology.
Dates or Sequential Designation: v. 1- May 1948-
 Record Information
Bibliographic ID: UF00027829
Volume ID: VID00177
Source Institution: University of Florida
Holding Location: Department of Special Collections and Area Studies, George A. Smathers Libraries, University of Florida
Rights Management: All rights reserved by the source institution and holding location.
Resource Identifier: oclc - 01569447
lccn - 56028409
issn - 0015-3893

Table of Contents
    Table of Contents
        Table of Contents
    Editor's Page
        Page 176
    Faunal Remains from the Palmer Site (8OS2), with a Focus on Shark Remains
        Page 177
        Page 178
        Page 179
        Page 180
        Page 181
        Page 182
        Page 183
        Page 184
        Page 185
        Page 186
        Page 187
        Page 188
        Page 189
        Page 190
        Page 191
        Page 192
    Zoological Indicators of Habitat Exploitation and Seasonality from the Shell Ridge Midden, Palmer Site (8SO2), Osprey, Florida
        Page 193
        Page 194
        Page 195
        Page 196
        Page 197
        Page 198
        Page 199
        Page 200
        Page 201
        Page 202
        Page 203
        Page 204
        Page 205
        Page 206
    Archaeobotanical Research at Shell Ridge Midden, Palmer Site (8SO2), Sarasota County, Florida
        Page 207
        Page 208
        Page 209
        Page 210
        Page 211
        Page 212
        Page 213
        Page 214
        Page 215
        Page 216
        Page 217
        Page 218
        Page 219
        Page 220
        Page 221
        Page 222
    Heineken Hammock, 8CR231: A Late Archaic Corridor Site in Collier County
        Page 223
        Page 224
        Page 225
        Page 226
        Page 227
        Page 228
        Page 229
        Page 230
        Page 231
        Page 232
        Page 233
        Page 234
        Page 235
        Page 236
        Page 237
        Page 238
        Page 239
        Page 240
    Comment: Status of Florida's Isolated Finds Program
        Page 241
        Page 242
    Book Reviews
        Page 243
        Page 244
        Page 245
    About the Authors
        Page 246
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 51 Number 4
December 1998


Editor's Page. Robert J. Austin 176


Faunal Remains from the Palmer Site (80S2), with a Focus on Shark Remains. Laura Kozuch 177

Zoological Indicators of Habitat Exploitation and Seasonality from the Shell Ridge Midden,
Palmer Site (8S02), Osprey, Florida. Irvy R. Quitmyer 193

Archaeobotanical Research at Shell Ridge Midden, Palmer Site (8SO2),
Sarasota County, Florida. Lee A. Newsom 207

Heineken Hammock, 8CR231: A Late Archaic Corridor Site in Collier County. Arthur R. Lee,
John G. Beriault, Jean Belknap, Walter M. Buschelman, John W. Thompson, and Carl B. Johnson 223


Status of Florida's Isolated Finds Program. Robert L. Knight 241

Acknowledgment of Reviewers 242


Milanich et al.: Archaeology of Northern Florida, A.D. 200-900: The McKeithen Weeden
Island Culture. Karl T. Steinen 243
Emerson: Cahokia and the Archaeology of Power. John Scarry 244

About the Authors 246

Cover: The Malleability of Material Culture by Scott Mitchell

Copyright 1998 by the
ISSN 0015-3893


Southwest Florida is the focus of this issue of The Florida
Anthropologist, with a special emphasis on the Palmer site
in Sarasota County. Excavations at this well-known coastal
site were conducted by Ripley and Adelaide Bullen in the late
1950s and early 1960s. Their report of the excavations was
published in FASP No. 8 in 1976. In 1991, excavations were
conducted in the Palmer site's Shell Ridge Midden by
archaeologists Marion Almy, George Luer, and Corbett
Torrence, assisted by members of Time Sifters Archaeology
Society. These excavations were conducted in preparation
for a permanent exhibition entitled "A Window to the Past,"
that features a profile of the shell mound. The first three
articles in this issue are based on environmental archaeology
studies conducted as part of that project.
Laura Kozuch's paper presents the results of a faunal
analysis of samples from the "Window to the Past" excava-
tion, as well as faunal remains recovered by the Bullens. In
addition to providing a detailed accounting of the species
exploited by native peoples during the late Manasota and
early Safety Harbor periods, she also presents methods for
recovering shark dermal denticles. The recovery and
quantification of these very small faunal remains provides
important information on shark exploitation and documents
their presence even when teeth are absent.
The second paper by Irv Quitmyer provides multiple lines
of faunal evidence indicating that the occupants of the Shell
Ridge Midden occupied the site during all seasons of the
year, although different resources were collected during
different times of the year.
Archaeobotanical data are the focus of Lee Newsom's
paper which describes the types of plant species that were
exploited at the site. An important component of her paper
is the comparison of the Palmer site's archaeobotanical
assemblage with similar assemblages from other sites in
south Florida. Her comparison indicates relatively close

similarities in the types of species exploited, with minor
variations due to geographic location. In general, south
Florida native peoples practiced a strategy of opportunistic
collecting of a wide range of edible plant species.
The fourth paper is by members of the Southwest Florida
Archaeological Society, Art Lee, John Beriault, Jean Bel-
knap, Walt Buschelman, Jack Thompson, and Carl Johnson.
They present the results of an excavation at a small, interior
site in Collier County. This report adds to the information
that has been accumulated by SWFAS on similar small sites
on the southwest coast. These studies provide an important
element to understanding prehistoric land use in southwest
In our Comments section, Robert Knight provides an
update on the Isolated Finds Program. After a year of
operation, the program has resulted in the recording of many
new sites and the overall response has been encouraging.
The program is an excellent example of how amateur and
avocational archaeologists can work together with profes-
sionals to better understand the prehistory of Florida.
Finally, we have book reviews by Karl Steinen and John
Before closing, I want to acknowledge the hard work and
assistance of Pam Vojnovski, who has helped me prepare the
journal for mail delivery since I began my editorship in
December of 1995. Pam has taken a position with the
C.A.R.L. Archaeological Survey in Tallahassee, and so had
to resign her position as Editorial Assistant. Pasting labels,
bundling journals, and running errands for the Editor are not
glamorous jobs, but Pam did them all conscientiously and
with good humor. Her work "behind-the-scenes" made my
job as Editor much easier.
I hope you enjoy this issue of the journal.




In the last issue of the journal (Vol. 51, No. 3), on page 147 of "Ninety-Nine Bottles of Beer in the Wall: A Turn-of-the-
century Bottle Well in New Smyrna Beach" by Dana Ste. Claire, Dorothy L. Moore, and Robert W. McKinney, the second
sentence of the second paragraph, right-hand column, mistakenly attributes the land referred to in the letter by W.A. McRae
as belonging to Douglas Dummett. The sentence should read: "The McRae letter records Walker's land claim..." and later
"...if it should be allowed to claimant [Walker], it being...." In addition, the caption for Figure 2 of the same article should
read "Jason Burns removes bottles from the wall of the well while above Debbie Mullins (left), Bob McKinney, and Melanie
Wendt clean and tag bottles."
Submitted by Dorothy L. Moore


VOL. 51 NO. 4



U.S. Army Corps of Engineers, MCX-CMAC, St. Louis, Missouri 63105
E-mail: kozuch@icon-stl.net

In this paper I present the results of an analysis of prehis-
toric archaeological faunal remains from a coastal site
near Osprey, Florida, known as the Palmer site (8SO2).
The remains are from two separate excavation episodes.
Bullen and Bullen (1976) excavated there from 1959 to
1962. Later excavations were undertaken in 1991 by
archaeologist Corbett Torrence, supervised by archaeologists
Marion Almy and George Luer and assisted in the field by
members of Time Sifters Archaeology Society, a chapter of
the Florida Anthropological Society.'
Materials reported here were recovered from a strati-
graphic profile during the 1991 excavations in the "Shell
Ridge," and from the earlier Bullen excavations. An in-
depth analysis is presented of materials from the 1991
excavations, as well as a summary of the faunal remains
collected by the Bullens.
The Bullens recovered a large amount of faunal material.
Though provenience information is available, they did not
state the recovery method in their report, and thus a detailed
analysis of the remains is not possible. In 1996, I was able
to identify the Bullens' faunal material, and the relatively
large sizes of the bones suggest that their choice of which
bones to include in their sample was largely size-dependent.
These circumstances greatly reduce the amount of informa-
tion that a faunal analysis of the Bullens' material can
The objectives of an analysis of faunal remains can be
quite diverse, but always we want to know how humans and
animals interacted, and how products from animals were
used. My analysis focused on the range of species utilized,
the methods) of capture for these species, environmental
exploitation by humans, and, in particular, the use of shark
products. I also attempted to identify the species that were
targeted by prehistoric peoples for meat consumption, both
vertebrates and invertebrates.

The Palmer Site

The Palmer site has components dating from the Archaic
Period all the way through historic times. Figure 1 is a plan
of the site, which shows the site components and test units
discussed below.
The Bullens excavated a number of test pits, designated
Tests A through M, at various locations on the site (Figure
1). Test pits at the Hill Cottage Midden were designated A,

B, C, D, and K; the Shell Midden tests were designated F
and G; and the Shell Ridge was the location of tests E, H, I,
and J (Bullen and Bullen 1976).
The Shell Ridgemidden dates to approximatelyA.D. 100-
1200, and thus ranges from the mid-Manasota Period
through the early Safety Harbor Period. The Hill Cottage
Midden dates to Archaic times (Bullen and Bullen 1976),
and may be a shell ring. Lower levels of the Hill Cottage
Midden are ceramic. The Shell Midden (an eastward
extension of the Shell Ridge midden) dates from 200 B.C. to
A.D. 200. The Burial Mound is constructed of sand and also
contained animal bones. The Burial Mound probably dates
to the mid-to-late Manasota Period based on ceramic sherds
and radiocarbon dates recovered from it (Luer and Almy
The 1991 excavations were conducted in preparation for
an exhibition that incorporates an actual profile of the shell
mound as part of the exhibition and so was called "A
Window to the Past." The site is open to the public as a
historical park and preservation area, known as Historic
Spanish Point, which is owned and operated by the non-
profit Gulf Coast.Heritage Association.
The stratification of the Shell Ridge consists of layers
comprised of shells, sand, bones, and other cultural debris,
such as pottery sherds and charred plant remains. Some
layers contain articulated remains of both sharks and bony
fishes. These articulated vertebrate remains strongly suggest
rapid accumulation of midden material, since scavengers
(such as raccoons) would have been attracted to rotting
carcasses lying on the surface and would have disturbed the
remains. This argues for a rapid accumulation of midden
material. Further radiocarbon dating of the strata may help
to clarify how quickly the layers accumulated and thus how
rapidly the Indians built up this large shell mound.


Only pre-European faunal remains are examined here.
All of the analyzed materials are curated by the Florida
Museum of Natural History (FLMNH), Environmental
Archaeology Laboratory (FLMNH accession number 53).

1991 Sample

I analyzed one bulk sample collected from Unit 8-2-2 at


VOL. 51 NO. 4


1,1 1" Vol[. 50

Figure 1. Site plan of the Palmer site (modified from Bullen and Bullen 1976). Locations of the Bullens' test units in
1959-1962 are shown by capital letters. Note the location of the 1991 excavation in the Shell Ridge where the earlier Tests
H and I were located.

8 991 Vol 51(4)

T Fm Arfulmorat r


130to 136 cm below datum. This provenience was described
in the field as a "dark organic layer." The volume of the
entire sample was approximately 10 liters (2.5 gallons) and
was sent in its entirety to the Environmental Archaeology
Laboratory at FLMNH in Gainesville. The same unit and
level has an associated provenience (8-2-1), which was
designated in the field as an "ash locus." Material from this
locus was submitted for radiocarbon dating (Beta 48866),
producing a date of 900 60 uncorrected radiocarbon years
before present (B.P., present = A.D. 1950) or a calibrated age
of A.D. 1032 to 1177. Two additional radiocarbon samples
from nearby in the profile produced very similar dates,
indicating that this is a valid age range.
In addition to this sample, all vertebrate faunal materials
that did not pass through the 6.4-mm- (1/4-in-) mesh screens
were saved. I examined these for taxa not represented in my
analyzed bulk sample (see Appendix I). Moreover, a detailed
list of shark remains from the 6.4-mm (1/4-in) screens is
given in Appendix II. Element counts were not made since
this is intended to be a presence/absence list. Molluscan
remains were saved from general proveniences only if they
were to be part of the seasonality analysis (see Quitmyer, this

The Bullens' Samples

Much more faunal material was recovered from the Bul-
lens' Test A than from their other test units. (For the
locations and ages of these tests, see above and Figure 1; also
Bullen and Bullen 1976). The test units were excavated in
6-in (15-cm), arbitrary levels. As well as can be determined,
no screening of materials was performed, and faunal remains
were recovered by a visual search through midden matrix
that was spread on a table in the field. Only vertebrate
remains were saved for curation.


1991 Sample

After fumigation, the faunal materials from Unit 8-2-2,
130-136 centimeters below datum (cmbd) were air-dried
inside at room temperature, since they were still very damp.
The dried matrix was then screened through a series of
nested geological sieves of the following mesh sizes: 4 mm,
2 mm, 1 mm, 600 microns, 420 microns, 212 microns, and
125 microns. These mesh sizes were chosen to facilitate
sorting ofmaterials, both faunal and botanical. All botanical
remains were given to paleoethnobotanist Lee Newsom for
analysis (see Newsom, this issue).
After screening, all faunal materials that did not pass
through the 2-mm mesh were identified to the lowest taxon
possible. Materials that did pass through the 2-mm mesh
were analyzed for shark dermal denticles and botanical
remains only. Materials that passed through the 4-mm mesh
but did not pass through the 2-mm mesh were sorted for taxa
and identified only to the family level, or lower, since there

was much crushed shell and bone in the sample. A small
amount of material from the 1-mm screen was sorted to
determine if any significant information was being lost by
not analyzing the 1-mm fraction. Only three significant
elements were found: one flounder (Paralichthys sp.) gill
raker, one caudal tubercle from an Atlantic stingray
(Dasyatis sabina), and mouth parts from barnacles (Balanus
sp.). Since additional species or individuals were not added
by further study of the 1-mm fraction, no further sorting of
these I-mm materials was deemed necessary.
All sorted material from Unit 8-2-2, 130-136 cmbd, was
weighed, and the minimum number of individuals (MNI)
was determined (for an explanation of the method for
determining MNI, see Wing and Brown 1979:123). Frag-
ments of bivalves, molluscs, and barnacles were weighed
only, while fragments from all other taxa were counted. The
number offragments per taxon was designated as the number
of individual specimens (NISP) (see Grayson 1979).

Dermal Denticle Recovery

Before analysis of the 1991 materials, I devised a strategy
to recover shark dermal denticles (also known as placoid
scales) from the shell midden matrix. This had never been
attempted before in Florida, nor anywhere else to my
In the laboratory, I measured shark dermal denticles from
preserved shark skin samples from known species. The
shark skin samples are preserved in isopropyl alcohol, and
are curated at the FLMNH Ichthyology Range. Figure 2 is a
scanning electron micrograph of shark skin from a scalloped
hammerhead (Sphyrna lewini), showing how the dermal
denticles interlock to form the rough "skin."
Dermal denticles are embedded in the skin, and it is very
hard to remove a single denticle. They are composed of
enamel and dentin (Moss 1977), and therefore survive well
in the archaeological record. Measurements revealed that
the smallest denticles have a greatest measurement of.2 mm,
and so probably would not pass through the 125 micron
mesh. However, my measurements also showed that den-
ticles of the nurse shark (Ginglymostoma cirratum) are much
larger (1.2 mm) than denticles of other species with similar
body sizes (Table 1). At the time of analysis, I had hoped
that the denticles could be identified to shark species, but it
has since been shown that this is impossible at the present
time (Bargar and Thorson 1995; Raschi and Tabit 1992).
In order to see if dermal denticles indeed existed in the
Shell Ridge midden, a sample was taken from the ash lens
exposed in the profile excavation (Unit 8-2, 124-141 cmbd).
This provenience was screened using geological sieves with
mesh sizes of 4 mm, 2 mm, 1 mm, and .42 mm. I took a
volumetric sample of 1.0 ml from material that did not pass
through the .42-mm mesh, and discovered a number of
dermal denticles. All of the shark dermal denticles were
counted. I believe that these are the first denticles ever
identified from an archaeological context.
In addition to the volumetric sample, subsamples of the





Figure 2. Scanning-electron micrograph (120x) of modern scalloped
hammerhead shark skin sample from the Florida Museum of Natural History.

Table 1. Shark dermal denticle sizes from modern comparative
Size in Size in
mm microns

Isurus oxyrhinchus, mako .2 200

Ginglymostoma cirratum, nurse 1.2 1200

Carcharhinusfalciformis, silky .2 200

Carcharhinus limbatus, blacktip .3 300

Carcharhinus obscurus. dusky .2 200

Galeocerdo cuvieri. tiger .3 300

Rhizoprionodon terraenovae, .2 200

sieved material from the analyzed
sample (8-2-2, 130-136 cmbd) also
were examined for the presence of
dermal denticles. All fractions smaller
than 1 mm were found to contain der-
mal denticles. They were not quanti-
fied, however, since counting them
would have been time-consuming and
of little usefulness or significance.

The Bullens' Samples

A recent National Science Founda-
tion curation grant awarded to FLMNH
zooarchaeologist Elizabeth S. Wing
allowed me to identify the great amount
of faunal remains recovered by the Bul-
lens. Since many remains of smaller
animals were not collected due to the
biased nature of recovery (see above), an
in-depth analysis was not performed.
Analysis consisted of identification,
weights, and counts of the materials.
This information is now accessible on a
computer database at the FLMNH Envi-
ronmental Archaeology Laboratory.


1991 Excavation: Species Composi-

A complete list of all taxa from the
in-depth analysis (Unit 8-2-2, 130
cmbd), including MNI, NISP, skeletal
weight, and percentages for each, can
be found in Appendix III. Scientific
and common names for the species

recovered in the analyzed sample are
found in Table 2. Appendix I lists the vertebrate species that
were recovered from 6.4 mm screened material from else-
where in the excavation profile, but which were not repre-
sented in the sample that was subjected to in-depth analysis.
Mammals. The only mammal bones recovered from the
analyzed sample were from a single deer (Odocoileus
virginianus), and the epiphyses were fused, indicating that
the animal was an adult at the time of death. The three
elements are: one right radius with a proximal fused
epiphysis and 2.5-cm long proximal diaphysis, one left femur
with a proximal fused epiphysis with partial greater trochan-
ter and neck, and one left metacarpal with a proximal partial
epiphysis (indeterminate fusion). The only other mammal
species from the profile is a rabbit (Sylvilagus sp.) bone from
the general 6.4-mm screen (Appendix I).
Reptiles. The turtle bones recovered from the sample are
not sufficient enough to allow precise identification, although


Table 2. Scientific and common names of taxa from the
detailed analysis, 1991 excavation (8-2-2, 130-136 cmbs).

Taxon Common Name

Odocoileus virginianus


White-tailed deer


Carcharhinus limbatus
Rhizoprionodon terraenovae
Sphyrna lewini
Sphyrna tiburo
Dasyatis sabina

Amia calva
Bagre marinus
Opsanus sp.
Belonidae cf. Strongylura
Chriodorus atherinoides
Fundulus sp.
Diplectrum formosum
Epinephelus sp.
Caranx sp.
Lutjanus sp.
Eucinostomus argenteus
Eucinostomus gula
Eucinostomus sp.
Orthopristis chrysoptera
Haemulon sp.
Diplodus holbrooki
Lagodon rhomboides
Bairdiella sp.
Cynoscion arenarius
Cynoscion nebulosus
Cynoscion sp.
Leiostomus xanthurus
Menticirrhus littoralis
Menticirrhus saxatilis
Sciaenops ocellata
Mugil sp.
Sphyraena sp.
Paralichthys sp.
Sphoeroides sp.
Chilomycterus sp.

Cartilaginous Fishes
Blacktip shark
Sharpnose shark
Scalloped hammerhead
Bonnethead shark
Requiem sharks
Atlantic stingray
Skates and rays

Bony Fishes
Herring family
Hardhead catfish
Gafflopsail catfish
Sea catfishes
Needlefish family
Hardhead halfbeak
Halfbeak family
Killifish genus
Killifishes & halfbeaks
Sea robin
Sand perch
Sea bass
Sea bass family
.Jack family
Spotfin mojarra
Silver jenny
Grunt family
Spottail pinfish
Porgy family
Sand seatrout
Spotted seatrout
Sea trout
Northern kingfish


Modulus sp.
Batillaria minima
Cerithium lutosum
Crepidula aculeata
Strombus alatus
Polinices duplicatus
Phyllonotus pomum
Urosalpinx perrugata
Busycon sinistrum
Olivella sp.
Marginella sp.
Melongena corona
Nassarius vibex
Fasciolaria hunteria
Fasciolaria tulipa
Pyramidella crenulata
Turbonilla sp.
Anadara sp.
Noetia ponderosa
Brachidontes exustus
Modiolus americanus
Argopecten irradians
Plicatula gibbosa
Anomia simplex
Crassostrea virginica
Ostreola equestris
Lucina pectinata
Trachycardium egmontianum
Trachycardium muricatum
Dinocardium robustum
Chione cancellata
Mercenaria campechiensis
Anomalocardia auberiana
Macrocallista nimbosa
Callinectes sapidus
Menippe mercenaria
Balanus sp.

Bony fishes

False cerith
Dwarf cerith
Spiny slipper shell
Atlantic slipper shell
Fighting conch
Shark eye
Moon shell family
Apple murex
Gulf oyster drill
Lightning whelk
Dwarf olive
Crown conch
Common eastern nassa
Banded tulip
True tulip
Atlantic pyram
Ponderous ark
Scorched mussel
Tulip mussel
Pen shell family
Bay scallop
Kitten's paw
Common jingle shell
Atlantic oyster
Crested oyster
Thick lucina
Lucina family
Broad-ribbed cardita
Prickly cockle
Yellow cockle
Giant Atlantic cockle
Cross-barred venus
Hardshell clam
Pointed venus
Sunray venus
Blue crab
Stone crab
Sea urchin

it is clear from their size and morphology that they are not
from a sea turtle. It is possible that they might be from a
pond turtle. Gopher tortoise (Gopheruspolyphemus) and sea
turtle bones were recovered from other proveniences (Appen-
dix I).
Sharks and Rays. Cartilaginous fishes are well-repre-
sented. Four species of sharks representing five MNI were
recovered from the analyzed sample: blacktip, Atlantic
sharpnose, scalloped hammerhead, and bonnethead sharks.
One blacktip shark was adult and one was juvenile; only
teeth were recovered for this species. The sharpnose (teeth)
and the bonnethead shark (represented by one vertebral
centrum) were both adults, and the scalloped hammerhead
was a juvenile (teeth).


shark dermal denticles recovered from
the ash lens (Unit 8-2, 124-141 cmbd).
In the volumetric sample of 1 ml, all but
three of the 319 dermal denticles had
been burned.
Bony Fishes. Skeletal remains of 33
different species of bony fishes are pres-
ent in the analyzed sample. The great-
est MNI comes from the pinfish (Lago-
don rhomboides), with 35 MNI, and the
second greatest is from sea trout (Cyno-
scion sp.), with 31 MNI. Remains of 15
hardhead catfish (Galeichthysfelis) also
were recovered. Atlas widths of pinfish
remains indicate juvenile individuals
with standard lengths averaging 95 mm.
Sea trout remains are mostly of adult
individuals. Five additional species of
bony fishes were represented in the 6.4-
mm samples (Appendix I).
Molluscs. Molluscs were important
to the diet, but their remains (e.g.,
shells) were collected only during the
1991 excavations. Gastropods of 18
species and bivalves of 19 species were
recovered. Not all of these species rep-
resent intentional harvesting; some of
the animals were transported inciden-
tally to the site with the gathering of the
intended species. The incidental species
include ceriths, slipper shells, oyster
drills, barnacles, pyrams, and tur-

The Bullens' Excavations

Figure 3. Scanning-electron micrograph (250x) ofsharkderm
the ash lens in the Shell Ridge, Palmer site.

Remains from elsewhere in the profile excavation show
that 10 other species of sharks also were deposited in the
midden (Appendix I); therefore, a total of 14 species of
sharks was utilized. It should be noted that teeth of the
Atlantic sharpnose, the juvenile blacktip, and the scalloped
hammerhead were recovered from the 2-mm fraction;
therefore, three shark MNI came from the 2-mm fraction.
One rostral cartilage from a shark represents one individual.
Shark remains throughout the midden are presented in
greater detail in Appendix II. Remains of rays include the
Atlantic stingray from the analyzed sample, and the sawfish
and cownose ray from the 6.4-mm screened samples.
Numerous shark dermal denticles were found in the
analyzed sample. All denticles were recovered in screens
with mesh sizes of less than .5 mm. Some of them were
articulated, i.e., still interlocked as if embedded in shark
skin. Figure 3 is a scanning-electron micrograph of articu-
lated dermal denticles from the Palmer site. There were 319

al denticles from The Bullens systematically saved
only vertebrate faunal remains from
their excavation units. Although their
recovery methods were not up to today's standards, the
faunal remains from their excavations may still reveal
behavioral aspects of the site's inhabitants. For example,
shark and deer remains were common. Appendix IV lists
species recovered from the Bullens' excavations that were
not found in the faunal sample from the 1991 excavation. In
addition, I prepared a presence/absence list for the Bullens'
faunal material by test unit (Appendix V).

Habitat Information

Terrestrial Resources

Deer occupy a variety of habitats in search of edible
plants, although they prefer oak mast. They could have been
captured close to the site, although large concentrations of
humans tend to repel them. It is unlikely that they were
found on the barrier islands to the west of the site. There are

1S. ..u -..-------------------" '

8 991 Vot 51(4)



a J1003

a number of smaller "camp" sites to the east of the Palmer
site that may have functioned as ephemeral hunting camps
(Luer 1995:207-208, Figure 6). Two species of rabbit were
available near the site, the eastern cottontail (Sylvilagus
floridanus) and the marsh rabbit (S. palustris). Both are
nocturnal, but the eastern cottontail inhabits dry, open areas,
while the marsh rabbit lives in swamps, bottom lands, and
hammocks. Gopher tortoises inhabit sandy areas where they
can make their burrows, and can be found during the day
foraging on grass, leaves, and wild fruits.

Aquatic Resources

Most of the species utilized at the site, including turtles,
fishes, and molluscs, were from aquatic environments. Most
of the molluscs and fishes (except bowfin) can be found in
the seagrass bed community adjacent to the site (Zieman and
Zieman 1989). Bowfins (Amia calva), as well as pond
turtles, are found in freshwater environments. Sea turtles
could have been captured when they came to the beach to
mate or to lay eggs, although adults spend most of their time
at sea.
Sharks are carnivores at the top of the food chain, feeding
on other fishes and on invertebrates. The Atlantic sharpnose
shark is the most common shark occurring on the west coast
of Florida (Burgess et al. n.d.), so it is not surprising that it
is one of the sharks whose remains are present throughout
the midden (Appendix II). The sharpnose is a small shark
that attains a maximum length of 110 cm (Compagno
1984:533). The other commonly represented shark is the
bonnethead shark, which can grow to be 150 cm in length
(Compagno 1984:551). Sea trout spend most of their time
over the seagrass flats, and are considered to be top carni-
vores (Springer and Woodburn 1960; Zieman and Zieman
1989). The pinfish is the most abundant fish in seagrass
beds of the eastern Gulf of Mexico (Zieman and Zieman
1989:49). Pinfishes eat seagrasses or detritus and plant
debris, and the size class recovered in the analyzed sample
belong to the intermediate stage during which they are
benthic (Zieman and Zieman 1989:72). The spottail pinfish
(Diplodus holbrooki) is also a benthic dweller. Hardhead
catfish are scavengers and detritus feeders, frequently feeding
on the bottom.


Targeted Species

An analysis of species that were main contributors to the
meat diet is possible only from the materials excavated in
1991. It is obvious from Table II that aquatic species were
targeted for subsistence. Sharks, sea trout, hardhead catfish,
pinfish, scallops, and oysters were eaten regularly. Other
species were added to the diet as supplements.
Fishes. All of the fishes represented in the sample inhabit
near-shore environments. Adult sharks were probably
captured by hook and line, though smaller sharks could have

been netted. Sea trout, catfish, pinfish, and toadfish were
probably taken with a net, although they could have been
caught with hook and line. Some of the fishes represented
are small and could have been taken only by small-meshed
net. These include herrings, killifishes, mojarras andjennys,
halfbeaks, and needlefishes.
Materials from the Bullens' excavations reveal that
sheepshead, jackfish, blacknose shark, burrfish, seatrout,
grouper, snapper, mullet, toadfish, black drum, sharpnose
shark, redfish, and snook were captured regularly (Appendix
V). These are the larger fish species that were taken, and
should not (due to the bias for large bones in the Bullens'
samples) be interpreted as main contributors to the diet.
Hook and line probably were used for larger fishes, although
they could be caught occasionally in nets.
Bivalves. The presence of large amounts of scallop shells
hints that scallops were regularly captured for food. They
are filter feeders that live in shallow-water seagrass beds.
Oysters live in estuarine waters, and grow on solid or soft
substrates. The tulip mussel lives on algae- or moss-covered
rocks and corals (Rehder 1981). Hardshell clams live buried
in sandy or muddy substrates, but are not completely buried.
All of the cockles found in the sample also bury themselves
in the substrate (Rehder 1981). The sunray venus is a filter
feeder that buries itself in the sand.
Gastropods. The apple murex feeds principally on
oysters, and so is a mobile predator. The crown conch
scavenges on dead fish, crabs, and other molluscs, but also
attacks and feeds on living bivalves. The fighting conch is
herbivorous, feeding on algae, and can be found on sand or
among seagrass beds. They bury themselves during low tide
and suddenly emerge just as the tide begins to return (Rehder
1981). Tulip species prey on a variety of gastropods and
bivalves. The lightning whelk eats a variety of gastropods
and bivalves, including the hardshell clam, cross-barred
venus, shark eye, oysters, razor clam, and the sunray venus
(Magalhaes 1948; Paine 1963). Dead fish also are eaten by
the lightning whelk. The lightning whelk leaves distinct
marks on bivalve shells in the process of attacking and eating
the animal inside. Predators of the lightning whelk include
the stone crab and sea gulls, both of which make distinct
marks on whelk shells after feeding on the animal (Magal-
haes 1948).

Shark Utilization

The data in Appendix II show that sharks were captured
throughout the aboriginal occupation represented in the
profile excavation in the Shell Ridge, and that a number of
sharks were taken. Several articulated sections of shark
centra found in the profile are incontrovertible evidence for
shark capture, since shark carcasses rarely wash up on the
shore (Budker 1971; Lineaweaver and Backus 1984:23).
These data argue for regular capture of different species of
sharks, with emphasis placed on a few species.
Denticles were found from the analyzed provenience, as
well as the adjacent ash locus. The volumetric sample from

DATI UET- CwTWa Er*T1ai Dus1 N I


1TH ..I. .UA AN........ ... 1998 VOL 1

the ash locus contained 316 burned shark dermal denticles
out of a total of 319. Many of the denticles were burned in a
pattern that suggests chunks of meat with skin attached were
placed skin down on a fire, or perhaps were roasted over a
fire with the skin toward the flames. Perhaps shark skin was
being dried over a fire to prepare it to be used as shagreen
("sandpaper"). Another possibility is that sharks were
roasted whole.
The finding of denticles, along with cranial, caudal,
vertebral, and tooth elements, indicates that most of the
shark carcass, if not all of it, was brought back to the site to
be butchered. This information tends to discount the idea
that sharks were butchered on the beach. It is possible that
shark livers were boiled for their oil, because they contain oil
that could have been used in paints or in cooking (Kozuch
1991, 1993).
Shark tooth tools also were recovered. These consist of
one tiger shark tooth that was drilled in the root, one tiger
shark tooth that had a lateral cusp broken off, and the lower
tooth of a blacktip shark that had both lateral cusps filed to
expedite hafting for use as a drill or carving tool. Shark
teeth might have been used to carve wood or as saws. Shark
tooth drills have been demonstrated to be superior instru-
ments for drilling other shark teeth (Brown 1994:102).
Shark tooth weapons with tiger shark teeth for blades have
been recovered from the Key Marco site (Cushing 1973).
Year-round capture of sharks is indicated by the tempera-
ture preferences of certain species identified in the Shell
Ridge sample. Blacknose sharks are present off the south-
west Florida coast from March through November. The
bonnethead shark prefers the colder waters of winter, being
almost totally absent from the shallow coastal waters in
summer, but present during other times of the year. There-
fore, it is possible to infer that sharks were being caught all
year long.

Animal Bones from the Palmer Burial Mound

The analysis of animal remains included as offerings with
some of the human burials may shed light on which animals
had ritual or ceremonial significance. At least 400 individ-
ual human burials were uncovered by the Bullens (1976:46).
In the burial mound, as in their other tests at the Palmer site,
the Bullens probably missed some of the smaller faunal
bones. Nonetheless, the animal remains they found include
alligator, domestic dog, requiem shark, snook, whale
(fossilized), sea turtle, opossum, gopher tortoise, West Indian
monk seal, white-tailed deer, a probable osprey (Pandion
sp.), sea trout, and siren (Appendix V).
Of particular note is the West Indian monk seal, which is
now extinct. Seal bones are rare from Florida sites, and have
been recovered from South Indian Field in Brevard County,
the Grenada site in Dade County, Horr's Island in Collier
County, and the Wightman site in Lee County (Cumbaa
1980; FLMNH data cards for accession number 188; Wing
and Loucks 1984). It also is noteworthy that the only seal
bone from the Palmer site was in a human burial context.

One alligator and four dog burials were excavated from
the burial mound by the Bullens. None was accompanied
with artifacts. The alligator burial was in association with
two strings of sawfish (Pristis sp.) vertebrae. These verte-
brae are not in the zooarchaeology collections at FLMNH,
having been loaned to Historic Spanish Point in the mid-
1980s. Dogs, alligators, and seals apparently had some
social, ceremonial, or ideological significance. Bullen and
Bullen (1976:46, 51-52) suggest that the alligator might have
represented a clan totem. At present, there is not enough
evidence for or against any conclusion.


Aquatic resources dominate the assemblage. Terrestrial
animals were captured with much less regularity. The data
indicate that at least two and probably three methods were
used to capture fishes: 1) nets with small mesh sizes were
used to capture small schooling fishes, 2) nets with larger
mesh sizes were used to capture medium-sized fishes, and 3)
hooks and lines or weirs were used to capture sharks and
larger fishes. Bottom fishes are represented by the catfish,
spottail pinfish, intermediate-size pinfish, toadfish, and the
flounder, indicating that nets were designed to reach the
bottom to capture fishes swimming there.
Molluscs were probably gathered at low tide. People
gathered species that were numerous or large enough to
warrant gathering. Clams and cockles, including the
hardshell clam, could have been felt with the feet and dug
out of the substrate. Crown conchs, tulips, pear whelks, and
lightning whelks (being carnivores) are less common than
some of the bivalves, and were collected when found.
Oysters and tulip mussels could have been gathered from
places with hard substrates, such as mangrove roots, oyster
bars, and half-submerged dead trees.
An important factor in the capture of all estuarine species
is tidal movement. The people had to be able to predict and
plan for the tides. Shallow water makes tides particularly
significant, since normally submerged sand bars, mud flats,
and oyster reefs may be exposed to the air during low tide.
Nets and weirs are more effective when natural barriers, such
as sand bars, prevent fish from escaping, and tides are a
factor in determining where fish are located. Sharks gener-
ally come into the estuaries during high tide to prey on the
fishes present there. Molluscs are more easily seen and can
be more easily felt with the feet during low tide.


A sample of midden material, dating to early in the Safety
Harbor Period (ca. A.D. 1000-1200), was excavated in 1991
from near the western end of the Palmer site's Shell Ridge
midden. Faunal analysis reveals that the sample contains the
remains of five individual sharks representing four different
species. Remains of 170 individuals of bony fishes and 272
individual invertebrates also were recovered from the
analyzed sample.


1998 VOL- 514)



Analysis of faunal remains reveals that the people who
built and lived on the shell mounds at the Palmer site knew
how to make a living from the natural environment. They
made nets to capture pinfish and other fish, and used hooks
and lines or fish weirs to catch sharks and other large fishes.
People also gathered shellfish such as scallops, oysters, and
clams by wading in the shallow-water seagrass beds. Their
lives were intimately linked with tidal movements, and they
knew where and when fishing nets would be most effective.
Sharks were caught regularly year-round. The whole
carcass was brought back to the site, as evidenced by cranial,
caudal, and skin elements in the analyzed sample. Shark
skin may have been used as shagreen. Shark teeth, most
importantly from tiger and blacktip sharks, were used as
tools. The recovery of shark dermal denticles from the Shell
Ridge midden represents the first time this has been accom-
plished, and a method for recovery and recognition has been
This study of faunal remains from the Palmer site also
includes an analysis of materials excavated by the Bullens
over 35 years ago. Their recovery methods present some
problems in interpretation of the remains, but the problems
are surmountable, and important information still can be
gleaned from the Bullens' excavations. The Bullens'
material dates to the Late Archaic Period as well as to the
Manasota and early Safety Harbor periods. The combined
taxa represented in all the Bullens' material and the 1991
sample show that 14 species of shark, 44 species ofbony fish,
and 37 species of mollusc were brought back to the Palmer
site. They also show that deer, rabbit, turtle, opossum,
raccoon, loon, and duck also were eaten.
As voluminous as the faunal materials are from the
Palmer site, the site is so large as to make further investiga-
tions fruitful. This may include a search for tool-making
areas, suggested by shell debitage, or cooking areas, perhaps
suggested by burned dermal denticles. The change in
subsistence through time is another area in need of system-
atic and intensive analysis. The Palmer site is now pro-
tected, and perhaps may yet render evidence for the answers
we seek.


The faunal analysis of the 1991 excavation sample was financed in part with
historic preservation grant assistance provided by the Bureau of Historic
Preservation, Division of Historical Resources, Florida Department of State,
assisted by the Historic Preservation Advisory Council. A state Special
Category Grant, plus local assistance, helped fund "A Window to the Past"
Exhibition at Historic Spanish Point, which was a project of the Gulf Coast
Heritage Association, a non-profit organization that owns the Palmer site.


Special thanks go to William Marquardt at FLMNH, who understands the
importance ofzooarchaeology, and who was instrumental in enabling the 1991
faunal analysis to be performed. George Luer contributed excellent comments,
ideas, editing, and much of the impetus in getting the work published. F.
Morgan Wallace's expertise in scanning electron microscopy is greatly
appreciated. George Burgess of FLMNH has always been my source for shark
biological and life history information, and he provided his data on the


availability ofsharks offthe west coast of Florida. Elizabeth Wing at FLMNH
allowed me to work on the faunal remains, and always gives her support in my
zooarchaeological undertakings. This research was supported in part by a
National Science Foundation Grant (#SBR 9511302) awarded to Elizabeth
Wing for collection management. Any omissions or errors are solely the
responsibility ofthe author.

References Cited

Bargar, Tom W., and Thomas B. Thorson
1995 A Scanning Electron Microscopy Study ofthe Dermal Denticles ofthe
Bull Shark, Carcharhinus leucas. Journal of Aquariculture and
Aquatic Sciences 7:120-137.
Brown, Robin C.
1994 Florida's First People, 12,000 Years of Human History. Pineapple
Press, Sarasota.
Budker, Paul
1971 The Life of Sharks. Weidenfeld and Nicolson, London.
Bullen, Ripley P., and Adelaide K. Bullen
1976 The Palmer Site. Florida Anthropological Society Publication
Number 8, Gainesville.
Burgess, George H., Franklin F. Snelson, Stephen J. Walsh, Steven Clark,
Kevin G. Abbot, Larry E. Barton, and W. Steven Otwell
n.d. Biological Fishery and Product Assessments of the Keoghfish, an
Underutilized and Unmanaged Gulf ofMexico Resource. Manu-
script on file, Florida Museum of Natural History, Gainesville.
Compagno, Leonard, J. V.
1984 FAO Species Catalogue. Vol. 4. Sharks of the world. An Annotated
and Illustrated Catalogue of Shark Species Known to Date. Part 2.
Carcharhiniformes. FAO Fisheries Synopsis No. 125, Vol. 4, Part 2,
pp. 251-655.
Cumbaa, Stephen L.
1980 Aboriginal UseofMarine Mammals in the Southeastern United States.
Southeastern Archaeological Conference Bulletin 17:6-10.
Cushing, Frank Hamilton
1973 Exploration ofAncient Key Dwellers' Remains on the GulfCoast of
Florida. AMS Press, Inc., New York. Reprint of original 1896
Grayson, Donald K.
1979 On the Quantification of Vertebrate Archaeofaunas. InAdvances in
Archaeological Methodand Theory, Volume2, edited by Michael B.
Schiffer, pp. 200-238. Academic Press, New York.
Kozuch, Laura
1991 Use of Shark Products by Prehistoric Peoples in South Florida.
M.A. thesis, Department of Anthropology, University of Florida,
1993 Sharks and Shark Products in Prehistoric South Florida. Institute
ofArchaeology and Paleoenvironmental Studies, Monograph Number
2, University of Florida, Gainesville.
Lineaweaver, Thomas H., III, and Richard H. Backus
1984 The Natural History ofSharks. Nick Lyons Books, New York.
Luer, George M.
1995 The Brookside Mound, Sarasota County, Florida: Notes on Land-
scape, Settlement, Scrub Habitat, and Isolated Burial Mounds. The
Florida Anthropologist 48:200-216.
Luer, George M., and Marion M. Almy
1982 A Definition of the Manasota Culture. The Florida Anthropologist
Magalhaes, Hulda
1948 An Ecological Study of Snails of the Genus Busycon at Beaufort,
North Carolina. Ecological Monographs 18(3):377-409.
Moss, Melvin L.
1977 Skeletal Tissues in Sharks. American Zoology 17:335-342
Paine, Robert T.
1963 Trophic Relationships of Eight Sympatric Predatory Gastropods.
Ecology 44(1):63-73.
Raschi, W. and C. Tabit
1992 Functional Aspects of Placoid Scales: A Review and Update.
Australian Journal of Marine and Freshwater Research 43:123-
Rehder, Harald A.
1981 The Audubon Society Field Guide to North American Seashells.
Alfred A. Knopf New York.


1'R FL D .....O.OL. 9 O I

Springer, Victor G., and Kenneth D. Woodburn
1960 An Ecological Study of the Fishes of the Tampa Bay Area. Florida
Board Conservation, Professional Papers Series 1:1-104.
Wing, Elizabeth S., and Antoinette B. Brown
1979 Paleonutrition: Method and Theory in Prehistoric Foodways.
Academic Press, New York.
Wing, Elizabeth S., and L. Jill Loucks
1984 Grenada Site Faunal Analysis. In Excavations at the Grenada Site,
edited by John W. Griffin, pp. 259-345. Florida Division ofArchives,
History, and Records Management, Tallahassee.
Zieman, Joseph C., and Rita T. Zieman
1989 The Ecology ofthe Seagrass Meadows ofthe West Coast ofFlorida:
A Community Profile. United States Fish and Wildlife Service,
Biological Report 85(7.25).

Appendix I. Remains of the following vertebrate taxa were found
in midden material removed during the 1991 excavation but were
not represented in the limited sample that underwent detailed
analysis. These taxa are based on fairly large fragments that did
not pass through a 6.4-mm (1/4-in) mesh screen. The list combines
remains from two separate cultural components dating to the
Manasota Period (ca. A.D. 300) and the early Safety Harbor Period
(ca. A.D. 1100), with most originating from the latter component.

Taxon Common Name

Sylvilagus sp. Rabbit
Gavia immer Common loon
Anatidae Duck family
Viperidae Poisonous snake family
Serpentes Snake
Emydidae Pond turtle family
Gopherus polyphemus Gopher tortoise
Cheloniidae Sea turtle
Sirenidae Siren or Manatee
Ginglymostoma cirratum Nurse shark
Mustelus norrisi Florida dogfish
Odontaspis taurus Sand tiger shark
Carcharhinus acronotus Blacknose shark
Carcharhinus brevipinna Spinner shark
Carcharhinus leucas Bull shark
Carcharhinus plumbeus Sandbar shark
Galeocerdo cuvieri Tiger shark
Negaprion brevirostris Lemon shark
Sphyrna mokarran Great hammerhead
Pristis sp. Sawfish
Rhinoptera bonasus Cownose ray
Megalops atlanticus Tarpon
Mycteroperca sp. Grouper
Lopholatilus chamaeleonticeps Tilefish
Caranx hippos Crevalle jack
Archosargus probatocephalus Sheepshead

Appendix II. 1991 Excavations: Shark remains by level in the
profile, from the 6.4-mm (1/4-in) mesh screen. Remains from
above approximately 250 cmbd date to ca. A.D. 1100 (Safety
Harbor Period), whereas remains from below this depth date to ca.
A.D. 300 (Manasota Period).

Taxon Element

Carcharhinus acronotus
Carcharhinus brevipinna
Carcharhinus limbatus
Carcharhinus plumbeus
Rhizoprionodon terraenovae
Sphyrna tiburo

UNIT 3-1
126-146cmbd, 91-M-7
Rhizoprionodon terraenovae

146-164cmbd, 91-M-8
Carcharhinus plumbeus

UNIT 4-1
78-93cmbd, 91-M-10
Rhizoprionodon terraenovae

93-133cmbd, 91-M-11
Rhizoprionodon terraenovae

133-180cmbd, 91-M-12
Carcharhinus acronotus
Rhizoprionodon terraenovae
Sphyrna tiburo

215cmbd, 91-M-18
Carcharhinus plumbeus

219-234cmbd, 91-M-40
Carcharhinus acronotus

UNIT 5-1
80-136cmbd, 91-M-19
Carcharhinus plumbeus
Rhizoprionodon terraenovae
Sphyrna tiburo

UNIT 6-1
49-106cmbd. 91-M-22
Carcharhinus limbatus
Sphyrna lewini

180-189cmbd, 91-M-62
Carcharhinus acronotus
Rhizoprionodon terraenovae

187-192cmbd, 91-M-60
Carcharhinus sp.
Sphyrna tiburo

192-204cmbd, 91-M-63
Carcharhinus limbatus
Rhizoprionodon terraenovae

UNIT 6-2
54-56cmbd, 91-M-24
Carcharhinus leucas
Sphyrna lewini



lower tooth



rostral cartilage


articulated centra



articulated centra


upper tooth



0-lOcmbd, 91-M-1
Odontaspis taurus

tooth & centra

56-87cmbd, 91-M-23
Carcharhinus limbatus
Carcharhinus plumbeus

lower tooth & centra


1998 VOL 51(4


Appendix II (Continued).

Taxon Element

Sphyrna lewini


Sphyrna sp. (juvenile)

12-39cmbd, 91-M-37
Carcharhinus plumbeus
Sphyrna tiburo

85-106cmbd, 91-M-25
Carcharhinus leucas
Rhizoprionodon terraenovae

Sphyrna lewini
Sphyrna tiburo

106-132cmbd, 91-M-26
Carcharhinus limbatus
Rhizoprionodon terraenovae
Sphyrna tiburo

130-160cmbd, 91-M-27
Mustelus cf. norrisi
Carcharhinus plumbeus
Carcharhinus cf. obscurus

158-181cmbd, 91-M-28
Rhizoprionodon terraenovae

UNIT 7-1
Rhizoprionodon terraenovae

UNIT 7-2
76-110cmbd, 91-M-32
Carcharhinus brevipinna
Sphyrna lewini
Sphyrna tiburo

110-155cmbd, 91-M-35
Ginglymostoma cirratum
Mustelus cf. norrisi
Carcharhinus plumbeus
Sphyrna tiburo

156-180cmbd, 91-M-36
Carcharhinus acronotus
Rhizoprionodon terraenovae
Sphyrna mokarran
Sphyrna tiburo

UNIT 8-1
173-181cmbd, 91-M-99
Carcharhinus limbatus
Carcharhinus plumbeus
Rhizoprionodon terraenovae
Ginglymostoma cirratum

186-206cmbd, 91-M-l 10
Rhizoprionodon terraenovae
Sphyrna tiburo

186cmbd, 91-M-102, Area I
Carcharhinus acronotus
Carcharhinus leucas

rostral cartilage &

upper tooth & centra






lower tooth

lower tooth & centra



39-50cmbd, 91-M-43
Rhizoprionodon terraenovae

50-79cmbd, 91-M-44

75-99cmbd, 91-M-47
Sphyrna tiburo

81-100cmbd, 91-M-48
Sphyrna tiburo

100-117cmbd, 91-M-49
Carcharhinus limbatus
Sphyrna tiburo

109-135cmbd, 91-M-54
Carcharhinus acronotus
Carcharhinus limbatus
Rhizoprionodon terraenovae

Sphyrna tiburo

118-130cmbd, 91-M-51
Rhizoprionodon terraenovae

130-136cmbd, Zooarch. Detailed Analysis
Carcharhinus limbatus
Rhizoprionodon terraenovae
Sphyrna lewini
Sphyrna tiburo

136-150cmbd, 91-M-64
Carcharhinus limbatus
Sphyrna tiburo

138-168cmbd. 91-M-94
Sphyrna tiburo

UNIT 20-1
320-330cmbd, 91-M-66
Rhi:oprionodon terraenovae

Carcharhinus acronotus
Carcharhinus limbatus

UNIT 21-1
320-330cmbd, 91-M-80
Carcharhinus limbatus
Rhizoprionodon terraenovae

350-360cmbd, 91-M-I I
Carcharhinus limbatus
Carcharhinus plumbeus


caudal centra




rostral cartilage &










UNIT 8-2


Kozuc 108





Appendix II (Continued).

Taxon Element

Rhizoprionodon terraenovae

UNIT 21-2
Carcharhinus acronotus
Rhizoprionodon terraenovae

340cmbd, 91-M-114
Carcharhinus limbatus

Rhizoprionodon terraenovae

340cmbd, 91-M-108
Rhizoprionodon terraenovae




157-310cmbs, under tree, Unit 21, 91-M-3

Carcharhinus sp.
Carcharhinus limbatus
Negaprion brevirostris
Rhizoprionodon terraenovae
Sphyrna lewini
Sphyrna tiburo



upper tooth & centra

Appendix III. Detailed faunal analysis of sample collected from Unit 8-2-2, 130cmbd, in 1991 profile excavation (smallest
mesh size is 2mm). MNI = minimum number of individuals. NISP = number of individual specimens.

Skeletal Weight
Taxon MNI Percent NISP Percent Peletal percent
in gm

Odocoileus virginianus

Total Tetrapoda

2 100.00 12 100.00

18.37 100.00

Carcharhinus limbatus
Rhizoprionodon terraenovae
Sphyrna lewini
Sphyrna tiburo
Dasyatis sabina
Amia calva
Galeichthys felis
Bagre marinus
Opsanus sp.
Belonidae cf. Strongylura
Chriodorus atherinoides
Fundulus sp.
Diplectrum formosum
Epinephelus sp.
Caranx sp.
Lutjanus sp.
Eucinostomus argenteus
Eucinostomus gula
Eucinostomus sp.
Orthopristis chrysoptera
Haemulon sp.
Diplodus holbrooki
Lagodon rhomboides
Bairdiella sp.







1998 VOL. 51(4)


Taxon MNI Percent NISP Percent Skeletal Weight Percent
in gm
Cynoscion arenarius 19 10.86 36 1.17 10.90 7.56
Cynoscion nebulosus 8 4.57 12 .39 4.04 2.80
Cynoscion sp. 4 2.29 19 .62 2.89 2.01
Leiostomus xanthurus 3 1.71 3 .10 .02 .01
Menticirrhus littoralis 2 1.14 2 .06 .11 .08
Menticirrhus saxatilis 1 .57 1 .03 .08 .06
Sciaenops ocellata 1 .57 2 .06 1.44 1.00
Mugil sp. 4 2.29 21 .68 1.00 .69
Sphyraena sp. 1 .57 7 .23 .05 .03
Paralichthys sp. 1 .57 2 .06 .01 .01
Sphoeroides sp. 1 .57 1 .03 .01 .01
Chilomycterus sp. 5 2.86 32 1.04 .47 .33
Osteichthyes 0 .00 2145 69.48 88.46 61.38

Total Fishes 175 100.00 3087 100.00 144.13 100.00

Vertebrata 0 .00 11 .00 1.75 .00

Modulus sp. 1 .37 1 .13 .18 .01
Batillaria minima 1 .37 1 .13 .05 .00
Cerithium lutosum 37 13.60 37 4.82 2.61 .19
Strombus alatus 2 .74 13 1.69 76.44 5.57
Crepidula aculeata 1 .37 1 .13 .98 .07
Crepidulafornicata 1 .37 1 .13 .53 .04
Crepidula sp. 59 21.69 59 7.68 1.78 .13
Polinices duplicatus 1 .37 1 .13 5.13 .37
Naticidae 1 .37 1 .13 .02 .00
Phyllonotus pomum 2 .74 21 2.73 30.50 2.22
Urosalpinx perrugata 5 1.84 6 .78 1.00 .07
Busycon sinistrum 1 .37 1 .13 9.58 .70
Olivella sp. 5 1.84 5 .65 .20 .01
Marginella sp. 3 1.10 3 ..39 .40 .03
Pyramidella crenulata 1 .37 1 .13 .20 .01
Turbonilla sp. 4 1.47 4 .52 .09 .01
Melongena corona 5 1.84 6 .78 16.96 1.24
Nassarius vibex 2 .74 2 .26 .26 .02
Fasciolaria hunteria 5 1.84 5 .65 13.35 .97
Fasciolaria tulipa 1 .37 1 .13 2.88 .21
Fasciolaria sp. 0 .00 6 .78 7.18 .52
Gastropoda 0 .00 0 .00 41.06 2.99
Anadara sp. 1 .37 2 .26 .01 .00
Noetia ponderosa 4 1.47 6 .78 9.52 .69
Brachidontes exustus 7 2.57 43 5.60 3.83 .28
Modiolus americanus 2 .74 32 4.17 13.58 .99
Mytilidae 10 3.68 39 5.08 4.90 .36
Pinnidae 1 .37 10 1.30 1.09 .08
Argopecten irradians 24 8.82 113 14.71 302.33 22.03
Plicatula gibbosa 1 .37 1 .13 .61 .04
Anomia simplex 17 6.25 20 2.60 4.00 .29
Crassostrea virginica 12 4.41 47 6.12 33.87 2.47
Ostreola equestris 12 4.41 18 2.34 3.67 .27
Lucina pectinata 1 .37 1 .13 .07 .01
Lucinidae 1 .37 1 .13 .05 .00
Carditamerafloridana 8 2.94 15 1.95 7.19 .52
Trachycardium egmontianum 7 2.57 13 1.69 49.12 3.58
Trachycardium muricatum 1 .37 1 .13 3.16 .23
Trachycardium sp. 1 .37 64 8.33 46.29 3.37
Dinocardium robustum 1 .37 18 2.34 72.24 5.26
Chione cancellata 4 1.47 10 1.30 7.96 .58
Mercenaria campechiensis 4 1.47 49 6.38 239.34 17.44



Taxon MNI Percent NISP Percent Skeletal Weight Percent
in m

Anomalocardia auberiana 5 1.84 9 1.17 1.78 .13
Macrocallista nimbosa 2 .74 20 2.60 52.36 3.82
Bivalvia 0 .00 0 .00 126.29 9.20
Mollusca 0 .00 0 .00 160.43 11.69
Menippe mercenaria 1 .37 7 .91 2.96 .22
Callinectes sp. 3 1.10 8 1.04 1.85 .13
Brachyura 2 .74 42 5.47 6.11 .45
Balanus sp. 1 .37 1 .13 6.37 .46
Echinoidea 1 .37 2 .26 .02 .00

Total Invertebrata 272 100.00 768 100.00 1372.38 100.0

Appendix IV. Additions to the species list for the Palmer site
based on faunal remains from the Bullens' excavations. These
faunal materials are from 10 different test units in the Hill Cottage
Midden, the Shell Midden, and the Shell Ridge, plus excavations
in the Palmer Burial Mound, The materials date to the Late
Archaic, Manasota, and early Safety Harbor periods.

Taxon Common Name

Didelphis virginiana
Lynx rufus
Monachus tropicalis

Neofiber alleni
Neotoma floridana
Procyon lotor
Sciurus cf. carolinensis
Urocyon cinereoargenteus

Anas sp.
Pandion sp.
Phalacrocorax auritus

ReptAgkistrodon piscivorus
Agkistrodon piscivorus

dog/coyote/wolf family
domestic dog
whale order
dolphin family
West Indian monk seal
(now extinct)
round-tailed muskrat
eastern woodrat
gray fox

duck genus
eagle/osprey genus
double-crested cormorant

cottonmouth snake

Alligator mississippiensis
Chelydra serpentina
Deirochelys reticularia
Drymarchon sp.
Farancia sp.
Kinosternon sp.
Natrix sp.
Pseudemys sp.

Terrapene carolina

Bufo marinus
Siren lacertina

Cartilaginous Fishes
Aetobatis narinari
Carcharhinus obscurus
Pristis sp.

Bony Fishes
Balistes sp.
Calamus sp.
Caranx crysos
Centropomus sp.
Micropterus salmoides
Pogonias cromis

American alligator
snapping turtle
chicken turtle
indigo snake genus
mud/rainbow snake genus
mud turtle genus
water snake genus
pond/cooter/slider turtle
box turtle
softshell turtle

giant toad
greater siren

spotted eagle ray
dusky shark

triggerfish genus
porgy genus
blue runner
largemouth bass
black drum
mackerel family

Appendix V. Presence/absence list by provenience from Bullen excavations (no European-derived domesticates included).
The ages of the test units are discussed in the text. NISP = number of individual specimens.

Test Units Burial
Taxon NISP
A B C D E F G H I J Mound

Aetobatis narinari
Agkistrodon piscivorus
Alligator mississippiensis
Amia calva
Anas sp.
Archosargus probatocephalus
Ardea herodias

14 X


9 X X
2 X
1 X
172 X X X X X X X X X
1 X

8 991 VoL 51(4)


o N P Test Units Burial
A B C D E F G H I J Mound

Balistes sp.
Bufo marinus
Calamus sp.
Canis sp.
Caranx crysos
Caranx hippos
Caranx sp.
Carcharhinus cf. acronotus
Carcharhinus cf. brevipinna
Carcharhinus cf. limbatus
Carcharhinus acronotus
Carcharhinus brevipinna
Carcharhinus limbatus
Carcharhinus obscurus
Carcharhinus plumbeus
Carcharhinus sp.
Centropomis sp.
Chelydra serpentina
Chilomycterus schoepfi
Chrysemys sp.
Cynoscion sp.
Dasyatis sp.
Deirochelys reticularia
Didelphis virginiana
Drymarchon sp.
Epinephelus sp.
Farancia sp.
Galeocerdo cuvieri
Gavia immer
Ginglymostoma cirratum
Gopherus polyphemus
Haemulon sp.
Kinosternon sp.
Lutjanus sp.
Lynx rufus
Malaclemys terrapin
Megalops atlanticus
Micropterus salmoides
Monachus tropicalis
Mugil sp.
Mycteroperca sp.
Natrix sp.
Negaprion brevirostris
Neofiber alleni

1 X
4 X
1 X
2 X
2 X
1 X
5 X
21 X
17 X
11 X
73 X
1 X
6 X
13 X
33 X
3 X
20 X
74 X
4 X
5 X
88 X
32 X
15 X
10 X
28 X
14 X
1 X
54 X
2 X
3 X
6 X
2 X
3 X
14 X
178 X
18 X
2 X
6 X
10 X
7 X
2 X
251 X

3 X


x x

x X




x x x x x x
x x

x x x x x x
x x


x x

x x


X X x x X X




X X x x X

x x x x x x x



x x x x



X X x x


D *u n CS-w^ F I.* -w

Irr ....................... ----

Test Units Burial
Taxon NISP AH Mound
A B C D E F G H I J Mound

Neotoma floridana
Odocoileus virginianus
Odontaspis taurus
Opsanus sp.
Pandion sp.
Paralichthys sp.
Phalacrocorax auritus
Phalacrocorax sp.
Pogonias cromis
Pristis sp.
Procyon lotor
Pseudemys sp.
Rhizoprionodon terraenovae
Sciaenops ocellatus
Sciurus cf. carolinensis
Siren lacertina
Sphyrna cf. lewini
Sphyrna lewini
Sphyrna mokarran
Sphyrna tiburo
Sylvilagus sp.
Terrapene carolina
[Iroevon cinereoarpenteus

1 X
365 X
31 X
2 X
3 X
1 X
16 X
21 X
8 X
12 X
108 X
78 X
32 X
22 X
9 X
273 X
13 X
2 X
26 X
36 X
7 X
1 X

X X x x X

xx x

x x

x x x x

x x
x x




X X x x



x x




X X x x





1998 VOL 51 (4)




Environmental Archaeology Laboratory, Florida Museum of Natural History, Gainesville, Florida 32611
E-mail: quitmyer@flmnh.ufl.edu

he shell middens of the southeastern coast of the
United States contain the subsistence remains of
precolumbian, maritime people. Such deposits range
in density from small shell scatters, containing the remains
of only a few animal species, to places like the Palmer site
(8S02), Sarasota, Florida, which is meters thick and covers
several hectares. These sites contain the remains of hun-
dreds of animal taxa that were collected by the early resi-
dents of Florida's coast. Like tree rings, a record of the life
histories of many of these marine organisms is recorded in
their skeletons (Hudson et al. 1976). The ability to read this
record can provide information about precolumbian human
ecology (Quitmyer et al. 1997). This record is particularly
important because of its taxonomic diversity. Humans are
broad-based omnivores that may target animal resources at
specific times during an annual cycle. This may occur when
animals attain their maximum seasonal biomass and or when
they form large seasonal aggregates. A single taxon there-
fore, does not provide the fullest picture of the seasonal
round of subsistence.
An understanding of human ecology and cultural com-
plexity requires an understanding of when and where people
lived through the annual cycle of life. Were the pre-
columbian people of the Gulf coast of Florida fully adapted
to a maritime existence or did they seasonally abandon
coastal sites for resources in the interior? In the archaeolog-
ical literature, the time of the year when people occupied an
archaeological site is referred to as "seasonality." Many of
the animals that were harvested by the prehistoric inhabitants
of the Palmer site can provide information about when and
where various resources were collected. This is one of the
underpinnings of estimating archaeological settlement
The goal of this research is to provide information about
the seasonality of two different cultural components in the
Shell Ridge midden, one dating to the mid-Manasota Period
(ca. A.D. 300) and the other dating to the early Safety
Harbor Period (ca. A.D. 1100). This will be done through
the analysis of the shells of three molluscan species that were
recovered during excavation. These species are: the bay
scallop (Argopecten irradians concentricus), the eastern
oyster (Crassostrea virginica), and the quahog (Mercenaria
sp.). The identification of the common loon (Gavia immer),

a migratory waterfowl, also provides information about site
seasonality. The preferred habitats ofthese animals also help
to identify the places that precolumbian people used in the
food quest.

Materials and Methods


The excavation of Shell Ridge at the Palmer site was
conducted by Corbett McP. Torrence (1991a, 199 Ib), Florida
Museum of Natural History, University of Florida in August
of 1991 and is titled "A Window to the Past" The excava-
tion was near the locations of Tests H and I where Edward
M. Dolan, Florida State University, excavated over 30 years
earlier (Bullen and Bullen 1976:4,22; see also Kozuch, this
issue, Figure 1). The purpose of "A Window to the Past"
was to expose a profile of the midden and preserve this
stratigraphic section for public interpretation and education.
Laboratory analysis of the animal and plant remains recov-
ered in this excavation was conducted at the Environmental
Archaeology Laboratory, Florida Museum ofNatural History
(see papers by Kozuch and Newsom, this issue) under the
direction of William H. Marquardt, Southwest Florida
The "Window to the Past" excavation exposed approxi-
mately a 3-m vertical profile of the midden. There appear to
be eight major strata representing a variety of activities at
this location. These major strata are referred to here as
"macro levels" and each is composed of a series of complex
lenses of crushed shell, bone, and botanical remains (Tor-
rence 1991a, 1991b; Figure 1). Macro levels IA, IB, IC, ID,
IE, IIA, IIB, and III belong to the early Safety Harbor Period,
ca. A.D. 1100. Macro level IV belongs to the mid-Manasota
Period, ca. A.D. 300.

Analytical Techniques

Three analytical methods were used to estimate the
seasonality of the Shell Ridge midden: the presence of
migratory waterfowl, the size-class composition of the shells
of the bay scallop (Russo and Quitmyer 1996), and the
analysis of the incremental growth structures found in the



VOL. 51 NO. 4

4 in. r6 wIIU~A Ani IRflkCIAfGO iV- C1A

3 4 5 6 7 8 9
0.0 -0.0

0.5 05

"'" (. *" .. .


Figure 1. Excavation profile illustrating the locations of the macro levels identified in
the "Window to the Past" excavation, Palmer site's Shell Ridge Midden (8S02).
Macro levels IA, IB, C, ID, IE, A, and represent the Safety Harbor Period (ca.

A.D. 1100), while Macro level IV is associated with the Manasota Period (ca. A.D.

2.0 ? .. L :.."'. '.' l ..! I:::: 2.0

2.5 .C'? :" 2.5
3.0 L.'.- k '3.0
S ,. ""IV "
3..5 3.5
0 1.0 2.0

Figure 1. Excavation profile illustrating the locations of the macro levels identified in
the "Window to the Past" excavation, Palmer site's Shell Ridge Midden (8SO2).
Macro levels IA, IB, IC, ID, IE, HA, and M represent the Safety Harbor Period (ca.
A.D. 1100), while Macro level IV is associated with the Manasota Period (ca. A.D.

shells of quahogs (Quitmyer et al. 1997). A morphometric
analysis of the shells of the eastern oyster was conducted to
determine what habitats were exploited for this resource
(Kent 1988). The application of these techniques requires
the presence of contemporary studies that serve as a compar-
ative series for the archaeological materials (Kent 1988;
Quitmyer et al. 1997; Russo and Quitmyer 1996).

Presence ofAnimal Resources as Seasonal Indicators

The presence of migratory animal species is one of the
earliest and most commonly applied methods for estimating
the seasonality of archaeological sites. Common loon was
routinely identified during the course of processing the
faunal materials from the "Window to the Past" excavation
and the Bullens' excavations (Table 1). This animal is
known to winter in southern Florida from October to May
(Howell 1932:73).

Annual Shell Size Composition of Bay Scallops as a Sea-
sonal Indicator

Today along the Florida Gulf coast the bay scallop is
collected recreationally during the summer season. Many
have thought that the bay scallop migrates into the shallows
of the estuary during the summer months, then disappears
during the winter. Empirical data indicate that there is no
evidence for the migration of this species (Broom 1976:20;
Winter and Hamilton 1985:241). Researchers have shown
that the bay scallop is a permanent resident of the grassy
shallows. The apparent disappearance of the animal in latest
fall/earliest winter is the result of a natural change in size-
class composition and mass mortality at the end of the
annual cycle (Barber and Blake 1983; Broom 1976). This

can be illustrated by examining part of the annual life history
of bay scallops from the Anclote estuary near Tarpon
Springs, Florida (Barber and Blake 1983).
The growth and reproductive cycle of bay scallops found
in the Anclote estuary commences during the winter when
water temperature is at a minimum. Juvenile scallops have
a winter mean shell height of less than 20 mm (Figures 2 and
3). By the spring quarter, the young animals begin to
increase rapidly in size until they reach their maximum
dimension in September; from May to September the mean
shell height increases from 39.9 to 60 mm (Barber and Blake
1983:248, Figure 3). This increase tracks increasing water
temperature (Barber and Blake 1983:252). Availability of
food also has been implicated in the development of the
animal (Broom 1976:13-19).
Spawning occurs in October (mean shell height > 60 mm)
as the water temperature starts to decline (Figure 3). Just
prior to spawning (August), the scallop adductor muscle
undergoes a significant reduction in mass. This is probably
related to the depletion of energy reserves during spawning
and may be a contributing factor to a mass mortality that
occurs in late November. It is at this time that the scallop
size-class composition quickly decreases as most ofthe adults
die and make way for the coming year's progeny (Barber and
Blake 1983:248, 252).
Latitudinal trends in the period of reproduction and
possibly size-class composition have been noted (Barber and
Blake 1983:252). The time of reproduction in southern
populations of bay scallops is earlier in the season when
compared to northern populations. This is probably related
to temperature increases, or factors related to temperature,
that occur earlier in southern latitudes. Due to its geographic
proximity (115 km) to the Palmer site, the Anclote scallop
shell size-class data correlated with the seasons of the year

1orn I7f: SA\


OUImYKR ~1~A~flWAIITV AT TUE aaa~ ~' I,:,

were used to interpret the season
that the zooarchaeological scallops
were collected.
Application ofthe Anclote scallop
data to archaeological assemblages
was first attempted by Russo (Austin
and Russo 1989:55-65, Russo and
Quitmyer 1996). In Russo's review
of the Anclote scallop data, he notes
that only the mean shell height sta-
tistic was reported for each of the
sampled months (Barber and Blake
1983). This is somewhat problem-
atic because the variation in shell
height around the mean cannot be
determined. If smaller individuals,
for example, were ignored in the
prehistoric collection strategy, there
is a chance of skewing the mean
shell height upward for any one
month, and thus causing an inaccu-
rate inference regarding the season
ofharvest. Russo suggests that com-
paring the modal peaks of the con-
temporary series with the archaeo-
logical series may be more accurate,
since the mode would include the
most commonly occurring size class-
es available for harvesting during
any one month. However, after

Table 1. The occurrence of the common loon (Gavia immer) at the Palmer site (8SO2).

Cat. No. Unit Levela Skeletal Element Archaeological
91-M-003 -- Surface Tarsometarus, left distal Early Safety Harbor
91-M-027 06-02 140 Coracoid Early Safety Harbor
91-M-035 07-02 110-155 Cervical vertebra Early Safety Harbor
91-M-036 07-02 156 Phalanx Early Safety Harbor
91-M-099 08-01 173-181 Femur, right Early Safety Harbor
91-M-043 08-02 39-50 Tibiotarsus, right distal Early Safety Harbor
91-M-104 08-02 166 Femur, left Early Safety Harbor
91-M-044 01-02 50 Coracoid, right Early Safety Harbor
91-M-044 01-02 50 Scapula, right Early Safety Harbor
91-M-064 08-02 136-150 Sternum Early Safety Harbor
91-M-098 21-02 313-340 Quadrate Manasota
Bullen and Bullen 1976
95681 Test E 36-42 Ulna, right proximal Manasota
95693 Test F 60-66 Cervical vertebra Manasota
95695 Test F 72-78 Humerus, right proximal Manasota
95696 Test F 78-84 Coracoid, left proximal Manasota
95699 Test F 108-120 Femur, right shaft Manasota
95705 Test 30-36 Humerus, right proximal Manasota

SLevel depths are in centimeters below surface except for Bullen and Bullen (1976) which
were recorded in inches below surface.

examining size-class

composition data from North Carolina, where sample size
and range of the number of specimens in each size class are
available, Russo concluded that mean shell height data may
be used in lieu of modal shell-growth patterns (Austin and
Russo 1989:63; Broom 1976:26-27). In that study the
removal of outliers from around the mean shell height did
not seem to hamper the ability to predict the season of death
of scallops using the mean size-class data (Austin and Russo
1989:63). Notwithstanding, the effects of removing outliers
on the interpretation of the data may ultimately depend on
the size classes and number of outliers removed.

Figure 2. Measurement of shell height taken on the shell
of the bay scallop (Argopecten irradians concentricus).

Samples of zooarchaeological scallop shells (1143) were
collected from three macro levels (IA, IB, IIA) of the Shell
Ridge excavation (Table 2, Figures 1 and 3). This represents
14 samples from Units 8-1 and 8-2 that date to the early
Safety Harbor Period. A measurement of the greatest
distance from the shell umbo to the ventral margin (shell
height in mm) of the right valve was made (Abbott 1974:11;
see Figure 2). The mean of the shell height and the maxi-
mum range were compared to the data collected from the
Anclote scallop series (Table 2, Figure3).

Quahog Incremental Shell-Growth Analysis as a Seasonal

The shells of many of the marine mollusca form seasonal
shell-growth increments through the annual cycle. These
increments, like tree rings, accurately record seasonal
environmental changes that occur during the course of
ontogeny. Quahog shells grow by accretion along the shell
margin and periodically form an alternating pattern of light
and dark increments (Figure 4) (Quitmyer et al. 1997).
These increments represent seasonal changes in shell
chemistry, microstructure, and growth rate (Jones and
Quitmyer 1996; Quitmyer and Jones 1992; Quitmyer et al.
The ability to correlate the annual cycle of incremental
shell-growth formation with the seasons of the year results
from a longitudinal study of living quahogs (n = 399) from


wAw NAi ITV ArT TrV PAI UPu "rr-

THE FtonmA Ajun fuVo~oLo2isT *77O Vtl. %,jI

Table 2. Mean shell heights, sample sizes, and ranges of shell heights of bay
scallop (Argopecten iradians concentricus) excavated from the Palmer site's
Shell Ridge Midden (8SO2).
Provenience Sample Mean Shell
(Unit, cmbs, S e en he Minimum Maximum
8-2, 12, IA 166 41.5 29.8 58.2
8-2, 39-50, IA 71 42.4 25.2 56.8
8-2, 60, IA 377 39.8 24.8 59.6
8-2, 75-99, IB 71 38.3 26.4 55.2
8-2, 90-100, IB 35 39.1 31.9 49.2
8-2, 100-117, IB 71 40.0 25.1 60.1
8-2, 100-135, B1 35 41.2 30.4 58.4
8-2, 138, IB 23 40.5 31.6 58.0
8-2, 135-160, IB 130 39.1 17.3 58.9
8-2, 166, IB 23 41.4 31.6 58.0
8-1, 173-181, IA 47 39.9 28.6 60.1
8-1, 173-181, IA 25 43.5 30.5 43.5
8-1, 186, IIA 47 43.4 28.7 61.1
8-1, 181-200, IIA 22 41.8 28.7 56.8
Total Sample 1143

two localities in Charlotte Harbor, Florida (Quit-
myer and Jones 1992). The observed periodicity
of incremental shell formation shows that the
Charlotte Harbor quahog sample sizes are large
enough to account for variability in the periodicity
of incremental shell formation that may exist
among individual quahogs within a population,
between years, and geographic location. These
data were therefore used to interpret the incre-
mental shell growth structures of archaeological
quahog shells excavated from the Shell Ridge
The incremental growth structures of quahogs
are best exposed by radially cross-sectioning the
shell along the greatest growth axis from the
umbo to the ventral margin with a water-cooled
lapidary saw equipped with a thin diamond-
impregnated blade (Figure 4). In cross-section,
the alternating dark and light shell-growth incre-
ments can be examined microscopically or with
the unaided eye.
The shell microstructure ofthe light-colored or
opaque increment is more porous, having less
tightly packed crystallites, and reflects light when
viewed in thin section with transmitted light.

Anclote Bay, Florida



.S 40

0 20







Figure 3. Seasonal change in the mean shell heights (mm) of bay scallops (Argopecten irradians
concentricus) collected from the Anclote estuary, Florida (Barber and Blake 1983). No sample size was
reported. Solid square boxes represent modern size-class data while open circles represent the mean
sizes of the zooarchaeological scallops.


1998 Vrl .1.\


This light-colored increment represents a
period of rapid shell growth and is usually
wider than the dark-colored or translucent
increment. In contrast, the translucent
increment is composed of well organized,
tightly packed crystallites that transmit
light when viewed in thin section. This
increment represents a period of slow shell
formation and is usually narrower than the
light-colored increment.
An annual growth cycle in Mercenaria
is composed of one opaque and one trans-
lucent increment. A count of these incre-
ments yields the age of the animal at the
time of its death. Individual specimens of
Mercenaria can be comparatively long-
lived, often reaching ages of 30-35 years of
age (Quitmyer and Jones 1992).
Temporal control of the alternating
opaque and translucent growth increments
was established by evaluating the incre-
ments found in the cross-sectioned quahog
shells with a convenient, six-part subdivi-
sion of the annual shell growth (Figure 5).
Each subdivision is referred to as a growth
phase. The formation of the translucent
growth increment was divided into three
growth phases: Translucent 1 (T) trans-
lucent growth just starting; Translucent 2
(T2)- translucent increment approximately
one-half the size of the previous year's
translucent increment; and Translucent 3
(T3) translucent growth is equal to or
greater than the previous translucent incre- central
ment. The formation of the opaque growth margin
increment was divided similarly into three
growth phases (01, 02, and 03).
The use of the shell-growth phases to
maintain temporal control recognizes that
growth is an exponential process through Figure 4. The p(
ontogeny and highly seasonal in nature, and the exposed a
Furthermore, there is some individual and Jones 1992).
variation among quahogs living in the
same bed or in different habitats (Jones et al. 1992). Even
though natural variability exists in quahog populations, the
six-part subdivision of shell growth describes the annual
progression of shell formation in the population. This is
accomplished by calculating the growth-frequency profiles
for each season and for the annual cycle for the population
(Figure 6).
A growth-frequency profile is constructed by determining
the percentage of the frequency of individuals in each of the
six growth phases and presenting the results in a histogram
(Figures 7). Growth-frequency profiles were constructed for
the contemporary quahogs collected from two separate
habitats in Charlotte Harbor: one at Catfish Creek in Char-
lotte County and another near Bokeelia in Lee County

Translucent increment
Opaque increment Annual growth cycle

Outer shell layer
Middle shell layer
Inner shell layer


sition of the radial cut of a quahog shell (Mercenaria sp.)
alternating incremental shell-growth structures (Quitmyer

(Quitmyer and Jones 1992: Figure 8). Seasonal and annual
growth-frequency profiles were observed to be rather discrete
and provide the best fit for assessing quahogs from archaeo-
logical contexts at the Palmer site.
The growth-frequency profiles of quahogs collected from
Charlotte Harbor proceed in the following manner through
the seasons of the year (Figure 6): During the winter quarter
(December through February), the translucent 3 (T3) growth
phase dominates the Charlotte Harbor growth frequency
profile. Growth phase TI is absent, while less than 8% of
the sample contains T2 shell growth. Opaque 1 (01)
represents the emergence of opaque shell growth in the
population. By the spring quarter (March through May), the
translucent 3 (T3) growth phase has substantially declined;




TsnE FLoREDA ANwuwnnPn, 1disT 1002v Vm 4lA

Growth Phase

Radial Cross-Section
Margin Ventral Margin

Translucent increment forming
on the marginal edges

Translucent increment one-half

Translucent increment complete

Opaque increment forming on
the marginal edges

Opaque increment one-half

0-e i t
Opaque increment complete

____________________ 4.

Figure 5. The arbitrary six-part subdivision of incremental shell growth of the
quahog (Mercenaria sp.) (Quitmyer and Jones 1992).

01, 02, and 03 represent a major part of the spring profile Sand oysters have
of shell growth. The trace of TI and T2 shell growth indicating collect
represents the first individuals in the population to start the substrate is con
forming the translucent growth increment in their shells. In in mixed sandy mu
the summer quarter (June through August), the growth 2.0. Channel oyster
profile may be characterized by the strong presence of all HLR of greater tha
phases of translucent growth and a decline of opaque shell oysters have small
growth over the spring quarter. Finally, in the fall quarter HLR of greater thai
(September through November), the TI growth phase is in small clumps,
much diminished from the previous season (summer) and 1988:30).
there is a marked increase in T2 and T3 phases. Only a trace The HLR means
of opaque growth phases is present during this quarter. This period (ca. A.D. 1
marks a sharp decline in opaque shell growth over the of 90-110 centimet
previous spring quarter. 250 cmbs (ca. A.D
When the frequency of the various growth phases is were made on oyst

summed for the year, translucent growth
phases (in particular, T3) dominate the
growth-frequency profiles (Figure 6).
The presence of opaque growth is less
frequent, thus indicating a shorter time
of opaque shell formation through the
year when compared to translucent shell
growth. A comparison of the archaeo-
logical quahog shell growth-frequency
profiles with the seasonal and annual
contemporary series provides an esti-
mate of archaeological quahog harvest.
Growth-frequency profiles were con-
structed for seven macro levels (Figure
1) to estimate the season of quahog
harvest for each level (compare Figures
6 and 7). There were 381 quahog shells
analyzed from these levels. The sample
size for the quahog shells excavated
from macro level 1C were not large
enough to be used reliably to estimate
season of harvest of the quahog clams
represented in that level.

Morphometric Analysis of the Shells of
the Eastern Oyster

Oysters live in a variety of habitats
that effect the general morphology of
their shells and the presence of
commensal species that occur on the
oyster beds. Some of the limiting fac-
tors of shell morphology and predator
activity are salinity, turbidity, and water
depth (Kent 1988:28). The shell-
height-to-shell-length ratio (HLR) ofthe
left valve (Figure 8) describes the shape
of the shell, which is correlated with the
habitats from which oysters were col-
lected (Kent 1988:28).
There are four general habitats that
can be described by the shell HLR
an HLR of 1.3 (short and broad shells)
on from intertidal beaches or bars where
posed of coarse sand. Bed oysters occur
id and have an intermediate HLR of 1.3-
rs are large, elongated forms that have a
an 2.0 and live in subtidal waters. Reef
elongated shells that are described by an
n 2.0. They grow as single individuals or
and live in the intertidal zone (Kent

urements were made on Safety Harbor-
00) oyster shells excavated from a depth
ers below surface (cmbs) and from 230-
1100) (Table 3). Similar measurements
er shell excavated from Manasota-period

T P mlarlA ANFrnmnpniorwICT

100RVnrl 21tA\


100 Charlotte Harbor Winter Charlotte Harbor Spring Charlotte Harbor SUMMER Charlotte Harbor Fall Charlotte Harbor- Annual
100 n=O1 n=116 100 100 100
86.1 n-=92 n=90 n=399
80 80 80 80 80

60 60 6 60 60 60

40 40 362 43.5 44.4 46.7 47.4
20 20 15.5 16.4 20 19.6 20 20 120 19.0 10.8
0.9 0.9 2.2 4.3 7.8 5.0 5.8
0 T1 T2 T3 01 02 3 0 T T2 T3 01 02 03 0 0LM
Ti T2 T3 01 02 03 Ti T2 T3 01 02 03 Ti T2 T3 01 02 03

Figure 6. Growth-frequency profiles illustrating the periodicity of incremental shell growth in quahog shells (Mercenaria sp.) collected monthly through the year
at Charlotte Harbor, Florida (Quitmyer and Jones 1992).

Figure 7. Growth-frequency profiles illustrating the timing of the formation of incremental shell growth in quahog shells (Mercenaria sp.) excavated from macro
levels IA, IB, IC, ID, IE, IIA, IH, (Safety Harbor Period ca. 1100) and IV (Manasota Period, ca. A.D. 300) the Shell Ridge, Palmer site.

Tnr U nnm:Arrnn..n. r~rIGO4R V-L.51I

Table 3. Mean shell heights, variances, and sample sizes of common
oysters (Crassostrea virginica) from the Palmer site's Shell Ridge Midden

Depth in Mean Shell e Sample Archaeolog
cmbs Height (mm) SVaize Period

90-110 51.5 607.6 22 Safety Harb

230-250 45.0 99.8 221 Safety Harb

313-340 50.4 129.5 397 Manasota

Figure 8. The positions of the measurements taken to
determine the height-to-length ratio (HLR) of the eastern
oyster (Crassostrea virginica) (after Kent 1988).

(ca. A.D. 300) deposits at a depth of313-340 cmbs (Table 3).
The HLR oyster measurements help to facilitate the compari-
son of the habitats that were exploited for this species during
the Safety Harbor and Manasota periods. The oyster shell
samples were randomly sampled and well preserved.


Migratory Waterfowl

In the course of conducting the "Window to the Past"
excavation, ten skeletal elements of common loon were
recovered from the Safety Harbor strata, while one element
was recovered from Manasota-period strata (Table 1) These
were present in levels extending from 39 to 340 cmbs (Table




1). In Dolan's excavation, six loon skeletal elements
were recovered from Manasota-period strata in Test
Units E, F, and G (Table 1). The presence of com-
mon loon is indicative of deposits that were formed
from October to May during the Safety Harbor and
Manasota periods.

Bay Scallops

The 1143 bay scallop shells excavated from three
macro levels (IA, IB, and IIA Safety Harbor Pe-
riod) of excavation units 8-1 and 8-2 had mean
heights that ranged from 38.3 to 43.5 mm (Table 2). The
minimum shell height measured was 17.3 mm, while the
maximum was 61.1 mm. When the minimum and maximum
ranges of shell height are plotted against the Anclote estuary
scallop data, a harvest period for all three Safety Harbor-
period samples extends from May to September (Figure 3).


The growth-frequency profile of the quahog shells (n =
47) excavated from macro level IA (Safety Harbor Period)
contains shells in all phases of incremental shell formation
(Figure 7). Most shells are in the translucent growth phases
(TI = 36.2%, T2 = 19.1%, T3 = 12.8%) or in opaque 1
growth phase (01 = 27.7%). The remaining 02 and 03
growth phases are each represented by 2.1% of the individu-
als in the sample. When these data are compared to the
modem growth series (Figure 6), a spring-through-summer
harvest period is indicated. This is when a small proportion
of the individuals is forming the opaque growth increment in
their shells and a larger number of individuals are forming
the translucent increment. A similar pattern is seen in the
sample of clam shells excavated from macro level IB (Safety
Harbor Period) (TI = 19.8%, T2 = 24.6%, T3 = 25.4%, 01
= 24.6%, 02 = 3.2%, and 03 = 2.4%) (Figures 6 and 7).
The growth-frequency profile ofthe quahog shells (Figure
7) analyzed from macro level ID (Safety Harbor Period)
(Figure 1) does not match any of the seasonal growth-
frequency profiles (Figure 6). A best fit is obtained by
comparing the archaeological profile (ID) with the annual
growth-frequency profile of the Charlotte Harbor data. Such
a profile contains individuals in all phases of growth;
however, most individuals are in the T3 (28.6%) and 01
(26.2%) growth phases. The remaining growth phases are
represented by a lower frequency of individuals that are
forming T (9.5%), T2 (16.7%), 02 (16.7%), and T3 (2.4%)
in their shells.
A season ofspring-through-earliest-summer clam harvest
was calculated for the quahog shells excavated from macro
level IE (Safety Harbor Period) (Figures 6 and 7). This
growth-frequency profile contains individuals that are
forming all phases of shell growth in their shells. None of
the growth phases dominate the growth-frequency profile
(TI = 19.4%, T2 = 23.6%, T3 = 19.4%, 01 = 12.5%, 02 =
11.1%, and 03 = 13.9%). This appears to match the data


8 991 Vot 51 4


OllimY~a EAsflNAI ITY AT THE Vn tr -- Q

from Charlotte Harbor where a large proportion of individu-
als is forming opaque growth in their shells through the
spring quarter and forming translucent shell growth during
the early summer. This would be before the fall quarter
when T2 and T3 incremental shell formation dominates the
growth-frequency profile.
Clam shells excavated from macro level IIA (Safety
Harbor Period) represented a fall-through-winter growth-
frequency profile (Figures 6 and 7). The sample is domi-
nated by individuals forming T3 (47.4%) growth in their
shells. Translucent 1 (10.5%), T2 (26.3%), and 01 (15.6%)
growth phases are well represented, while no individuals are
observed in the 02 and 03 growth phases.
The growth-frequency profile of clam shells excavated
from macro level III (Safety Harbor Period) indicates a
summer quahog harvest period (Figures 6 and 7). This
profile is dominated by individuals that are in the three
phases of translucent shell growth (TI = 32.0%, T2 = 24.0%,
T3 = 36.0%). A trace of Ol (4.0%) and 02 (4.) is observed,
while no 03 incremental shell formation was identified. A
similar growth-frequency profile was calculated for the
quahog shells excavated from macro level IV (TI = 14.3%,
T2 = 37.7%, T3 = 40.5%, and 01 = 9.5%). Macro level IV
belongs to the mid-Manasota Period, while the overlying six
macro levels are approximately 800 years younger and date
to the early Safety Harbor Period.

Eastern Oysters

The HLR of the oyster shells from the 90-110 cmbs
(Safety Harbor Period, ca. A.D. 1100), 230-250 cmbs (Safety
Harbor Period, ca. A.D. 1100), and 313-340 cmbs (Manasota
Period, ca. A.D. 300) levels of the midden range from 1.5 to
1.6 (Figure 9). Over 90% of the sample show signs of
attachment points to a solid substrate. It seems that these
oysters represent forms that occur in clustered groups;
perhaps in the intertidal zones of the estuary. The HLR falls
just below that observed for reef oysters, but within the range
of bed oysters. The difference is, however, that nearly all of
the valves show evidence of attachment, and this is most
closely associated with reef oysters (Kent 1988:28).
Another environmental indicator is the size that indi-
viduals within a population of oysters attains (Table 3). As
explained above, the greatest shell height of the left valve
was used for this measure. Oyster shells sampled from 90-
110 cmbs have a mean height of 51.5 mm (variance =
607.6). The oyster shells sampled from 230-250 cmbs had
a mean shell height of 45.0 mm (variance = 99.8), while
those excavated from 313-340 cmbs had a mean shell height
of 50.5 mm (variance = 129.5).
The question arises: do the size classes of the oysters
represented in these three samples change over time? A two-
tailed Student's t test was calculated for the three samples,
using the measured mean shell height of the oyster left valves
(Table 4). The null hypothesis is that the three samples
could have been collected from the same population. The
alternative hypothesis is that the three samples could not

Table 4. Results of two-tailed student's t test comparing the
mean shell heights of left oyster valves sampled from Safety
Harbor- and Manasota-period contexts at the Palmer site's
Shell Ridge Midden (8S02).
Samples t Null Hypothesis (Ho)

A, B 2.4 Rejected in favor of the alterna-
tive hypothesis (H,)

A, C .4 Accepted

B, C 5.9 Rejected in favor of the alterna-
tive hypothesis (H,)

The area of rejection of the null hypothesis, at degrees of freedom
equal to infinity, at the .98 test level, is 2.326 (Cloake 1988:422).

H- = There is no difference in the measured mean shell heights of
the left valves of oysters sampled from the three levels of the
excavation indicating that these shells could have been collected
from the same population.
H, = The measured mean shell heights from the three excavated
levels are significantly different indicating that these shells were
not collected from the same population.

A = Level 90-110 cmbs, Safety Harbor Period.
B = Level 230-250 cmbs, Safety Harbor Period.
C = Level 313-340 cmbs, Manasota Period.

have been collected from the same population. The area of
rejection of the null hypothesis is 2.326 at the.98 test level,
degrees of freedom equal to infinity (Cloake 1988:422).
Table 4 shows that the mean shell height of the Safety
Harbor-period oysters from 90-110 cmbs is significantly
different from those sampled from 230-250 cmbs. Oysters
found in 90-110 cmbs (Safety Harbor Period) and 310-340
cmbs (Manasota Period) could have been collected from the
same population. The mean shell height of the oysters from
230-250 cmbs (Safety Harbor Period) and 310-340 cmbs
(Manasota Period) are significantly different and could not
have been sampled from the same population.


Radiometric ('4C) analysis of the top and bottom of the 3-
m vertical section exposed in the "Window to the Past"
excavation indicates that it accumulated in two separate
episodes approximately 800 years apart, one during the
Manasota Period (ca. A.D. 300) and the other during the
Safety Harbor Period (ca. A.D. 1100) (Torrence 1991a).
Deposition of the faunal and botanical constituents was not
constant through time. Some levels appear to have been
deposited rather quickly. The evidence for this is the
presence of articulated or partially articulated skeletons of
animals (e.g., shark vertebrae; see Kozuch, this issue).
There also are successive, fine lenses of bone that are
dominated by skeletal material from bony fishes (Osteich-
thyes). Further indications of relatively quick deposition of
strata are places where mollusk shells are uncrushed.


RE FloannA Arnamm t

Palmer Site (90 100 cmbs)
35 n = 21
Z 20 19.0
P 15 14.314.3 14.3
10 9.5 9.5 9.5
5 4.8 4.8
0 0.o 0.0 .0 0.0 0.0
1.0 1.1 1.2 1.3 1.4 1.5 1.6 1.7 1.8 1.9 2.0 2.1 2.2 2.3 2.4
Shell Height/Shell Length Ratio

Palmer Site (230 250 cmbs)
35 n = 215

S25 2
5 20
S15 12.1 13.0

S6.59.5 09 1 0.0
0 0.0
1.0 1.1 1.2 1.3 1.4 1.5 1.6 1.7 1.8 1.9 2.0 2.1 2.2 2.3 2.4
Shell Height/Shell Length Ratio

Palmer Site (313-340 cmbs)
35 n =396

20 18.9
S11.4 12.4
10 8.3 8.6
2.5 1.0 28 2.3 1.8 1.0 0.8

1.0 1.1 1.2 1.3 1.4 1.5 1.6 1.7 1.8 1.9 2.0 2.1 2.2 2.3 2.4
Shell Height/Shell Length Ratio

Figure 9. The distribution of the height-to-length ratios of eastern oyster
(Crassostrea virginica) shells excavated from the Palmer site (8S02).

HL ___ ____ ^____ -.S V---_, A-- I-.

8 991 Vol 51{4)



Taxon Level I D





r Fall

Safety Harbor (ca. A.D. 1100)



Quahogs ID



|Quahogs III

Manasota (ca. A.D. 300)
Loons 313-340
Loons Test E
Loons Test F
Loons Test G
Quahogs IV

Figure 10. Season of resource collection for three zoological indicators excavated from Palmer site (8S02).

In contrast to the strata discussed above, there are places
in the midden that are marked by heavily crushed mollusk
shell and bone. The crushing and abrading of these materi-
als might have resulted from humans living on the midden
surface for relatively long periods of time.
The various strata are evidence of a rather complex and
diverse set of human activities that occurred through part of
the history of the Palmer site. Analysis of the bay scallop,
quahog, oyster, and loon remains preserved in these strata
provide information about when and how people lived on
Little Sarasota Bay.
The seasonality of seven of the eight macro levels is
summarized in Figure 10. Analysis of the quahog and
scallop shells contained in macro levels IA and IB (Safety
Harbor Period, ca. A.D. 1100) seems to indicate that the
harvest of these animals occurred during the spring and

summer quarters of the year. This does not rule out the
occupation of the site during other seasons because it is
known that various plant and animal resources are targeted
through the seasonal round of subsistence (see papers by
Kozuch and Newsom, this issue).
Macro level ID (Safety Harbor Period) was interpreted as
generalized backfill in a pit or hole that was dug into the
midden and filled during the Safety Harbor Period. It
contains quahog shells that were collected through all
seasons of the year (Figures 1 and 10). Mixing of shells
from underlying levels and/or the long-term, open accumula-
tion of shells and other debris in the pit would have produced
such a result.
The estimated season of quahog and scallop harvest for
macro level II (Safety Harbor Period) shows that shellfish
collection was not necessarily confined to a particular part of









aMS At .


the year during the Safety Harbor Period. Quahogs repre-
sented from macro level IIA (Safety Harbor Period) were
collected during the fall and winter quarters, while the
scallops were harvested during the spring and summer
(Figure 10).
The season of harvest for the quahogs represented in
macro levels III (Safety Harbor Period) and IV (Manasota
Period) is the summer season. This is a narrower period of
quahog harvest than is represented in the overlying strata.
The presence of loons in the Safety Harbor-period and
Manasota-period faunal assemblages, along with quahogs
and scallops, seems to show that Little Sarasota Bay was
used during all seasons of the year (Figure 10). The seasonal
gaps that are present during the late spring and early fall of
the Manasota Period, are probably more a function of the
number of seasonal indicators analyzed, than of people
abandoning the site during these brief periods of the year.
The known habitat preferences of the animals that were
part of this study provides information about some of the
micro habitats that were exploited in the vicinity of the
Palmer site. Inferences about the methods of resource
exploitation also can be made.
Scallop shells contained in the Safety Harbor strata of the
Shell Ridge midden are indicative of collecting in the
shallow grass beds of the estuary. The largest ofthe individ-
uals are easily collected by treading or by visually exploring
for the animals in the shallows. However, there is a suffi-
cient number of very small bay scallop shells in all of the
measured samples to indicate that, in addition to hand
harvesting, some other mass-capture device might have been
used in their procurement. It is known that aboriginal people
had a sophisticated fine-mesh-net technology that could have
been used to collect scallops in the shallow grass flats of the
estuary. The presence of so many small individuals may
indicate that the whole animal was being consumed instead
ofjust the muscle. This is in contrast to the practices of most
present-day collectors of scallops.
Quahogs occupy a variety of micro-habitats in the estuary.
They may be acquired in greatest numbers in the sandy mud
of the grass flats. The most likely ways of obtaining these
animals are by treading or raking. Scallop and quahog
habitats overlap and both could have been acquired during
the same collecting trip. Quahogs were regularly identified
from Safety Harbor-period and Manasota-period strata.
As discussed above, oysters occupy a variety of habitats
that effect their shell morphology (Kent 1988:28). The four
habitats characterized by the shell-height-to-shell-length
ratio (HLR) is known for commercially viable beds located
in more northern latitudes. It is clear that oyster beds of a
non-commercial nature need to be studied. The measured
oyster shells from the "Window to the Past" excavation seem
to indicate that they were collected from clusters of oysters
connected to solid substrates. The HLR is slightly less than
that observed for reef oysters, but most individuals show
signs of connection to a solid substrate. The oysters from all
levels seem to have been collected from the same habitat type
during the Safety Harbor and Manasota periods

A difference exists in the size of the oyster shells that were
collected through time at the Shell Ridge. A similar phe-
nomenon was observed by Bushnell (1965) in his study of the
oyster shells excavated by the Bullens from Tests A and B in
the nearby Hill Cottage midden (Bullen and Bullen 1976:3-
20). The Hill Cottage midden contains food remains of the
Late Archaic-period people who lived at the Palmer site.
Bushnell measured the shell height of oyster shells sampled
from eleven, successively deeper levels of the Archaic-period
midden and found differences in the sizes of the mean shell
height. He concluded that these differences represent
"ecological-environmental" changes (Bushnell 1965). The
sizes of Bushnell's oysters and other species of mollusks
from 96-132 inches below surface (244-335 cmbs) were
interpreted as individuals that lived in an environment that
was cooler and siltier. Bushnell goes on to suggest that, after
this period in history, there was gradual warming of the
earth's climate and thus a rise in sea level. This is character-
ized by the presence of oysters that have a mean shell height
greater than the more recent excavation level.
It is clear from the data that changes in the shell height of
the oyster shells found in both the Shell Ridge and Hill
Cottage middens occurred. The causes for these differences
are not easily determined at this level of research. Differ-
ences in the shell height of oysters might be explained by
differing salinity regimes from changes in sea level or even
the natural closing and opening of inlets into the bay.
Estuarine salinity also can be effected by freshwater runoff
resulting from wet and dry climate events. Another explana-
tion simply could be that the Palmer site people were
collecting oysters from slightly different habitats. Finally,
oysters could have become smaller over time due to human
harvesting pressure. The causes of the changes in oyster
shell height remains a topic of further research.


The purpose of this paper was to provide information
about seasonality of the early Safety Harbor-period (ca. A.D.
1100) and mid-Manasota-period (ca. A.D. 300) peoples that
occupied the Palmer site's Shell Ridge midden. The known
habits and habitats of the animals studied during the course
of this research also provides information about how and
where these people obtained some of their food resources.
The seasonality study of scallops, quahogs, and loons
presented in this paper indicates that the Safety Harbor- and
Manasota-period people formed the Shell Ridge midden
during all seasons of the year. Measurements of scallop
shells from the Safety Harbor Period indicate they were
collected between March and late August. Zooarchaeological
quahog shells were cross-sectioned to expose their annual
growth increments. Using a modern analog that correlates
quahog incremental shell formation with the season of the
year, it was found that Safety Harbor people collected
quahogs during all seasons of the year, but predominantly
during the spring and summer. A sample of quahogs from
the Manasota Period indicated that quahogs were collected

CprAmcr A rrv AT rTUn PDA Uw..r Crrr


during the summer season. Loons were present in both the
Safety Harbor-period and Manasota-period strata, thus
indicating they were obtained between the fall and spring
quarters of the year.
The known habitats of oysters, scallops, and quahogs
show that the shallow grass flats of the estuary and shallow
oyster reefs were favorite places for the collection of shell-
fish. Acquisition of these resources was a simple gathering
task. However, the presence of very small scallop shells
seems to indicate that, in addition to hand harvesting, some
sort of mass-capture device (e.g., nets) was being used.
Some of the scallops were so small that it seems that these
animals were collected with nets and the whole animal was
eaten. This is in contrast to the way that most present-day
people collect and prepare scallops. In contrast to our own
modern taboo against consuming shellfish during the warm
months, Safety Harbor and Manasota people had no such


Funding for "A Window to the Past," was provided to the Gulf Coast
Heritage Association, the non-profit owners of the site, by a Florida Department
of State Special Category Grant. The excavation project was organized by
Marion M. Almy, Archaeological Consultants, Inc., Sarasota, under the general
direction of Linda Williams Mansperger, Executive Director of the Gulf Coast
Heritage Association.
I acknowledge Elizabeth S. Wing, Environmental Archaeology Laboratory,
and William S. Marquardt, Southwest Florida Project, Florida Museum of
Natural History for their fiscal and intellectual support of this study. Identifica-
tion of the common loon was made by Robert H. Chandler, Vertebrate
Paleontology, Florida Museum of Natural History. I appreciate the help of
George Luer who encouraged the publication of this manuscript. I also
acknowledge those that reviewed and commented on this research.

References Cited

Abbott, R. Tucker
1974 American Seashells. Van Nostrand Reinhold Co., New York
Austin, Robert J., and Michael Russo
1989 Limited Excavations at the Catfish Creek Site (8S0608), Sarasota
County, Florida. Report prepared for Palmer Venture, Inc. by Piper
Archaeological Research, Inc. St. Petersburg. Copy on file, Department
of Natural History, Zooarchaeology Laboratory, Florida Museum of
Natural History, Gainesville.
Barber, Bruce J., and Norman J. Blake
1983 Growth and Reproduction of the Bay Scallop Argopecten irradians
(Lamarck) at its Southern Distributional Limit. Journal of Experi-
mental Marine Biology and Ecology 66:247-256.
Broom, M. J.
1976 Synopsis of Biological Data on Scallops, FAO Fisheries Synopsis
No. 114. Food and Agriculture Organization of the United Nations,
Bullen, Ripley P., and Adelaide K. Bullen
1976 The Palmer Site. Florida Anthropological Society Publication No.8,
Bushnell, Francis F.
1965 Preliminary Report on the Ecology of Little Sarasota Bay Based on
Shellfish Samples from the Palmer Site. Manuscript on file, Depart-
ment of Anthropology, Florida Museum of Natural History, Univer-
sity of Florida, Gainesville.
Cloake, William T.
1988 Statistical Modeling Using 1-2-3: Release 2. Management Informa-
tion Source, Inc. Portland, Oregon.
Howell, Arthur Holmes
1932 Florida Bird Life. Coward-McClain, Inc., New York.

Hudson, J. H., E. Shinn, R. Halley, and B. Lidz
1976 Sclerochronology: A New Tool for Interpreting Past Environments.
Geology 4:361-364.
Jones, Douglas S., Irvy R. Quitmyer, William S. Arnold, and Dan C. Marelli
1992 Annual Shell Banding, Age, and Growth Rate of Hard Clams
(Mercenaria spp.) from Florida. Journal of Shellfish Research
Jones, Douglas S., and Irvy R. Quitmyer
1996 Marking Time with Bivalve shells: Oxygen Isotopes and Season of
Annual Increment Formation. Palaios 11:340-346.
Kent, Bretton
1988 Making Dead Oysters Talk: Techniques for Analyzing Oysters from
Archaeological Sites. Maryland Historical Trust, Historic St. Marys
City, Jefferson Patterson Park and Museum.
Quitmyer, Irvy R., and Douglas S. Jones
1992 Calendars of the Coast: Seasonal Growth Increment Patterns in the
Shells ofModem and Archaeological SouthemQuahogs, Mercenaria
campechiensis, from Charlotte Harbor, Florida. In Culture and
Environment in the Domain of the Calusa, edited by William H.
Marquardt, pp. 247-264. InstituteofArchaeologyandPaleoenviron-
mental Studies, Monograph Number 1, Gainesville.
Quitmyer, Irvy R., Dogulas S. Jones, and Williams S. Arnold
1997 The Sclerochronology of Hard Clams, Mercenaria. spp., from the
South-Eastern U. S.A.: A Method ofElucidating the Zooarchaeological
Records of Seasonal Resource Procurement and Seasonality in
Prehistoric Shell Middens. Journal of Archaeological Science
Russo, Michael, and Irvy R. Quitmyer
1996 Sedentism in the Coastal Populations of South Florida. In Case
Studies in EnvironmentalArchaeology, edited by Elizabeth J. Reitz,
Lee A. Newsom, and Sylvia J. Scudder, pp.215-231. Plenum Press,
New York.
Torrence, Corbett McP.
1991 a Unpublished field notes, Window to the Past, Palmer Archaeological
Site, 8S02. On file, Department of Natural History, Southwest
Florida Project Laboratory, Florida Museum of Natural History,
University of Florida, Gainesville.
1991b Untitled class paper. Manuscript on file, Department of Natural
History, Florida Museum of Natural History, University of Florida,
Winter, M. A., and P. V. Hamilton
1985 Factors Influencing Swimming in Scallops, Argopecten irradians
(Lamarck, 1819). Journal of Experimental Marine Biology and
Ecology 88:227-242.



-_ el

Archeology of the
Florida Gulf Coast
Gordon R. Willey
With a new preface by the author
"By the end of 1950, only about a
dozen publications in American
archaeology might be said to stand
as monumental contributions....
Of these, the reviewer regards
Gordon Willey's great volume on
the Florida Gulf Coast as perhaps
the best of all."-American Antiquity
Paper, $29.95

Archaeology of
Colonial Pensacola
Edited by Judith A. Bense
"Dr. Bense and her colleagues are
resurrecting a lost community,
the colonial predecessor of
modern Pensacola, from the
long-neglected archaeological
remnants and archival docu-
ments of Spanish and British
colonists."-Gregory A.
Waselkov, University of South
April. Cloth, $49.95

Space and Time
Perspective in Northern
St. Johns Archeology,
John M. Goggin
Introduction by James J. Miller
"Will long stand as a highly
important reference.... The wealth
of material summarized, and the
inclusiveness of the bibliographical
coverage, will assure this....
Undeniably the most significant
single work which has appeared on
this important area."
-John W. Griffin, The Florida
Paper, $29.95

Early History of the
Creek Indians and
Their Neighbors
John R. Swanton
First published in 1922, copies of
this respected classic have been
coveted, hoarded, and worn ragged
ever since by archaeologists,
anthropologists, and historians
across the Southeast and beyond.
Paper, $29.95

Florida Indians and the
Invasion from Europe
Jerald T. Milanich
"This book is excellent. Its detail,
readability, and references make it a
necessary addition to the library of
every student of southeastern
Indian cultures and culture
change."-North American
Paper, $19.95

Florida's Indians
from Ancient Times
to the Present
Jerald T. Milanich
"An exceptional book.... It is a
wonderful synthesis, and will be
avidly read by both professional
archaeologists and the general
public."-Marvin T. Smith,
Valdosta State University
Native Peoples, Cultures, and Places
of the Southeastern United States Series
Cloth, $39.95 Paper, $19.95

An Environmental
History of Northeast
James J. Miller
"A significant contribution to an
overall understanding of northeast
Florida.... Frees the reader to
consider... our relationship to the
land. This book makes you
think."-Robert L. Thunen,
University of North Florida
Museum of Natural History:
Ripley P Bullen Series Cloth, $49.95

Southeastern Classics in Archaeology,
Anthropology, and History
Series Editor: Jerald T. Milanich

Order through full-service
booksellers, with VISA or M/C
toll free: 1-800-226-3822, or
though our website at http://


The Apalachee Indians
and Mission San Luis
John H. Hann
and Bonnie G. McEwan
Financed in part with historic preservation
grant assistance provided by the Bureau of
Historic Preservation, Division of Historical
Resources, Florida Department of State,
assisted by the Historic Preservation
Advisory Council.
Cloth, $49.95 Paper, $19.95

Hernando de Soto
among the Apalachee
The Archaeology of the
First Winter Encampment
Charles R. Ewen and John H. Hann
Paper, $29.95

The Timucuan Chiefdoms
of Spanish Florida
Volume I: Assimilation
Volume II: Resistance and Destruction
John E. Worth
Volume I: Cloth, $49.95
Volume n: Cloth, $49.95

Ancient Earthen
Enclosures of the
Eastern Woodlands
Edited by Robert C. Mainfort, Jr,
and Lynne P. Sullivan
Cloth, $49.95



Center for Archaeological Investigations, Southern Illinois University at Carbondale, Carbondale, Illinois 62901-4527
E-mail: lnewsom@(siu.edu

Archaeological investigations at the Palmer Shell
Ridge Midden (8S02) in Sarasota County were
carried out in August 1991 in conjunction with a
Special Category Grant project entitled "A Window to the
Past." The grant was awarded by the Florida Department of
State, Division of Historical Resources, to the Gulf Coast
Heritage Association, which owns and manages the site as
part of an historical and archaeological preserve, "Historic
Spanish Point." Corbett McP. Torrence of the Florida
Museum of Natural History (FLMNH) supervised the field
operations under consultation with Marion Almy, George
Luer, and William H. Marquardt.
The Palmer site is a prehistoric village complex located in
a gulf coastal or near-shore setting, south of Tampa Bay near
Osprey, Florida (Figure 1; also Figure 1 in Kozuch, this
issue). The 1991 excavations concentrated on one portion of
the Palmer site, the Shell Ridge Midden (8S02), which
consists of an extensive series of prehistoric living surfaces
and refuse deposits that range in age from approximately
A.D. 200-1200. From the material culture remains, particu-
larly ceramics, and accompanying radiocarbon dates, the
deposits have been assigned at the earliest to the prehistoric
Weeden Island, or more specifically, the middle-to-late
Manasota, culture (ca. A.D. 300-700) (Luer and Almy 1982;
Milanich 1994:221-227). Later deposits are associated with
the Safety Harbor culture (post ca. A.D. 900), South-central
or Manasota "variant" (Milanich 1994:400).
Among the research questions guiding the excavations at
the Palmer Shell Ridge Midden was the role of the economy,
and more specifically the subsistence system, in terms of the
long-term sustainability of the prehistoric occupation and
ultimately the development of social complexity. The
paleoethnobotanical aspects of the deposits and what the
archaeobotanical data indicate about the plant-use compo-
nent of the economy are the focus of this report.
A number ofenvironmental archaeology studies have been
conducted in this gulf-coastal region of Florida (e.g., Austin
and Russo 1989; Brown and Cohen 1985; Luer 1977a,
1977b, 1992, 1995; Luer and Almy 1982; Russo and Quit-
myer 1996). In combination, this research has provided
information on paleoenvironment, settlement dynamics, and
the faunal component of human diets. There nevertheless
exist significant gaps in basic information, particularly
concerning prehistoric subsistence systems and especially in

regard to plant use. Manasota subsistence is characterized in
general as having been strongly oriented toward natural
wetlands and maritime environments, a complex system of
fishing, hunting, and foraging, with the last component
including wild plants, shellfish, and other natural resources
(Luer and Almy 1982:39-43; Milanich 1994:224-225).
Staple crop production involving maize and/or root-crop
horticulture, does not appear to have been part of Manasota
subsistence patterns (at least there has been no evidence for
the presence of these plants; see below). Nevertheless, it is
reasonable to infer that homegardens and the deliberate
maintenance of a series of useful plants and perhaps quasi-
domesticates, such as gourds, gourd-squashes, as well as
medicinals and tobacco, are very likely to have occurred,
given the evidence for these plants and plant-use systems at
other Florida peninsular sites representative of various time
periods (Decker and Newsom 1988; Newsom 1987a, 1998;
Newsom and Scarry 1998; Newsom et al. 1993; Scarry and
Newsom 1992; and see below).
The excavations at Shell Ridge Midden afforded an
opportunity to conduct additional research that would help
illuminate food acquisition, environmental setting, and
prehistoric human adaptations along the central Florida gulf
coast. Thus, the research design and testing of the Palmer
site's Shell Ridge Midden incorporated specialized
paleoethnobotanical and zooarchaeological sampling and
analyses. The paleoethnobotanical component ofthe research
at the Shell Ridge Midden is described here.

Previous Paleoethnobotanical Research
in Peninsular Florida

At the 1989 Southeastern Archaeological Conference
meetings in Tampa, Margaret Scarry and I, along with
Marilyn Masson (Scarry et al. 1989), presented a synthesis
of all the paleoethnobotanical research that had been con-
ducted in the southern peninsula up to that time. Since then,
additional research has been completed in the Caloosa-
hatchee region (Newsom 1991a; Newsom et al. 1998;
Newsom and Scarry 1994, 1998), and limited analyses of
plant specimens from Carr's excavations at Cutler Ridge in
Miami also have occurred (Newsom 1987b). Relatively
extensive paleoethnobotanical studies have been conducted
of recovered plant remains from sites farther north in the



VOL. 51 NO. 4

U.S. 41

Sr\ Vamo Rd




Figure 1. Map showing the location of the Palmer site on the west coast of Florida.



8 991 Vot 51(4)

NIwsom Un-nA .rnnnr-,i A. HELL .nrn MUDDEN 2

central peninsula, from about Sarasota and Indian River
counties, north to around St. Augustine and the Gainesville
area. Among these are Hontoon Island, 8V0202 (Newsom
1987a), Groves' Orange Midden, 8V02601 (Newsom 1994),
Windover, 8BR246 (Newsom 1998), Fig Springs (Newsom
and Quitmyer 1992), and Little Salt Spring (Newsom n.d.).
Limited analyses of plant remains from the Seminole Rest
site, 8V0124, have been conducted by Ruhl (1995). The
results of all these studies are briefly summarized in the
following paragraphs to provide some background perspec-
tive in advance of presenting the Palmer Shell Ridge Midden
data. Since this research is focused on the archaeobotany of
the subtropical southern and central peninsula, archaeo-
botanical studies from areas north of approximately St.
Augustine-Gainesville-Cedar Key are not reviewed here.
Paleoethnobotanical research has demonstrated much
consistency in plant use across the entire peninsula and even
the various temporal sequences. Viewing the sites collec-
tively and beginning with carbonized wood, fuel-wood
selection seems to have been strongly focused on native
"yellow" pines (Pinus spp., section Diploxylon, taeda group;
the "southern hard" or "yellow pine" anatomical group
[Panshin and deZeeuw 1980; Phillips 1941]). Oaks, espe-
cially live oak (Quercus virginiana), are the next most
frequently identified taxa among carbonized wood speci-
mens, particularly with regard to the more northerly sites. In
southern coastal locations (e.g., Charlotte Harbor), various
mangrove taxa, especially black mangrove (Avicennia
germinans), assumed primary or secondary importance as
fuelwoods, depending on the particular deposit tested. Hard-
group pine also was frequently burned in this region as fuel
or for other purposes. Also identified from carbonized wood
remains retrieved from the various sites are hickories (Carya
spp., true and pecan anatomical groups), ash (Fraxinus spp.),
and a series of less common tropical trees (e.g., false mastic
[Mastichodendronfoetidissimum], wild lime [Zanthoxylum
sp.], cocoplum [Chrysobalanus icaco]). These taxa and
others appear consistently in lower frequencies than the pine,
oak, and mangrove genera.
Where waterlogged anaerobic conditions have allowed for
the preservation of uncarbonized wood-working debris or
wood debitage (e.g., Hontoon Island, Groves' Orange
Midden, and Pineland), it is abundantly evident that cypress
(Taxodium sp.) was heavily utilized (90-100% of the adzed
chips from these sites) for large wood-working tasks, perhaps
for dugout canoes, bowls and other utilitarian items, or totem
manufacture. Wooden artifacts from Hontoon Island, Groves'
Orange Midden, and Windover indicate very deliberate and
careful selection and matching of genera-specific wood
characteristics with the functions of particular objects (e.g.,
holly and maple [resilient woods] for stakes or tool handles;
live oak [impact-resistant wood] for pounding implements
[see Newsom 1987a, 1994, 1998]). Thus, the collective data
on wood remains from all these sites clearly demonstrate that
native Florida groups were very skilled and knowledgeable
about local wood resources.
Likewise, although a diverse array of plant foods poten-

tially was available to the various cultural groups inhabiting
the different parts of the peninsula, there was strong and
consistent emphasis on a few apparently high-priority
resources. Wild fruits from the coastal and interior tropi-
cal/subtropical hardwood hammock vegetation were widely
and regularly procured as food items. Among these are
cocoplums (Chrysobalanus icaco), sea grape (Coccoloba
uvifera), prickly-pear (Opuntia spp.), false mastic
(Mastichodendronfoetidissimum), and especially grape (Vitis
spp.) and saw palmetto (Serenoa repens). Certain other
plants that may have been exploited for their edible seeds
and/or greens also appear with regularity among the various
site deposits. Among these are members of the goosefoot
family Chenopodiaceae, purslane (Portulaca spp.), trian-
thema (Trianthema portulacastrum), pokeweed (Phytolacca
americana), and grass seeds belonging to the Panicoid tribe
(Poaceae, tribe Paniceae, e.g., Setaria spp., Panicum spp.;
hereafter "panicoid grass" [see Newsom 1991a:612-613]).
These five taxa are typically components of the native weed
flora, which means that their presence in various site
deposits may be purely incidental to the human occupations.
Nevertheless, their consistent appearance at sites, often in
association with definitive food remains (e.g., the stomach
contents from Windover inhumations), as well as relatively
high concentrations of the seeds at some sites (e.g., 491
panicoid seeds from one feature at Pineland [Newsom and
Scarry 1998]) are good indications that these taxa were
utilized resources, perhaps for food, fibers, medicines, and
Bottle gourd (Lagenaria siceraria) and Cucurbita pepo
gourd/squash remains are another interesting and very
consistent element of peninsular Florida archaeobotanical
assemblages. One or both of these Cucurbitaceae genera are
present at nearly all of the peninsular sites where archaeo-
botanical research has been conducted. Bottle gourd has an
association with human groups dating from at least as early
as the Florida Archaic Period based on the Windover site,
where bottle gourd remains were recovered from mortuary
contexts (Doran et al. 1990), and Little Salt Spring (Newsom
n.d.). Still earlier evidence (late Pleistocene) for the presence
of gourd/squash Cucurbita pepo comes from the Page-
Ladson and Latvis-Simpson sites in the Florida panhandle
(Newsom et al. 1993; Newsom and Mihlbachler n.d.). These
gourd/squash remains occur in association with Archaic
and/or Paleoindian strata, as well as earlier deposits that
presumably predate the presence ofhuman groups in Florida.
Gourds of all types and human cultures have a long
history of association and on a nearly world-wide basis
(Heiser 1979). The presence and ubiquity of gourd remains
from Florida sites invoke the question of whether peninsular
Florida cultures cultivated these and other plants in addition
to gathering useful species from the natural vegetation.
Anatomical and morphometric analyses of the various
Florida archaeological gourd accessions have been largely
indicative of wild forms (Decker and Newsom 1988; New-
som 1998; Newsom et al. 1993). Moreover, several recent
studies of modern gourd populations have demonstrated that


Aoro rcno~srur r ~ron ro~~ r P~~ D~~r ~~m~r-~


TH WI _mPLGS 99 O 14

these or similar gourd species are capable of sustained
dispersal and survival without human intervention (e.g.,
Cowan and Smith 1993). Thus, most of the prehistoric
gourds from Florida sites could have been gathered from
their natural riparian habitats. There is, however, enough
variation among seeds from some of the deposits at Groves'
Orange Midden (Newsom 1994), Hontoon Island (Newsom
1987a), and Pineland (Newsom and Scarry 1998) to suggest
the presence of distinct gourd varieties. These may be an
indication that some level of selection and perhaps mainte-
nance (gardening) of gourd or squash varieties occurred.
Extreme variation in Cucurbita sp. seed and peduncle
morphology is, however, associated exclusivelywith historic-
period strata from Hontoon Island (Decker and Newsom
1988; Newsom 1987a). The specimens from these late levels
are almost certainly evidence for deliberate selection and
gardening of gourd and/or squash cultivars.
The possibility that gourds and perhaps other useful plants
were cultivated or were on some level maintained and
encouraged by prehistoric people inhabiting the Florida
peninsula implies a step beyond simple gathering. However,
this is not, in and of itself evidence for agriculture, which is
a completely separate level of reliance and control of plant
resources. The question of more sustained or intensive plant
cultivation by prehistoric people in peninsular Florida
remains open, particularly since evidence of maize (Zea
mays), a plant so definitively associated with crop produc-
tion, is largely lacking from peninsular sites. Where maize
has been identified it is minimally present, suggesting that it
was grown as a minor crop. Peninsular sites from which
archaeological maize macroremains have been recovered
include the extreme northwest area (Fig Springs and Baptiz-
ing Spring), the central-northern reaches of the St. Johns
River (Hontoon Island and Mount Royal), the Riverbend site
near Ormond Beach, the Spanish-Indian ranch known as the
Zetrouer site in Alachua County, and 8DA411, a Seminole
site in Dade County (for a summary of all these maize
assemblages, see Newsom 1989; and see Hutchinson et al.
1998, regarding additional identifications of maize from
north Florida and Georgia). Even in these cases, the maize
remains have been documented from the latest prehistoric
and/or historic deposits (see Milanich 1987 for a complete
discussion of problems with the reported maize from Fort
Center; also Johnson 1990; Kessel 1991; Miller-Shaivitz and
Iscan 1991).

Shell Ridge Midden (8SO2)

The Shell Ridge Midden study initiated systematic paleo-
ethnobotanical research for terrestrial deposits in the Central
Gulf Coastal region of Florida. The general sampling
protocol for the recovery of biological remains, both plant
and animal, from Shell Ridge Midden was designed to
provide relatively dispersed coverage of the midden deposits,
rather than intensive study of any particular strata. By this
approach, we sought to develop a generalized understanding
of the occupation as reflected by the midden refuse and gain

preliminary insights into local subsistence practices, perhaps
other behavioral variables (e.g., activity areas), and to some
extent, environmental conditions at the time of occupation.
Thus, the paleoethnobotanical research at the Shell Ridge
Midden focused on basic questions related to occupation
One question concerned the overall patterns of plant use
and food production, and how these might compare with
what has been documented from other areas of the Florida
peninsula. Settlement patterns formed another key area of
interest, particularly any evidence of population mobility
and/or season of occupation. Finally, the general environ-
mental conditions at the time of occupation, to the extent
they can be deduced from vegetation reconstructed from
plant remains, were an important concern.

Sample Selection and Recovery

Our selection of samples suitable for both archaeo-
botanical and zooarchaeological analyses was guided by an
attempt to satisfy both subdisciplines and to incorporate as
much heterogeneity of deposits from the midden as possible
for study within the two-month period allotted for analysis.
We concentrated first on a vertical column (30 x 30 cm),
designated 8-2, that extended downward through a long
series of midden layers and isolated concentrations of
biological debris. Initially selected for analyses were 13
samples from particular horizontal strata and/or features that
were intersected by Column 8-2 (Table 1). The four samples
designated 8-1 (Table 1) are from an adjacent unit and
represent a continuation into deeper deposits than were
penetrated by Column 8-2. Samples from three additional
areas of the midden (Unit 20-1, 21-2, and an ash sample in
the southwest wall of the exposed midden) also were ana-
lyzed for plant remains (Table 1, bottom).
The individual samples that underwent analysis were
selected during excavation by noting the presence of distinc-
tive and contrasting attributes such as color, texture, high
bone/ organic content, shell, charcoal, and/or concentrations
of ash. Particular emphasis was placed on obtaining samples
from a variety of depositional and serial units. In the midden
profile, samples from above approximately 250 cm below
datum (cmbd) date to ca. A.D. 1100 (early Safety Harbor
Period), whereas samples from below approximately 250
cmbd date to ca. A.D. 200-300 (mid-Manasota Period) based
on ceramics and radiometric dates (George Luer, personal
communication, 1997). Thus, all the samples from columns
8-1 and 8-2 are associated with a Safety Harbor occupation.
The three samples from deeper within the midden samples
20-1, 21-2, and the "Southwest Wall, ash sample" belong
to the earlier Manasota Period. The ash sample from the
southwest wall of the excavation area came from near the
base of the profile.
Beyond selection, the recovery and processing of individ-
ual samples proceeded according to standard archaeo-
botanical practice for shell midden deposits. As is typical of
Florida middens, the differently layered deposits comprising

1998 VOL. 51(4)


Table I. Shell Ridge Midden archaeobotanical samples.
Sieve Fraction
Unit/Leve esripion Sample (in mm)
(cm below datum) Volume (liters) 2
4 2 1 .4
8-2, 80-92 cm dark organic layer 19.0 x
8-2, at 97 cm sample from SE corer of level .4 x x x x
8-2, 100-117 cm organic concentration .4 x x
8-2, 109-135 cm organic concentration .4 x x x
8-2, at 114 cm sample from SE corer of level .2 x x x
8-2, 114-119 cm dark organic zone 19.0 x x
8-2, 124-141 cm SE ash locus .2 x x x x
8-2-1, at 130 cm large ash lens 19.0 x
8-2-2, at 130 cm organic concentration 6.0 x x x x
8-2, at 136 cm dark organic layer 19.0 x x x
8-2, at 157 cm dark organic layer 19.0 x x
8-2, at 163 cm dark organic layer 19.0 x x x x
8-2, at 166 cm mixed ash, organic layer 19.0 x x
8-1, at 176 cm sample from SE corer of level .6 x x
8-1, at 176 cm organic concentration, SW comer of level .2 x x x
8-1, at 181 cm dark organic layer 4.0 x x
8-1, at 186 cm Area 1, north side of unit 1.0 x x x x
20-1, at 354 cm dark organic layer 19.0 x x
21-2, at 351 cm concentration of plant materials 1.0 x x
SW Wall, ash large ash lens 15.0 x x

' In the profile, materials from above approximately 250 cm below datum date to ca. A.D. 1100 (early Safety Harbor Period), whereas
materials from below approximately 250 cm below datum data to ca. A.D. 300 (mid-Manasota Period).

separate midden strata varied greatly in terms oftotal volume
and spatial configuration (horizontal and vertical extent).
Samples of approximately 19 liters were collected whenever
a particular deposit/stratum was extensive enough to permit
collection of relatively large samples. Smaller deposits and
strata or features (including soil stains) were sampled by
collecting volumes of sediment ranging from 200 milliliters
to 15 liters, depending on the size of the deposit. The
volumes of individual samples are identified in Table 1. All
samples were bagged unscreened in the field and sent to the
Florida Museum of Natural History for further processing
and analysis.

Laboratory Methods and Procedures

At the FLMNH, preliminary tests ofthe effects ofwater on
the carbonized remains from the Shell Ridge Midden were
conducted in order to determine the appropriateness of
water- versus dry-processing and separation techniques.
First, a few large, dry fragments of charcoal from the
general-excavation screens were placed on a paper towel. A
small amount of water was dropped on one edge of the towel
and was allowed to permeate the paper to dampen the
individual charcoal fragments. This gentle hydration resulted
in severe fragmentation of the charcoal specimens, clearly

demonstrating that the carbonized wood could not withstand
moisture stress.
A second simple experiment was conducted with a
complete sample from the midden (8-2 at 136 cmbd) as an
additional test of the effects of subjecting long and deeply
buried, dry carbonized materials to sudden moisture ex-
tremes. The sample first was gently sieved dry through a
nested sieve series (4 mm, 2 mm, 1 mm, and .42 mm mesh
sizes) to presort the sample and remove large shells, artifacts,
and identifiable charcoal. The charcoal was removed since I
had already demonstrated the negative effects ofwater shock
to large charcoal fragments, and this avoided sacrificing
identifiable wood specimens. Next, the sample fraction
captured in the 2-mm sieve (>2 mm <4 mm) was selected for
the test. This sample component was split into two parts,
then one subsample was directly sorted to separate plant and
animal remains, while the other portion underwent typical
water flotation in a bucket placed in the laboratory sink,
again to separate the various sample constituents (see
Pearsall 1989). The final results showed that while the
overall ratios (the bulk weights/volumes) of plant remains to
faunal material stayed nearly equal for both subsamples, the
carbonized plant remains nevertheless suffered considerable
damage. The wood remains were recovered only 78%/ intact
(based on the amount of material that was still greater than



iii .. .... A O1 O

r :1'.4

Figure 2. Partitioned subsamples from midden sample 8-2/136 cmbd to assess the effects of
handling and separation procedures. The upper three subsample fractions were derived from the
sample split that underwent direct sorting to separate faunal and botanical remains. The lower
three fractions were derived from the other half of the sample that was processed by water
flotation. Plant remains in the flotation subsample (lower three images) were adversely impacted
by exposure to water, resulting in considerable amounts of material in the < 2-mm size range
relative to the other half of the sample (upper three images). Note the glossy appearance of the
flotation-sieve fractions, which is due largely to fresh breaks on the carbonized wood remains.

2 mm in size) in the flotation subsample versus 98% intact
in the sieved sample. That is, wood remains were reduced on
average to much smaller sizes in the flotation subsample,
with considerable quantities ending up in the I-mm and .42-
mm size ranges (Figure 2). By extension, carbonized seeds
from the Shell Ridge Midden deposits should also be affected
by water stress. Thus, to avoid the demonstrated negative
effects of using water methods (e.g., flotation or water-
sieving) to separate the sample components of this particular
midden, all samples subsequently were processed and
maintained in a dry state.
Initiating laboratory analysis, the volumes of all samples
were first verified and recorded, and then, in preparation for
the sieving process, each sample was spread out on a tray
and large shells and ceramics were extracted by hand to
avoid potential mechanical breakage to plant remains by
heavy materials. Larger fragments of wood charcoal also
were removed in this initial stage of sample preparation. The
individual samples were next gently passed through nested
sieve series with mesh sizes of 4 mm, 2 mm, 1 mm, and .42

mm, as above, and the resulting sieve fractions were then
processed through the various stages of sample separation
and analysis.
Ultimately, plant remains from 20 samples underwent
archaeobotanical analysis (Table 1). Time constraints did not
allow for each of the 20 samples to undergo complete
analysis (hence, the right-hand columns in Table I, which
show the extent of analysis for the individual samples by
sieve fraction). Five samples were completely analyzed for
plant materials, incorporating all four sieve fractions. Four
additional samples were analyzed to the level of the I-mm
sieve fraction, analysis of nine samples extended to the 2-mm
fraction, and two samples were limited to the 4-mm compo-
It is important to consider that the differing levels and
intensity of sample analysis will necessarily affect the data
compiled. Thus, in making comparisons between the proven-
iences and their respective identifications, the level to which
each sample was analyzed must be considered to ensure that
apparent differences between samples are not simply the


1998 VOL. 51(4)





result of analytical procedures. In general, identifiable seeds
were not recovered in the finest subfraction, and this also
tends to be true of archaeological deposits from peninsular
Florida. The .42-mm sieve fractions tend to include only
unidentifiable fragments of seed coats, occasional fungi
spores, and very fine bits of wood. This means that samples
analyzed to the level ofthe I-mm sieve fraction are generally
comparable with those analyzed through the complete sieve
series. Hence, nine of the 17 samples from Column 8-1/8-2
are comparable (Table 1). Unfortunately, the seed assem-
blage from the older Manasota Period samples (21-1, 21-2,
and SW Wall ash in Table 1) is not absolutely comparable
with the later Safety Harbor deposits of Column 8-1/8-2,
since analysis of the former group extended only to the 2-mm
sieve fractions. The wood data, however, are fully compara-
ble across the cultural and temporal sequence because
archaeological wood identifications are based primarily on
specimens recovered in the 4-mm, and to some extent the 2
mm, sieve fractions, i.e., all samples.
Beyond the particulars and strategy described above, the
laboratory analyses of the Shell Ridge Midden samples
proceeded according to general practice. Wood remains from
all samples were identified as the size and condition of the
specimens permitted. Generally, all plant materials captured
in the 4-mm and 2-mm sieves were completely sorted and
identified. Finer fractions were scanned under a dissecting
microscope for identifiable plant remains, but were not
otherwise partitioned.
Seed identifications were made using pictorial guides
(Martin and Barkley 1961) and comparative specimens
housed at FLMNH. Whenever possible, seed counts used in
this study are based on whole seeds and/or seed scars to
provide as realistic a count as possible. In other cases -
primarily nutshells and small sections of palm seeds (here
designated "palm family") individual fragments are
reported. Because the overall seed and nutshell counts from
the Shell Ridge Midden samples are relatively low, no
attempt was made to standardize the raw data from individ-
ual samples according to sample volume or weight.
Included among the non-wood remains described below is
a category of plant materials designated "unidentified soft
tissue." This is a catch-all group for small, unidentifiable
fragments of parenchymatous plant tissues. These starchy
materials may derive from nut meats, seed endosperms,
roots, tubers, or other similar unlignified tissues.
Wood charcoal was identified on the basis of three-
dimensional anatomy under magnifications ranging from
40x to 1200x, and using a combination of incident-light
(dissecting) and compound microscopes. Keys to diagnostic
anatomical features (Record and Hess 1942-1948; Urling and
Smith 1953; Wheeler et al. 1986) were the primary tools
used to identify wood remains, supported by reference to
comparative specimens housed at FLMNH. All identifica-
tions were pursued to the lowest possible taxon, usually to
On a final note about archaeological wood identification,
previous research (Newsom 1991a; Scarry and Newsom

1992) has demonstrated that a minimum number of at least
30 specimens of wood identified per sample is necessary
before the relative importance of individual taxa in a given
provenience can be estimated. The benchmark of 30 mini-
mum identified was employed in the Shell Ridge Midden
analysis. The relative frequencies of wood types (see below)
from samples with less than 30 specimens identified should
be interpreted as provisional.

Results of Analysis

The results of this analysis are presented in the sections
that follow beginning with general overviews of the samples
and plant identifications, then proceeding to specific details
based on individual samples. First the sample data are
examined in total, and as groups and stratigraphic sets
according to their primary constituents and how the quanti-
ties of these materials vary between the proveniences. The
classes of remains in each sample are summarized and
compared using counts, weights, and other general descrip-
tive measures. Next discussed are the plant identifications,
followed by comprehensive analyses of individual samples
and the constituent plant remains.

Sample Overview

Archaeobotanical summary data for the midden samples
appear in Table 2. In all, 395 wood specimens were identi-
fied, along with greater than 200 seeds or seed fragments.
On an individual sample basis, the amount of wood and
seeds present varied considerably, which is very typical of
midden deposits. Carbonized wood ranged from greater than
100 gm to less than 1 gm, or entirely absent (Table 2).
Likewise, seed remains ranged from a complete absence to as
many as 75 individuals (sample 8-2/136 cmbd).
Sample size partially explains this variability; fewer
remains were associated with smaller samples and more
remains were recovered with larger samples. However, using
the standardized measures of wood and seed density (grams
per liter and count per liter of sample, respectively), it
becomes apparent that real differences in sample content
exist. Indeed, some samples included considerable concentra-
tions of carbonized wood and/or seed remains (Table 2). For
example, sample 8-2/100-117 cmbd contained 31.1 gm of
wood per liter of sample, whereas the identically sized (.4
liters) sample from immediately above, 8-2/97 cmbd,
included only 3.1 gm of wood per liter (Table 2). The sample
from 109-135 cmbd also included a relatively dense concen-
tration of carbonized wood. The average wood density for the
complete sample assemblage is 5.5 gm/liter.
It is instructive to note that the actual wood weights and
numbers identified from the two samples with relatively
dense concentrations of wood show that the charcoal speci-
mens within the samples were generally small in size. The
two samples had fewer specimens of an identifiable size
(generally <4 mm in size) and comparatively light weights
(Table 2). This should make clear that the most information



24 LIZ
HE wamA 6awo C

Table 2. Summary data for Shell Ridge Midden samples.
Unit l Sample Wood
Volume Weight
(cm below datum) (lites (gm)

8-2, 80-92 cm 19.0 15.56"
8-2, at 97 cm .4 1.23
8-2, 100-117 cm .4 12.44
8-2, 109-135 cm .4 7.33
8-2, at 114 cm .2 1.72

8-2, 114-119 cm
8-2, 124-141 cm
8-2-1, at 130 cm
8-2-2, at 130 cm
8-2, at 136 cm
8-2, at 157 cm
8-2, at 163 cm
8-2, at 166 cm
8-1, at 176 cm, SE
8-1, at 176 cm, SW
8-1, at 181 cm
8-1, at 186 cm
20-1, at 354 cm
21-2, at 351 cm
SW Wall, ash

Combined Totals




a Indicates wood weight should vary slightly; 2-mm wood remains are

is gleaned when the three different wood measures are
examined in tandem. Thus, the apparently dense concentra-
tions of wood in the samples between 100 and 135 cmbd
possibly resulted because the charcoal was once subjected to
more fragmentation than charcoal in other areas sampled
(e.g., high foot traffic). In contrast, other samples, such as 8-
2/136 cmbd or the SW Wall ash deposit, produced greater
amounts of large, somewhat dispersed, charcoal fragments.
If and when additional data are collected, the combination of
measures (wood weight, density, MNI) as described here may
be useful to define activity areas and/or loci of greater or
lesser activity (high/low traffic), and/or feature function.
The summary data for the Shell Ridge Midden seed
remains, including nutshell, provide similar insights (the
right-hand columns of Table 2). As with the wood data, the
combination of descriptive measures, in this case, raw count
and seed density, helps to reveal patterns and enables better
interpretations of the various deposits. The column samples
from 97 cmbd and, in particular, 176 cmbd have exceptional
concentrations of seeds, ranging from 25 to 106 specimens
per liter (Table 2). Interestingly, wood remains were not
conspicuous among these particular samples, which pro-
duced very little or no identifiable wood. In fact, without the
aid ofmagnification, the presence of plant materials in these
samples would not have been apparent. That is, during

abundant and were not completely extracted as in other samples.

excavation plant materials were not observable; the samples
were specifically selected because of their relatively dark
appearance. We can now specify that the deep brown
coloration of these deposits is the result of abundant carbon-
ized seed remains.
Viewing the sample assemblage as a whole, it becomes
apparent from the counts and density data that seed remains
were most abundant and strongly associated with the dark
organic strata from approximately 130 cmbd to around 180
cmbd (Tables 1 and 2), though the samples from 157 cmbd
and 166 cmbd deviate from this observation. However, both
samples were analyzed only to the 2-mm sieve fraction;
therefore, the corresponding seed data may not be truly
representative. Two of the ash deposits sample 8-2-1/130
cmbd and the SW Wall sample produced two and 14 seed
remains, respectively, and both samples included relatively
large quantities of wood (Table 2). In contrast, another ash
sample, from the locus 124-141 cmbd, which was an area of
intense burning activity, produced no identifiable plant
remains whatsoever (Table 2). It appears that if once present,
all plant and other organic materials from this deposit
underwent complete combustion and conversion to ash. It is
noteworthy, however, that a considerable quantity of burned
shark dermal denticles (.42-mm sieve) is present (Kozuch,
this issue). Thus, preservation biases in the form of higher







Th~~m F A~ M~

8 991 Voo 51 4

-l--..-.- IlVC N n a. onat... nnnr. In-u.- 1--

Table 3. Archaeobotanical identifications, Shell Ridge Midden.
Taxon Common Name Plant Part

Avicennia germinans black mangrove wood
Bursera simaruba gumbo-limbo wood
Carya sp. hickory nutshell fragments
Celtis sp. hackberry seed (mineralized)
cf. Chiococca alba snowberry wood
Conocarpus erecta buttonwood wood
cf. Eugenia sp. stopper seed/fruit
Mastichodendron false mastic seed
Palmae palm family wood; seed fragments
Pinus sp. pine wood
Quercus virginiana live oak wood
Quercus sp., red group e.g., laurel oak wood
Quercus sp., oak, unidentified wood; nutshell frag-
indeterminate ments; nutmeat
Rhizophora mangle red mangrove wood; ?hypocotyl/
Sabal palmetto cabbage palm seed
Serenoa repens saw palmetto seed
Vaccinium sp. blueberry wood
Vitis sp. wild grape wood (vine)

NOTE: cf. = provisional identification.

intensity and/or longer term or repeated burning (e.g.,
cooking areas) may account for the absence of seeds and
other plant materials from sample 8-2/124-141 cmbd. The
differential presence of seed and wood remains is further
discussed below.

Archaeobotanical Identifications

A comparatively diverse array of plant taxa was identified
from among the Shell Ridge Midden samples. The
remains derive from at least 16 types of plants (Table 3).
Eleven wood taxa were identified, including pine (Pinus sp.,
section diploxylon), live oak (Quercus virginiana), and
another oak (Quercus sp.) belonging to the red oak anatomi-
cal group. Also identified were three members of the coastal
mangrove community including black mangrove (Avicennia
germinans), red mangrove (Rhizophora mangle), and button-
wood (Conocarpus erecta), as well as several other taxa
often associated with various types of hardwood hammock
vegetation and/or shell midden deposits, including gumbo-
limbo (Bursera simaruba), palm (Palmaceae, probably
cabbage palm [Sabal palmetto]), blueberry/sparkleberry
(Vaccinium spp.), and grape vine (Vitis sp.). Snowberry
(Chiococca alba), typically found in subtropical hardwood
hammocks and likewise often in association with shell
midden deposits, was provisionally identified from a single
sample (see below). Vaccinium spp., palms, and grape also
occur sporadically in various types of pine woodland and
scrublands, and in different edaphic associations (e.g.,
cabbage palm hammocks) (United States Department of
Agriculture, Soil Conservation Service [U.S.D.A.

The seed identifications (Table 3) include hickory nut
(Carya sp.), hackberry (Celtis sp.), false mastic (Mastic-
hodendron foetidissimum), acorn (Quercus sp.), cabbage
palm (Sabal palmetto), saw palmetto (Serenoa repens), and
provisionally identified stopper (Eugenia sp.). All of these
derive from trees or shrubs, and again are generally associ-
ated with mature hardwood forests. The saw palmetto, like
the pine and other taxa noted above, including perhaps the
red-type oak, may represent pine-dominated woodlands such
as the longleaf pine-turkey oak or pine flatwoods associations
(U.S.D.A. S.C.S. 1989). The seeds or fruit of each of these
taxa are edible; documented uses for Eugenia spp. are
generally for medicinal purposes (e.g., Ayensu 1981:135).
False mastic is relatively rare in the Sarasota area, where it
tends to occur almost exclusively on shell middens (G. Luer,
personal communication, 1997).

Contextual Analysis

The distributions of the various archaeological plant taxa
from the Shell Ridge Midden appear in Tables 4 and 5
according to their respective proveniences. The seed data are
direct counts of individual seeds and seed fragments (Table
4). As was apparent from the summary data presented in
Table 2, seed remains are most abundant in sample 8-2/97
cmbd and the dark organic deposits from 136-181 cmbd (the
latter include 67% of the absolute seed total; 83% of the total
if the organic concentration at 130 cmbd [sample 8-2-2/130
cmbd] is included). The right-hand column in Table 4 tallies
the absolute number of taxa represented in each sample,
showing that the heavily organic-stained deposits also
include the greatest diversity of species, with as many as
seven individual taxa. Conspicuous among these deposits are
acorn shell fragments, but they also include a nearly whole
carbonized acorn nutmeat (sample 8-2/163 cmbd), as well as
cabbage palm seeds, saw palmetto seeds and seed fragments,
and additional small palm family seed fragments (Pal-
maceae, most likely also cabbage palm). The types and
frequencies of remains in sample 8-2/97 cmbd are similar.
These all are relatively large seeds and seed parts readily
captured in the 4-mm and 2-mm sieve fractions. Thus, their
differential presence among the total deposits is not the result
of our internal sieve sizes, sample processing, or analysis
In contrast, seeds are absent from five of the seven samples
recovered from the deposits lying between about 100-135
cmbd. These samples are generally within the range of
comparability because one was completely analyzed, two
were analyzed to the level of the I-mm sieve fraction, while
the remaining two samples were processed through the 2-mm
fraction (Table 1). With this in mind, it seems that the
absence of seeds may accurately characterize these particular
strata. It was noted earlier that the feature sample from 124-
141 cmbd derives from a heavily burned locus. The absence
of seeds in this feature may be due to the fact that any plant
materials originally present probably did not survive the
intense burning. However, wood remains are generally well


b~rorPn~rrrrm~rl PPnProrm rr alrm Dm~D ~lm-m,

Table 4. Shell Ridge Midden seed and miscellaneous plant identifications. Column values are direct counts; values in parentheses indicate tentative taxonomic placement.
Seeds Other Remains
uid plant # Taxal
Period/Sample Cabbage Saw Palm False St Hackberr uid uid soft ?seedling >2mm Sample
Salm Palmetto Family Mastic paper Hacberry seed tissue mangrove

Safety Harbor
8-2. 80-92 cm
8-2. 97 cm
8-2. 100-117 cm
8-2. 109-135 cm
8-2. 114 cm
8-2. 114-119 cm
8-2. 124-141 cm
8-2-1. 130 cm
8-2-2. 130 cm
8-2. 136 cm
8-2. 157 cm
8-2. 163 cm
8-2. 166 cm
8-1. 176 cm SE
8-1. 176 cm SW
8-1. 181 cm
8-1. 186 cm
20-1. 354 cm
21-2. 351 cm
SW Wall. ash

Combined Totals

1 1 1

9 4
5 7

1 3

29 1

20 12

1 0

3 1 1 4 3
4 4 3 2 1 1 4 6

60 6 68 34 66 1 1 1 8 10 4 75
53 18 47 47 29 6 6 6 35 18

Table 5. Shell Ridge Midden wood identifications. Column values represent relative frequencies; values in parentheses indicate tentative taxonomic placement.
/Pe Live Red Indeterminate Gumbo- Black Red
Period/Sample MNI Pine Oak Oak Oak limbo Blueberry Manrove Manrove Button Snowberry Grapevine Palm # Taxa/Sample

Safety Harbor
8-2, 80-92 cm 30 .70 .16 .10 .03 4
8-2, 97 cm 6 .67 .17 (.17) 3
8-2, 100-117 cm 6 .83 (.07) 2
8-2, 109-135 cm 15 1.00 1
8-2, 114 cm 0 0
8-2, 114-119 cm 30 .90 .03 (.03) .03 4
8-2, 124-141 cm 0 0
8-2-1, 130 cm 40 .60 .23 .10 .07 4
8-2-2, 130 cm 30 .83 .17 2
8-2, 136 cm 40 .93 .07 2
8-2, 157 cm 17 .88 .12 2
8-2, 163 cm 42 .95 .02 .02 3
8-2, 166 cm 12 .83 .17 2
8-1, 176 cm SE 0 0
8-1, 176 cm SW 1 1.00 1
8-1, 181 cm 30 .60 .40 2
8-1, 186 cm 10 .80 .20 2
20-1,354 cm 40 .93 .07 2
21-2, 351 cm 5 .60 .20 (.20) 3
SW Wall, ash 41 .73 .05 .07 .10 .02 .02 6
Ubiquity 100 75 6 13 19 6 20 6 13 6 6 13


preserved in these samples (Table 2), so preservation biases
do not appear to explain the overall absence of seeds between
100 cmbd and about 135 cmbd. Depositional factors or
perhaps feature function may better explain wood-versus-
seed presence. For example, in a hide-smoking context or
other non-food-related activity, the residual fuelwood
remains would be expected to survive, whereas seeds,
representing primarily edible items, would be absent. While
sample size also may be a factor, since all but two of these
samples are relatively small (Tables 1 and 2), the size of the
samples is relative to the size of the individual deposits,
which on some level reflects use or function. Moreover, the
two largest samples, 8-2/114-119 cmbd and 8-2-1/130 cmbd,
contain only 0 and 2 seeds, respectively.
Turning now to specific samples and plant taxa, taxon
ubiquity is a standardization tool that paleoethnobotanists
employ to compare among site proveniences. Ubiquity
documents the appearance of a given taxon in individual
proveniences on a presence/absence basis only, thus factoring
out potential biases based on unequal fragment counts that
make specimens of seed and wood types that are prone to
fragmentation and breakage appear to have been more
important than was actually the case (Popper 1988:60-64).
For example, if a certain archaeological taxon (e.g., "seed
type seven") is present in a single provenience out of a total
of 10 proveniences sampled, it is assigned a ubiquity value of
10; if the same taxon were to appear instead in seven
separate proveniences, its ubiquity value would be 70. It
follows that a taxon that is fairly conspicuous (e.g., appear-
ing in 70% of the hypothetical site proveniences) was more
important or more intensively relied upon than one that is
recorded from one or a few proveniences.
Taxon ubiquity values for the individual seed types
recovered from the Shell Ridge Midden are shown at the
bottom of Table 4, which combines all time periods and
strata. Oak and cabbage palm seeds (not to mention "palm
family" seed fragments which may also derive from cabbage
palm) have ubiquity values of 60% and 68%, respectively.
They are identified from the dark organic midden levels, as
well as the ash-laden (hearth or fire-pit?) contexts. These two
taxa were probably important and regularly sought plant
foods for the people who inhabited the site. Saw palmetto,
another edible fruit, also is fairly prominent among the
proveniences tested, with a ubiquity value of 34%. In
contrast, hickory nut, mastic, stopper, and hackberry each
were identified from three or fewer proveniences, resulting
in ubiquity values that range from 6% to 18%, suggesting
that these plant foods were less important and/or were used
only on an occasional basis. Provisionally identified red
mangrove seedlings (sprouts), another potential food re-
source, were identified from three proveniences. Finally, if
the remains grouped under the category "unidentified soft
tissue" are residues of edible plants (e.g., rootstocks), then
another, if only minor, food resource is suggested. These are
present in five proveniences, resulting in a ubiquity value of
Also noteworthy among the non-wood plant remains is one

small fragment of carbonized parenchymatous tissue that
resembles gourd rind (Lagenaria siceraria). This specimen
was recovered with the deeply buried ash sample from the
southwest wall (Table 4, last sample, under "unidentified soft
tissue"). A definitive identification of this specimen was
precluded by its size and poor condition. If it is indeed gourd,
then the presence of another economic plant is indicated.
Uses for bottle gourds include containers, fish-net floats, and
seed oil, among others (Heiser 1979).
The sample-by-sample wood identifications from the Shell
Ridge Midden are shown in Table 5 according to the relative
frequencies of the individual wood types. The absolute
number of wood types per sample varied from one to six
species, and the sample with the greatest diversity was the
ash sample from near the bottom of the excavations. Three
samples (8-2/114 cmbd, 8-2/124-141 cmbd, and 8-1/176
cmbd) did not generally contain wood of a size and condition
suitable for identification, and, as earlier indicated, one of
these (8-2/124-141 cmbd) is from an ash deposit with no
identifiable plant remains.
On an individual provenience basis, pine is overwhelm-
ingly the predominant wood type, averaging 80% relative
frequency (range 60-100%). Live oak tends to be the second
most abundant wood type, and six of the 20 samples include
only pine and live oak. The red anatomical type of oak and
other wood taxa comprise no more than 10% to perhaps 20%
of the identified taxa in individual samples (Table 5).
Ubiquity values for the wood taxa from the Shell Ridge
Midden are listed at the bottom of Table 5. Pine again is
prominent and was identified in all proveniences with
identifiable wood (100% presence). Live oak was nearly as
ubiquitous, having been documented in 75% of the proven-
iences. Black mangrove, buttonwood, gumbo-limbo, and
palm each appeared in two or three proveniences, while the
rest of the taxa were identified from single contexts only.


With the exception of Little Salt Spring, the Central Gulf
Coast region where the Palmer site is located has gone
largely unstudied from a paleoethnobotanical perspective.
Thus, by any measure, the Palmer site's Shell Ridge Midden
research is a significant addition to the paleoethnobotany of
peninsular Florida, and the gulf-coastal region in particular.
Plant identifications from the Palmer site's Shell Ridge
Midden demonstrate that the present-day forest composition
of the central gulf coast, with its distinctive combination of
temperate (e.g., blueberry, hickory) and tropical (e.g., false
mastic, gumbo-limbo, mangrove) elements, was long ago
established. Plants of both biogeographic affinities appear
throughout the midden strata, suggesting that the local
environment has changed very little over the past two
millennia, or at least that the current floristics approximate
the vegetation and climatic conditions at ca. A.D. 300 and
A.D. I100.
Most of the carbonized wood, given the admixture of
species in each provenience, probably represents spent

1998 Vot. 51(4)


NLiTrURc fMltNfAgflH nnrt fOaa Al OfLtflL flflfls u IrVILDLfl 1,
IW -. AI I IVI l l,/tb] l ltl~..lhi l~~ Itlll llll l ]

Table 6. Wood ubiquity scores for Palmer site's Shell ridge Midden and selected south Florida sites."
Josslyn Cash Big Mound Horr's
d Shell Ridge Buck Key sl Pineland Granada
Wood I2 8 island 3 Mound Key Island
8LL32 8CH38 8CH10 8CR208/209
Black 20 74 100 80 75 50 77 6
Button 13 52 16 25 17 27 21
Red 6 43 80 64 50 60 60
White 4 4 4 12
Gray Nicker 9
Seagrape 30 4 23
Blueberry 6
Live Oak 75 4 20 17 32
Mastic 9 4
Cocoplum 4 10
Grape vine 6
Gumbo-limbo 19
Palm 13
Rapanea 17 50
Sapotaceae 4
Snowberry 6
Stopper 4 (4)
Wild lime 4
Red Oak 6
Pine 40 100 88 100 100 91 100
Cypress 4
Total Taxa 11 10 3 11 5 4 9 7

SSource data are from Newsom (1991a), Newsom and Scarry (1998), Scarry (1984), and Scarry and Newsom (1992).
b Pineland data are from Operation I.

fuelwood. Some fragments, palm for example, may have
come from the remains of house structures (e.g., roof thatch)
or similar construction materials. Pine is present in every
sample with identifiable wood, and live oak is second in
terms of sample representation (Table 5). To the extent that
these data are representative, pine and live oak were the most
important fuelwoods used by Shell Ridge Midden inhabit-
Overall, wood-species selection based on the Shell Ridge
Midden contexts is consistent with patterns of fuelwood use
documented at other subtropical Florida sites. The Shell
Ridge Midden data are compared in Table 6 with wood
charcoal assemblages from seven other south Florida sites.
Pine is clearly prominent among all site assemblages,
ranging from a low of 40% presence at Buck Key to 88% at
Pineland, 91% at Horr's Island, and, strikingly, 100%
presence at all the other sites. Live oak is documented from

five sites, but except for the Shell Ridge Midden (75%
presence), it is not nearly as conspicuous. Instead, mangrove
taxa, especially black mangrove, tend to assume dominance
at the other sites, with ubiquity values ranging between about
10% to 100% (Table 6). Black mangrove from the Shell
Ridge Midden nevertheless has a site ubiquity value of 20%,
and two other mangrove taxa also are present (Table 6).
Gumbo-limbo, another tropical tree, was recorded from 19%
of the proveniences at the Shell Ridge Midden. It is instruc-
tive to note here that the Shell Ridge Midden is located
farthest north of all the sites listed in Table 6, very near the
northern limits of the range for most mangroves and other
tropical trees (see Little 1978, in particular maps 168, 172,
190, 219, and 238; also Myers and Ewel 1990:527). That the
Shell Ridge Midden is exclusive among the sites shown in
Table 6 in the presence of the more temperate species (red-
group oak and blueberry) is another reflection of its biogeo-



TnM R~ CRMA A,.rRfoPoL OC1-T 19 O l

graphic position.
The wood charcoal assemblage from the Shell Ridge
Midden, with the combination of temperate and tropi-
cal/subtropical species, is actually very similar to carbonized
wood assemblages from midden deposits at Hontoon Island
(Newsom 1987a:56), Groves' Orange Midden (Newsom
1994:409), Edgewater Landing, 8V0115 and 8VO1705,
(Russo et al. 1989:73-83), and Seminole Rest (Ruhl 1995),
all located slightly farther north than the Shell Ridge
Midden, but on the east side of the peninsula in the Indian
River and St. Johns River systems (see also Newsom 1985),
and in a different culture region (St. Johns tradition). For
example, among Edgewater Landing samples pine ranged
from 12-93% of the identified wood. Black mangrove was
second in overall frequency of appearance, followed by live
oak and three other taxa. Dominant taxa in the Seminole
Rest identifications were oak, hickory, and bay, but also
identified were pine and, provisionally, red mangrove,
among others (Ruhl 1995:43-44, 48).
The overall pattern for fuelwoods suggested by the penin-
sular sites as a group is one of relatively localized collection
of wood for the various purposes in which it was used. Such
overall consistency of species selection, with an emphasis on
pine, live oak, and mangroves, is certainly related to the local
abundance and availability of these trees in the subtropical,
coastal, and/or near-shore environments. Sites like the Shell
Ridge Midden and Hontoon Island, located at or near the
temperate/subtropical boundary, include a greater or more
equal representation of temperate relative to subtropical taxa,
but in general have assemblages very similar to the other
prehistoric sites and occupations (Table 6).
So similar a pattern across a considerable time span (five
millennia all together) also may indicate a certain level of
cultural preference for different wood species. For example,
buttonwood and black mangrove may have been routinely
employed to smoke fish, pine for quick ignition and to smoke
hides, and oak for a hot, enduring cooking fire (Newsom
1991a). The matching or blending of species-specific
characteristics and use/function may be evident in inverse
perspective also, because other common woods, such as
cypress, appear infrequently among the carbonized wood
assemblages. That cypress was present and available is
certain based on pollen data (e.g., Watts and Hansen 1988)
and by the presence of uncarbonized cypress wood debitage
at several sites, demonstrating functions other than fuelwood
and suggesting very thoughtful, preferential use.
The non-wood plant identifications from the Shell Ridge
Midden which represent seeds and other reproductive parts
from at least ten edible or medicinally valuable species
provide additional insights into indigenous subsistence
practices. Oak, hickory, grape, and palm fruits were com-
monly used as food sources by prehistoric people throughout
Florida, and spanning the full occupational history of the
state. Blueberry, or huckleberry, has been documented
previously only at two sites: seeds from Hontoon Island on
the St. Johns River (Newsom 1987a) and carbonized wood at
the Fort Walton Period Lake Jackson site (Newsom 1991b).

Hackberry and mastic fruit consumption has been inferred
from seed and wood presence at a number of coastal sites
south of the Palmer site, particularly in the Charlotte Harbor
region, and also at sites on the southeast coast and in the
Florida Keys (Scarry et al. 1989). Hackberry also was
identified from Seminole Rest on the east coast (Ruhl 1995).
The sprouts or young seedlings of red mangrove are edible,
and while this is a tentative identification for the Shell Ridge
Midden specimens, Scarry has identified red mangrove
sprouts in Charlotte Harbor material (Scarry and Newsom
Finally, it is possible to suggest medicinal and other uses
of plants from the Shell Ridge Midden, based broadly on
ethnographic medicinal practices. The snowberry's presence
(tentative identification from wood) at the Shell Ridge
Midden may be related to the use of its roots for muscle
cramps and other ailments (Ayensu 1981:161). Red man-
grove sprouts have been employed to treat backache during
pregnancy(Ayensu 1981:159). Gumbo-limbo, cabbage palm,
and species of stopper also are known to be used for home
medicinal remedies in the Caribbean region (Ayensu
1981:135, 139; Honychurch 1986:20). These plants may
have been similarly used by Florida Indians. Gums and resin
for glues and medicinal purposes may have been extracted
from gumbo-limbo and pine, while tannins for hide-curing
and other purposes were potentially available from the
mangrove taxa (Austin and McJunkin 1978).


This limited study of plant remains from the prehistoric
Shell Ridge Midden at the Palmer site has been productive,
yielding insights into species selection and environment. No
definitive evidence of cultivated and/or domesticated plants
was discovered. This is relatively consistent with
archaeobotanical analyses from other areas of the central and
southern peninsula, which suggest that maize and other
crops were a minor component of prehistoric subsistence, if
they were grown at all.
Instead, plant cultivation prehistorically in peninsular
Florida appears to have been limited to small-scale garden-
ing and/or low-maintenance crops, such as gourds and
perhaps Zamia sp. (Newsom 1991a; Newsom and Scarry
1994; Scarry et al. 1989). That gourds were useful plants
worthy of care and propagation is suggested by the apparent
uses to which the various types of gourd and gourd/squash
were put, including fishnet floats (e.g., Key Marco and
perhaps also Hontoon Island and Pineland [Cutler 1975;
Gilliland 1975; Newsom and Scarry 1998]), bottles and
containers (e.g., Windover, Little Salt Spring), and perhaps
also for rattles, seed oil (the rinds are bitter and inedible, but
the seeds and oil are edible), and even medicinal purposes
decoctionss of flower or fruit for measles and small pox;
seeds as vermifuge; fruit pulp and rinds said to contain
anthelmintic substances; root for jaundice, and more [see
Ayensu 1981]).
Viewed collectively, the paleoethnobotanical data from

8 991 Vol- 51(4)



Florida seem to indicate that plant-food use in peninsular
locations focused on the natural diversity of edible species.
This may have manifested itself in an essentially opportunis-
tic collection strategy that took advantage of abundant and
assorted, seasonally available wild fruits, nuts, and grains.
This is not meant to imply blind and unplanned collection.
Nevertheless, it seems clear from the repeated appearance of
the same basic suite oftaxa at site after site, that what may
be in evidence is the use of an optimal set of resources
(Winterhalder and Goland 1997), which implies more
focused, deliberate plant collection and use. Likewise,
fuelwood selection appears to have been strongly influenced
by three factors: biogeography (latitude, inland versus
coastal), local floristics (stand composition, species densi-
ties), and inherent species-specific wood characteristics
(high- versus low-density woods, grain orientation, etc.). The
Palmer site's Shell Ridge Midden identifications, few as they
are, fit this pattern of focused wild plant use, with local
forests providing, at a minimum, wood, nuts, and fresh
fruits. In terms of edible species, both early season (mastic,
blueberry, hackberry) and summer-fall (nut foods, grape, and
palm berries) plant foods are represented among the Shell
Ridge Midden deposits. Since in general only single samples
were collected and studied from individual midden deposits
or strata, future testing might be directed toward replicating
samples from units or strata to assess the representativeness
of the identifications presented here.


Special thanks to Marion Almy, George Luer, and William H. Marquardt for
making this research possible, and to Corbett Torrence for facilitating the
archaeobiological sample collections. Funding was provided by a Florida
Department of State, Division of Historical Resources "Special Category"
grant. The Florida Museum of Natural History and the Ethnobiology Labora-
tory under the direction and supervision of Elizabeth S. Wing provided
laboratory space and general support for this study. I am grateful also to Bill
Marquardt, George Luer, and anonymous reviewers for their comments and
review ofearlier drafts of this report.

References Cited

Austin, D.F., and D.M. McJunkin
1978 An Ethnoflora of Chokoloskee Island, Collier County, Florida.
Journal ofthe Arnold Arboretum 59:50-67.
Austin, Robert J., and Michael Russo
1989 Limited Excavations at the Catfish Creek Site (8S0608), Sarasota
County, Florida. Report prepared for Palmer Venture, Inc., Sarasota,
by Piper Archaeological Research, Inc., St. Petersburg.
Ayensu, Edward S.
1981 Medicinal Plants of the West Indies. Reference Publications, Inc.,
Algonac, Michigan.
Brown, J.G., and A.D. Cohen
1985 Palynologic and Petrographic Analyses of Peat Deposits, Little Salt
Spring. National Geographic Research 1(1):21-31.
Cutler, Hugh C.
1975 Two Kinds of Gourds from Key Marco. In The Material Culture of
Key Marco, Florida by Marion S. Gilliland, pp. 255-256. University
Press of Florida, Gainesville.
Cowan, C. Wesley, and Bruce D. Smith
1993 New Perspectives on a Wild Gourd in Eastern North America. Journal
ofEthnobiology 13(1):17-54.

Decker, Deena S., and Lee A. Newsom
1988 Numerical Analysis of Archaeological Cucurbita pepo Seeds from
Hontoon Island, Florida. Journal of Ethnobiology 8(1):35-44.
Doran, Glen H., Dave N. Dickel, and Lee A. Newsom
1990 A 7,290-year-old Bottle Gourd from the Windover Site, Florida.
American Antiquity 55:354-360.
Gilliland, Marion S.
1975 The Material Culture of Key Marco, Florida. University Press of
Florida, Gainesville.
Heiser, Charles B.
1979 The Gourd Book. University of Oklahoma Press, Norman.
Honychurch, Penelope N.
1986 Caribbean Wild Plants and their Uses. Macmillan Publishers, Ltd.,
Hutchinson, Dale F., Clark Spencer Larsen, Margaret J. Schoeninger, and
Lynette Norr
1998 Regional Variation in the Pattern of Maize Adoption and Use in
Florida and Georgia. American Antiquity 63:397-416.
Johnson, William G.
1990 The Role of Maize in South Florida Aboriginal Native Societies: an
Overview. The Florida Anthropologist 43:209-214.
Kessel, Morton H.
1991 The Role of Maize in South Florida Aboriginal Societies: a Comment.
The Florida Anthropologist 44:94.
Little, Elbert L., Jr.
1978 Atlas of United States Trees, Volume 5, Florida. Miscellaneous
Publication No. 1361, U.S.D.A. Forest service, Washington, D.C.
Luer, George M.
1977a Excavations at the Old Oak Site, Sarasota, Florida: a Late Weeden
Island-Safety Harbor Period Site. The Florida Anthropologist 30:37-
1977b The Robert's Bay Site, Sarasota, Florida. The Florida Anthropologist
1992 The Boylston Mound: a Safety Harbor Period Shell Midden; with
Notes on the Paleoenvironment ofSouthern Sarasota Bay. The Florida
Anthropologist 45:266-279.
1995 The Brookside Mound, Sarasota County, Florida: Noteson Landscape,
Settlement, Scrub Habitat, and Isolated Burial Mounds. The Florida
Anthropologist 48:200-216.
Luer, George M., and Marion M. Almy
1982 A Definition of the Manasota Culture. The Florida Anthropologist
Martin, A.C., and W.D. Barkley
1961 Seed Identification Manual. University ofCalifornia Press, Berkeley.
Milanich, Jerald T.
1987 Corn and Calusa: DeSoto and Demography. In Coasts, Plains, and
Deserts: Essays in Honor of Reynold J. Ruppe, edited by S.W.
Gaines, pp. 173-184. Arizona State University Anthropological
Research Papers No. 38, Tempe.
1994 Archaeology ofPrecolumbian Florida. University Press of Florida,
Miller-Shaivitz, P., and M.Y. Iscan
1991 The Prehistoric People of Fort Center: Physical and Health Character-
istics. In What Mean these Bones, edited by M.L. Powell, P.S.
Bridges, and A-M.W. Mires, pp. 131-147. The University of Alabama
Press, Tuscaloosa.
Myers, Ronald L., and John J. Ewel
1990 Ecosystems ofFlorida. University ofCentral Florida Press, Orlando.
Newsom, Lee A.
1985 Analysis of Wood Artifacts. In Archaeological Site Types, Distribu-
tion, and Preservation within the Upper St. Johns River Basin,
Florida, edited by B. Sigler-Eisenberg, pp. 140-155. Florida State
Museum Miscellaneous Project Report SeriesNumber27, Gainesville.
1987a Analysis of Botanical Remains from Hontoon Island (8Vo202),
Florida: 1980-1985 Excavations. The FloridaAnthropologist40:47-
1987b Analysis of Archaeobotanical Remains from Cutler Ridge, Miami.
Unpublished Laboratory Data, Center for Archaeological Investiga-
tions, Southern Illinois University, Carbondale.
1989 Botanical Analysis. In Phase II Archaeological Excavations at the
Riverbend Site (8Vo2567), Volusia County, Florida by Michael
Russo, Janice R. Ballo, Robert J. Austin, Lee Newsom, Sylvia
Scudder, and Vicki Rowland, pp. 49-56. Report prepared for Charles

N~nv ~ ~ ~ ~ I~IUlr1R~ C_ LUI~~~NI-LL~ L

* o-n nr\or\~ru~r D~on ar~ r7 E~n D~n ~m~RI


E. Burkett & Associates, Inc. by Piper Archaeological Research, Inc.,
St. Petersburg.
1991 a Horr's Island Archaeobotanical Research. In Final Report on Horr 's
Island: The Archaeology of Archaic and Glades Settlement and
Subsistence Patterns, Part II, by Michael Russo, Ann S. Cordell, Lee
ANewsom, and Sylvia Scudder, pp. 591-644. Report prepared for Key
Marco Developments, Marco Island, Florida by Department of
Anthropology, Florida Museum of Natural History, Gainesville.
1991b Plant Identifications from Lake Jackson. Report submittedto Claudine
Payne, Department of Anthropology, Florida Museum of Natural
History, Gainesville.
1994 Archaeobotanical Data from Groves' Orange Midden (8Vo2601),
Volusia County, Florida. Florida Anthropologist 47:404-417.
1998 The Paleoethnobotany of Windover (8BR246): An Archaic Period
Mortuary Site in Central Florida. In Multidisciplinary Investigations
of an Archaic Cemetery: Windover, Florida, edited by Glen H.
Doran. University Presses of Florida, Gainesville (in preparation).
n.d. Archaeobotanical Analysis of Waterlogged Materials from Little Salt
Spring, Sarasota County, Florida. Unpublished Laboratory Data,
Center for Archaeological Investigations, Southern Illinois University,
Newsom, Lee A., and Matthew Mihlbachler
n.d. 32,000-year-old Cucurbitapepo SeedsfromNorth Florida. Nature (in
Newsom, Lee A, Wendy Natt, and Robin Brown
1998 Calusa Wood and Fiber Industries: Wood Debitage and Fiber Cordage
from Pineland Mound A Midden Deposits. In The Archaeology of
Pineland: a Coastal Southwest Florida Village Complex, A.D. 100-
1600, edited by Karen J. Walker and William H. Marquardt. Institute
of Archaeology and Paleoenvironmental Studies, Monograph Number
3, University of Florida, Gainesville (in preparation).
Newsom, Lee A., and Irvy R. Quitmyer
1992 Archaeobotanical and Faunal Remains. In Excavations on the
Franciscan Frontier by Brent R. Weisman, pp. 206-243. University
Press of Florida, Gainesville.
Newsom, Lee A., and C. Margaret Scarry
1994 Homegardens and Mangrove Swamps: Pre-Columbian PlantUseatthe
Pineland Site, Southwest Florida. Paper presented at 51st annual
meeting of the Southeastern Archaeological Conference, Lexington.
1998 Homegardens and Mangrove Swamps: Pineland Archaeobotanical
Research. In The Archaeology of Pineland: a Coastal Southwest
Florida Village Complex, A.D. 100-1600, edited by Karen J. Walker
and William H. Marquardt. Institute of Archaeology and Paleo-
environmental Studies, Monograph Number 3, University of Florida,
Gainesville (chapter submitted for publication).
Newsom, Lee A., S. David Webb, and James S. Dunbar
1993 History and Geographic Distribution ofCucurbita pepo Gourds in
Florida. Journal ofEthnobiology 13:75-97.
Panshin, A.J., and C. deZeeuw
1980 Textbook of Wood Technology (4th edition). McGraw Hill, New
Pearsall, Deborah M.
1989 Paleoethnobotany: A Handbook of Procedures. Academic Press,
New York.
Phillips, E.W.J.
1941 The Identification of Coniferous Woods by their Microscopic Struc-
ture. The Journal of the Linnean Society of London 52:259-320.
Popper, Virginia S.
1988 SelectingQuantitativeMeasurements inPaleoethnobotany.InCurrent
Paleoethnobotany, edited by C.A. HastorfandV.S. Popper, pp. 53-71.
University of Chicago Press, Chicago.
Record, Samuel J., and Robert W. Hess
1942-1948 Keys to American Woods. In Tropical Woods 72:19-29 (1942),
73:23-42 (1943), 75:8-26(1943), 76:32-47(1944),85:1-19 (1946),
94:29-52 (1948).
Ruhl, Donna L.
1995 Paleoethnobotanical Investigations. In Final Report on the Archaeo-
logical Investigations at the Seminole Rest Site (CANA-0-
63/8V0124), Canaveral National Seashore, Volusia County,
Florida, edited by E.A. Horvath, pp. 41-58. Southeast Archaeological
Center, National Park Service, Tallahassee.
Russo, Michael, Ann Cordell, Lee Newsom, and Robert Austin
1989 Phase III Archaeological Excavations at Edgewater Landing, Volusia

County, Florida. Report prepared for Radnor/Edgewater, Inc.,
Edgewater, Florida by Piper Archaeological Research, Inc., St.
Russo, Michael, and Irvy R. Quitmyer
1996 Sedentism in Coastal Populations of South Florida. In Case Studies in
EnvironmentalArchaeology, edited by E.J. Reitz, L.A. Newsom, and
S.J. Scudder, pp. 215-231. Plenum Press, New York.
Scarry, C. Margaret
1984 Paleoethnobotany oftheGranada Site. InArchaeology and History of
the Granada Site, Volume I: Excavations at the Granada Site, by
J.W. Griffin, S.B. Richardson, M. Pohl, C.D. McMurray, C.M.
Scarry, S.K. Fish, E.S. Wing, L.J. Loucks, and M.K. Welch, pp. 181-
248. Division ofArchives, History, and Records Management, Florida
Department of State, Tallahassee.
Scarry, C. Margaret, and Lee A. Newsom
1992 Archaeobotanical Research in the Calusa Heartland. In Culture and
Environment in the Domain of the Calusa, edited by William H.
Marquardt, pp. 375-401. Institute of Archaeology and Paleo-
environmental Studies, Monograph No. I, University of Florida,
Scarry, C. Margaret, Lee A. Newsom, and Marilyn A. Masson
1989 Calusa and Tequesta Plant Use: Evidence Gleaned from Archaeo-
botanical Data. Paper presented at the 46th annual meeting of the
Southeastern Archaeological Conference, Tampa.
Urling, G.P., and R.B. Smith
1953 An Anatomical Study of Twenty Lesser Known Woods from Florida.
Quarterly Journal of the Florida Academy ofScience 16:163-180.
United States Department of Agriculture, Soil Conservation Service
1989 26 Ecological Communities ofFlorida. United States Department of
Agriculture. Reprinted by The Florida Chapter, Soil and Water
Conservation Society, Gainesville. Originally published, 1985.
Watts, William A., and Barbara C.S. Hansen
1988 Environments ofFlorida in the Late Wisconsin and Holocene. InWet
Site Archaeology, edited by Barbara A. Purdy, pp. 307-323. The
Telford Press, Caldwell, New Jersey.
Wheeler, E.A., R.G. Pearson, C.A. LaPasha, T. Zack, and W. Hatley
1986 Computer-Aided Wood Identification. North Carolina Agricultural
Research Service, Bulletin 474. Raleigh.
Winterhalder, Bruce, and Carol Goland
1997 An Evolutionary Ecology Perspective on Diet Choice, Risk, and Plant
Domestication. In People, Plants, and Landscapes: Studies in
Paleoethnobotany, edited by KristenJ. Gremillion, pp. 123-160. The
University of Alabama Press, Tuscaloosa.

1998 VOL. 51(4)





Southwest Florida Archaeological Society, P. O. Box 9965, Naples, Florida 34101
E-mail:'arlee@naples. net

As the waters of the Gulf of Mexico rose during
southwest Florida's Late Archaic Period, they cov-
ered campsites of the area's early settlers, except for
those on higher ground, such as the sites on Horr's and
Useppa islands, the dune ridge on south Marco Island, and
those well inland. In the Naples Gulf Shore area, most of the
sites excavated to date have been of the Glades Period;
indeed, sherds from the entrance to Naples Bay were used in
establishing a Glades-period pottery sequence (Goggin
1939:35-40). Somewhat earlier is the Firebreak site, located
on a 2-m-high sand dune which contained fiber-tempered
pottery radiocarbon dated at approximately 2900 B.P.
(Beriault 1979).
New aspects of the area's early settlement are being
revealed as beach areas are built up and real estate develop-
ers operate their bulldozers farther inland, followed by
those, like members of the Southwest Florida Archaeological
Society (SWFAS), who would ensure that not all evidence of
these early settlements is lost. Such was the case with a site
designated as 8CR231 and called Heineken Hammock, after
the bottles latter-day hunters left on its surface. The site is
located on a limestone unconformity in a slough area 8.8 km
(5.5 mi) east of the City of Naples. Excavation there discov-
ered marine shell that was radiocarbon dated at 4500 years
B.P. The shell was recovered from the same test pits as
sherds of sand-tempered plain and later pottery types which
are associated with the Glades I Late Period.
The dates provided by the marine shell are of a time
generally regarded as the beginning of a "Pre-Glades I Late"
period characterized by the introduction of Orange Plain
(fiber-tempered) pottery (Widmer 1988:Table 5). Sand-
tempered plain does not appear in most currently accepted
chronologies until the terminal Archaic or Florida Transi-
tional periods, ca. 3000 B.P.
In a discussion of sites on nearby Marco Island, Russo
(1996:193) noted a lack of dates to help define the temporal
range of the Late Archaic (fiber-tempered ceramic era) and
Middle Archaic (pre-ceramic) periods. While findings at
Heineken Hammock were modest, its artifacts and radiocar-
bon dates may contribute to the general knowledge of the
peopling ofthe region and, more specifically, to the lifeways
of early coastal dwellers. The site also adds to the growing
fund of information about "corridor" sites presumed way-
stations between coastal communities and the interior, which

also were used as bases for hunting and gathering by
extending this use to an early, possibly pre-ceramic period.


Heineken Hammock (8CR231) is located on a rise in the
limestone base of a swampy area, forming the south bank of
a shallow slough which historically, and more certainly prior
to modern drainage, would have held water through most of
the year. It is conceivable that the slough and other low areas
in the region were flooded as the climate became wetter
during the Late Archaic Period (Widmer 1988:165). The
slough remnant is now preserved as a pond, surrounded by
the one-story houses of Golden Gate's Berkshire Village
(Figures 1 and 2). The site now is part of a green-space area
set aside by the developers and guarded by the community
organization, which prizes its archaeological distinction.
Heineken Hammock was first noticed in February, 1983,
by three SWFAS members, John Beriault and Gail and
Charlie Strader, during a pre-development survey conducted
in cooperation with then Collier County Environmentalist
William Vines. It was one of several areas that were selected
from aerial photographs as likely to contain archaeological
sites (and the only one that investigation proved to have been
occupied). The trio, following a clearing apparently opened
in the brush cover by light earth-moving equipment, found
some potsherds and named the site for the green beer bottles
left behind by people hunting in what was then a wilderness
area. Five years later, in November, 1987, SWFAS obtained
permission from the developer to investigate the site's
prehistory (Figures 3 and 4).
Two test pits were dug, one 1.5 x 2 m, the other 1 x I m,
and a 50 x 50 cm column sample was obtained from the west
wall of Test Pit 1, in what proved to be SWFAS's most
elegant dig. Saturday, November 14, the lady of a nearby
house brought to the site a silver tray with her finest china
and the crew (after washing hands as best it could from a
canteen) was served a proper afternoon tea.
When material from the site was analyzed in 1993, marine
shell from the lowest level (just above the limestone cap at a
depth of about 30 cm) yielded a radiocarbon date of 3930
70 years before present or cal. 2155-1745 B.C., which the
investigators considered early for the quite sophisticated
thin-walled, well-compacted, sand-tempered-plain pottery



VOL. 51 NO. 4

.Tn.. F .....A ANTR.opo.......T 1998 Vu 514



Pine Rde I Rd
1 ,m

Figure 1. Location of 8CR231, Heineken Hammock, shown relative to the City of Naples, Florida.


8 991 VOL $1(4)


Figure 2. Location of the Heineken Hammock site in a designated open-space area (Outparcel B) within the Berkshire
Village subdivision, a suburb of Naples. The expanded view of the outparcel shows the irregular outline of the hammock,
the approximate boundary of 8CR231 (cross-hatching), and the locations of SWFAS's test units.

found in the same level. So it was decided to try to located
other dateable material at the same level in an extension of
Pit No. 1. SWFAS members Jean Belknap, Walt Buschel-
man, and Art Lee dug two narrow trenches, a shovel test,
and two post holes (all later called Test Pit 3) abutting the
southern edge of Test Pit 1 but found nothing dateable, and
it was obvious that a more extensive search was needed. This

was performed on January 21-22, 1995, when seven 1 x 1 m
pits were dug to bedrock at about 30 cm. On the basis of
surface indications the pits were located somewhat to the east
of the first exploration, three of them being contiguous
(Figure 2). A plane table was used to locate pits and eleva-
tions on a plat map.
The test pits, oriented to magnetic north, were dug in



8 991 Vot 51 4)


Figure 3. SWFAS members excavating at Heineken Hi

arbitrary 10-cm levels, with the exception of Test Pit 4,
whose first level was made somewhat greater than 10 cm to
allow for an irregularity believed to have been caused by the
uprooting of a nearby tree. Flat-bladed shovels were used for
schnitting unpromising areas, along with trowels, brushes,
wooden picks, and 6.4-mm (1/4-in) screens. Recovered
materials were taken to the Craighead Laboratory at the
Collier County Museum in Naples for washing, identifica-
tion, analysis, and preservation. Adequate shell for radiocar-
bon dating was found in the same (30 cm) levels as pottery
in two pits, but not in direct as-
sociation with the sherds, and ',..j --"
samples were submitted to Beta ..
Analytic Radiocarbon Dating
Laboratory in Miami for analy- "

Site Environment


The Heineken Hammock site
is situated on the north side of
an ovoid .6 hectare (1.5 acre)
hardwood/cabbage palm "is-
land," which is one of several
elevated areas found in a loose
cluster near the northern termi-
nus of Rattlesnake Hammock
Slough (T49S, R26E, Section
32). Elevation at the site is ap-
proximately 4.15 m amsl. Al-
though the area is quite flat, it is Fu

an extension of a drainage pattern that
once fed into Haldeman Creek and Naples
Bay before being disturbed by modern
development. A part of an original marsh
to the north of the Heineken Hammock
site has been retained as a pond.
The high areas are natural bedrock
nonconformities that may be the remnants
of Pleistocene marine reefs or ridges (Mis-
simer 1984:392-3) commonly found in
linear bands several kilometers inland
from the present-day coastline. Their
slight elevation, 80 cm or more above their
surroundings, made them attractive as
prehistoric campsites.


Soils at the site generally fit the Collier
County Soils Survey's description of
Hallandale Fine Sand (Yamataki
immock. 1990:11): "Typically, the surface layer is
very dark gray fine sand about 3 inches
(7.8 cm) thick. The subsurface layer is
grayish brown fine sand to a depth of about 9 inches (22.9
cm). The subsoil is yellowish brown fine sand to a depth of
about 12 inches (30.5 cm). Limestone bedrock is at a depth
of about 12 inches."
Although the soils encountered in the test pits were
darkened in varying degrees by humic discoloration, they
generally corresponded to this description, lightening from
a dark grayish- brown, 10YR2/2 (Munsell Color 1975) in the
upper layers to 10YR4/4 at the 30-cm depth, where the
limestone base was generally encountered. Figure 5 shows

-C':~r~~glR1' iv., .

4. SWFAS members excavating Test Pit 2.

"trv~sr L u

Ln Yl Ai. HvmEKEN HAmm 22WD7U

1I I I I
0 50 100 150 200

S Very dark gray sand (Munsel 10YR 2/2) containing a scattering of potsherds. bone.
marine shell and unburned limestone fragments as well as forest floor litter.

j11T] Soil lightens to a grayish-brown fine sand (Munsel 10YR 3/2): it contained the
heaviest concentration of cultural material.

Color of soil changes to a lighter yellowish-brown (Munsel 10 YR 4/4). cultural
material is less frequent.

S The limestone bedrock has an irregular surface with shallow depressions.

Marks indicate the relative abundance of cultural material, not individual artifacts.

Figure 5. Profile of Test Pit 1. The sequence of soils in this test pit was repeated in other excavation units
at the site.

. -. i

Figure 6. Sanibel Incised pottery sherds from Test Pit 9, Level 1 showing characteris-
tic chevron incisions.





Tnr n Au vrrnuwflDO flCI-T 1998 VoL. 5 1 M

Table 1. Floral species present at Heineken Hammock.

Common Name Latin Name

Canopy Trees

Cabbage palm Sabal palmetto
Live oak Quercus virginiana
Red bay Persea borbonia
Sweet bay Magnolia virginiana

Midstory Vegetation

Hog plum Ximenia americana
Willow bustic Dipholis salicifolia
Camphor wood Myrcianthesfragrans
Wax myrtle Mmyrica cerifera

Understory Plants

Wild coffee Psychotria nervosa
Boston fern Nephrolepis exaltata
Fern sp. Inelypteris sp.
Bracken fern Pteridium aquilinum
Poison ivy Toxiccodendron radicans

this soil profile in Test Pit 1. The sequence of soils was
repeated in all other test pits.


Heineken Hammock is located in an area which, before
modern development and concomitant drainage, consisted
of low-lying dwarf cypress strands interspersed with areas of
slash pine and saw palmetto. Its floral inventory, typical of
such live oak/cabbage palm hammocks, is detailed in Table



Potsherds were recovered from all of the site's 10 test pits
and the column sample. Total yield was 619 sherds weighing
3176.7 gm (Table 2) representing three ceramic types. Test
Pit 9 produced the greatest amount, 1656.4 gm, with adjacent
Test Pit 4 yielding 472.4 gm and Test Pit 1, some 9 m west
of Test Pit 9, containing 292.2 gm. Other excavation units
contained lesser amounts.
The most common pottery is sand-tempered plain which
was recovered from nearly every level of every test pit. Ofthe
several rim sherds large enough to permit accurate measure-
ment, analysis showed that they came from vessels with
estimated orifice diameters of either 20 or 24 cm. Lips of
nine of the rim sherds are rounded and seven are beveled; six
rim sherds are incurved, seven are vertical, and three are
outcurved. Most body sherds, 107, are 6-8 mm thick, 33 are
between 9 and 11 mm, and 22 are between 4 and 5 mm. Of

the 16 rim sherds, five are 4-5 mm thick and 11 are between
6 and 8 mm (Table 3). Colors are predominately shades of
brown, ranging from 10YR4/6, a yellowish-brown, to
7.5YR5/6, a strong brown, by way of red and gray tints. A
few sherds are shades of yellow.
Sanibel Incised is the most common decorated pottery
(Figure 6) and most sherds of this ware were found in the
upper two levels of Test Pit 9. This ware is well compacted
and has small- to-medium size quartz tempering. Colors
range from 5YR3/2, reddish-brown, to 5YR3/1, very dark
gray. The rim sherds are incurving, with rounded lips and a
thickness of 8 mm (Figures 7 and 8). Level 1 of Test Pit 9
yielded 11 rim sherds weighing 237.3 gm and 31 body sherds
weighing 237.6 gm. Level 2 had two rim sherds weighing
37.2 gm and 20 body sherds weighing 212.7 gm. Level 3 had
a single rim sherd weighing 30.3 gm. Although no effort at
reconstruction has been made, all the Sanibel Incised sherds

Table 2. Inventory of ceramics recovered from Heineken
Rim Body
Unit Lvl. Type Sherds Sherds Total in gm
CS I FD(S?)T 2 14 16 28.1
2 STP 1 1 4.0
TPI S STP 3 3 .5
1 STP 3 21 24 129.0
2 STP 2 30 32 137.7
3 STP 3 2 5 25.0
TP2 1 STP 7 7 13.6
2 STP 28 28 72.4
TP3 2 STP 2 2 1.1
3 STP 7 7 1.9
ST3 2 STP 2 2 .5
TP4 S SI 1 1 5.2
1 STP 1 43 44 273.6
2 STP 3 3 42.3
3 STP 21 21 151.3
TP5 1 STP 1 28 29 57.4
2 STP 1 74 75 154.1
TP6 2 STP 9 9 76.7
3 STP 4 4 9.5
TP7 S STP 14 14 24.6
1 STP 19 19 55.9
TP8 1 STP 3 14 17 168.1
2 STP 10 10 34.1
3 STP 1 3 4 18.9
TP9 I SI 11 31 42 474.9
STP 43 43 102.2
2 SI 2 20 22 250.4
FD(S?)T 4 17 21 51.7
STP 103 103 736.5
3 SI 1 1 30.3
STP 2 2 10.4
TPI0 1 STP 8 8 34.8
Totals 35 584 619 3176.7

KEY: CS = column sample; TP

test pit; ST = shovel test; S =

surface; FD(S?)T = Fort Drum (Sanibel?) Ticked; STP = sand-
tempered plain; SI = Sanibel Incised.

1998 VVl S1144

Tmr fi mi A mTwiarnDnhl1 nwerT



Table 3. Technological characteristics ofsand-tempered-plain pottery, all test units combined.

sherds indicates a pot diameter
of 30-32 cm at the orifice. Surl
Thirty-seven sherds of Fort Variables N
Drum or Sanibel Ticked ware
were found; 21 from Level 2 of
Test Pit 7 and 16 from Level 1 Beveled
of the column sample (Figure Rounded
8). The two vertically oriented, Totals
ticked rim sherds from the first
level of the column sample
adjacent to Test Pit 1 are 8 mm Incurved
thick with diagonally incised Vertical
lines on the exteriors and tops Outced
of the lips. The rims are too
small to permit an estimation ota
of pot size. The 14 body sherds
found with them in the column
sample are the same thickness. 20 cm
All are a dark brown color,
7.5YR3/4, and the coring is
light. The paste is well com-
pacted and the quartz sand
temper is small with occa- 4-5 mm
sional larger grains. 6-8 mm
The four ticked rims and
corresponding 17 body 9-11 mm 3
sherds found in Level 2 of Test Totals 3
Pit 9 are 7 mm thick with
small quartz sand grains in-
4-5 mm
cluded in the paste as temper. m
Rim sherds are incurving with 6-8 mm
rounded lips. The paste is well Totals
compacted. Color is 5YR4/6, a
yellowish-red. The sherds are
too small to permit an estimation of the vessel sizes.

Artifacts of Industry

The distribution of tools and other industry-related
materials was thin and scattered, with only 26 found in five
test pits (Table 4). Test Pits 1 and 4 held the greatest number
at 11 and 5, respectively. Stratigraphically, most tools (16)
were recovered from Level 2, with 6 in Level 1 and 2 in
Level 3.
The shell tools include an adze or celt (Figure 9), three
clam anvil/choppers (Figure 10), two cutting-edge tools
made from lightning whelk columellae, a columella hammer
(Figure 11), a knife/scraper, and a dipper. The pointed end
of a bone projectile point was found in Test Pit 5's Level 1
(Figures 12). The point, apparently from a deer long bone,
was comparable in size to others found in the Naples area,
measuring 31.2 mm long and 6 mm wide. There were no cut
marks at the basal end, which appeared to have been broken
off cleanly. Enigmatic are a number of irregularly shaped

ace Level 1 Level 2 Level 3 Totals

Pct. N Pet. N Pet. N Pct. N Pct.

Rim/Lip Shape
5 71.4 2 28.6 7 100.0
2 22.2 3 33.3 4 44.4 9 100.0
7 43.8 5 31.3 4 25.0 16 100.0

Wall Orientation
2 33.3 4 66.7 6 100.0
2 28.6 5 71.4 7 100.0
2 66.7 1 33.3 3 100.0
4 25.0 8 50.0 4 25.0 16 100.0

Vessel Diameter
1 1 100.0

2 33.3 1 16.7 3 50.0 6 100.0
2 28.6 2 28.6 3 42.9 7 100.0

Body Thickness
15 68.2 7 31.8 22 100.0
93 86.9 1 .9 13 12.2 107 100.0
9.09 2 6.06 25 75.8 3 9.1 33 100.0
1.85 110 67.9 33 20.4 16 9.9 162 100.0

Rim Thickness
2 40.0 3 60.0 5 100.0
7 63.6 3 27.3 1 9.1 11 100.0
7 43.8 5 31.3 4 25.0 16 100.0

pieces of limestone found in Test Pits 1 and 2 whose edges
show apparent wear.
A small, circular, glass flake of unknown use and origin
was found just under the surface litter of Test Pit 7's Level 1,
as was an apparently unfired round musket or pistol ball. The
lead ball is spherical, 12 mm in diameter, and shows mold
ridges at its equator. There are no marks from firing, nor
deformation from impact, so presumably it had been mis-
placed or fallen from a pouch. It is of a size compatible with
arms of the Seminole wars period, according to Brent Weis-
man, who is familiar with such weapons (personal communi-
cation, 1997). Its white patina suggests a long exposure to
the soil. The ball was found near the roots of a fallen tree.
Test Pit 1 had a modern "Nitro Express" shotgun shell just
below the surface litter. The shotgun shell is 12-gauge and
most of the paper of which it was made has rotted away,
leaving only its brass base. The percussion cap is indented,
indicating that it had been fired. About 20 m south of the dig
area, a buckshot-size pellet was located by a metal detector
on the surface. It disappeared before it could be fully ana-
lyzed but examination at the site showed no deformation.

appear to have come from
one pot. Measurement of rim






Sanibel Incised

8CR231-34 Test pit 9 Level 2

8CR231-39 Test pit 9 Level 3

Ft. Drum (or Sanibel?) Ticked

Top of rim



Exterior Section

Top of rim

8CR231-13 Test pit 1

Level 1


Test pit 9 Level 2

Sand Tempered Plain


8CR 231-21
TP 7 Lev 1

TP 8 Lev 1

Figure 7. Pottery sherds and rim profiles shown approximately actual size.

230 EIRD
9r.-L% -- A -

TP1 Lev 1

TP1 Lev 2

TP1 Lev 3

8 991 VOL 51(4)




Table 4. Artifacts of industry.

Artifacts Material N Provenie

Projectile point Bone 1 TP 5, L 1
Adze/celt Quahog clam 1 TP 2, L 2
Adze/celt Lightning 1 TP 4, L 2
Anvils Quahog clam 2 TP 4, L 2
Columella cutting-edge Lightning 2 TP 4, L 2
tools whelk
Columella cutting-edge Horse conch 1 TP 6, L 3
Knife/scraper Quahog clam 1 TP 1, L 1
Dipper Lightning 1 TP 1, L2
Abraders Limestone 5 TP 1, L 2
Abraders Limestone 3 TP 2, L 2
Abrader Coral 1 TP 2, L 2
Abraders Limestone 3 TP 1, L 1
Glass flake Glass 1 TP 7, Sur
Musket or pistol ball Lead 1 TP 7, L 1
Shotgun shell Paper, brass 1 TP 1, Sur


Vertebrate species. Vertebrate remains reinforce the image
of Heineken Hammock having served as a camp for the
provisioning of a settlement elsewhere as well as having
provided on-site nourishment for its hunters (Table 5).
Forest-dwelling animals, notably deer and raccoon, are
present in the form of skeletal
fragments, as well as local
water-oriented fauna such as
alligator, amphibians, and tur-
tle. Provisions obviously
brought in from the coast in-
clude shark and several species
of salt-water fish.
Deer remains include five
astragali demonstrating an
MNI of three, but identifiable
bones of the principal meat-
bearing portions are absent.
Other mammals represented
are rodent (probably rabbit),
opossum, otter, and raccoon.
Reptiles present in the faunal
assemblage are turtle, snake,
and alligator. There also are
amphibians and fish, including
catfish, shark, jackfish, sheeps-
head, and drumfish. Fragments
of bird long bone were also
found. Figure 8. Sanibel Incised
Even when its size (nearly wall.

twice that of the other test pits) is considered, Test
Pit 1 produced the greatest weight of bone, with
nce more than 700 gm. Test Pit 4 was a close runner-up
at 602.5 gm, followed by Test Pits 8 and 9 at about
250 gm each, and Test Pit 5 with 208.4 gm. The
remaining pits had lesser amounts. Test Pit 1 was
unique in having greater amounts of bone in Levels
1 (333.4 gm) and 3 (250.3 gm) than in Level 2
,3 (141.1 gm). In the more eastern test pits all except
Test Pits 7 and 10 had the greatest amounts of bone
in Level 2.
Invertebrate species. Across the site, and across
the time represented by test pit levels, those who
camped on Heineken Hammock relied on shellfish
for sustenance, at least to some small degree (Table
6). Although deposits differed in species and num-
bers, each test pit and the single column sample
yielded a variety of shell, totaling 23 species of
marine invertebrates plus a freshwater snail.
ace Most common was the crown conch, Melongena
corona, of which 161 individuals could be identi-
fied, followed by the sunray venus, Macrocallista
face nimbosa (70), and lightning whelk, Busycon con-
trarium (48). The distribution of food shell was
fairly uniform across the site, although of the nine
species most commonly found, Test Pit 1 on the western edge
of the site produced the greatest number of individuals, 80.
Test Pit 6, somewhat to the east, had 63 and Test Pit 8, still
farther east, had 55. Vertically, most of the invertebrate
remains, 191 identified individuals, were found in the
combined second level of the test pits, with the combined
Level 3 having 86 and Level 1 having 96.

rim sherd showing distinctive incurved lip and sloping vessel

UHamErm HAnnuar REWD^II


3 '4 5 16 7 i 9 1 i o

Figure 9. Adze/celt fashioned from a lightning

Figure 10. Quahog clam shell showing evidence of

Radiocarbon Dating

Whelk shell from the lowest level of Test Pit 1 (20-30 cm
below the surface), where sand-tempered-plain ceramics had
been discovered, was submitted to Beta Analytic Radiocarbon
Dating Laboratory in Miami. The sample (Beta-61546) yield-
ed a conventional radiocarbon age of 3930 70 B. P., or
3935 75 when adjusted for local reservoir correction. The

3C/12C ratio was estimated at
0.0. Calibrated results were
2155 to 1745 B.C. at 2 sigma,
95% probability. The tested
shell had been found at a
depth of 23 cm below the sur-
face of Test Pit I and there
were other shell, burned rock,
and charcoal below it at depths
of 24 and 26 cm. Sand-
tempered-plain pottery had
been found at a level of 22 cm.
The charcoal was not dated
because it was in small frag-
ments and could have been left
by brush fires, to which this
part of Florida is subject.
The third (20-30 cm) level
of Test Pit 6 also yielded both
S 2 1 sand-tempered-plain pottery
and crown conch shell. The
shell (Beta-85496) yielded a
conventional radiocarbon age
of 4530 70 B. P., or 4535
vhelk whorL 75 when adjusted for local
reservoir correction. The
"'C/"'C ratio was measured at
l9.. o0.3. The cali-brated date was
2915 to 2560 B.C. at 2 sigma,
95% probability, nearly 800
years earlier than the first
So early a date called for
a rationalization. The only other
test pit with dateable material
at the lowest level was Test Pit
w-a 4, but a sample of lightning
whelk from it, after treatment,
proved unreliable for radiocar-
bon dating. As a result, an-
other crown conch shell sam-
ple from Test Pit 6, Level 3
was submitted for dating. This
sample (Beta-106983) yielded
a conventional radiocarbon
age of 4450 80 B. P., or
use as an anviL 4460 80 when adjusted for
local reservoir correction. The
'3C/'2C ratio was estimated at 0.0. The calibrated date was
2875 to 2455 B.C. at 2 sigma with 95% probability, statisti-
cally the equivalent of Beta-85496.
In both of the test pits where shell material used for dating
was collected, the shell, some of which was fragmented, was
scattered both vertically and horizontally within the 10 cm
of the lowest levels, as were the sand-tempered-plain pot-
sherds, with no distinct vertical, stratigraphic demarcation
between the shell and pottery. As noted above, the pottery in

1998 VOL 51(4)

IE ET' Al- HI-.-- HAmm 8C--R2--. --23---3

Test Pit 1 was found at a depth
of 22 cm and the dated shell was
collected from a depth of 23
cm, but in different parts of the
Test Pit not a reliable spatial


Previous Investigations

Late Archaic sites are known
in this coastal area, although
those described have generally
been located on higher ground.
This was the case of one site
explored by SWFAS on south
Marco Island, 8CR766, which
proved to be a specialized shell-
tool manufacturing station with
fiber-tempered potsherds (Lee et
al. 1997). In that general area
other sites of comparable age

i I I I I !

Figure 12. Fragment of a projectile point made from

have been explored, some also with fiber-tempered ware
(Cockrell 1970; Widmer 1974).
On Horr's Island, just south of Marco Island, Russo
uncovered a number of sites without ceramics which he
believes may date to the mid-Holocene (Russo 1996:191). In
an earlier exploration of the island, McMichael (1982:73)
discovered a single Late Archaic site with ceramics: eight
fiber-tempered sherds containing sand, which he attributed

F urv 1 Fae i
FI 1 I i i i ii I
Figure 11. Fragment of a lightning whelk columella.

to the natural content of the
clay. He did not consider the
ware to be of the transitional,
sand-and-fiber-tempered type.
Widmer (1988:71) discusses
an early site on nearby Marco
Island's high dune area,
8CR112, which had a rich and
varied ceramic content and a
date of 4965 + 100 B.P., al
though he holds out the possibil-
Sity that the date may represent
an early, non-ceramic, occupa-
Stion. It did contain two early
ceramic types, St. Johns Plain
(fiber-tempered) and Perico Is-
land Plain (crushed limestone-
tempered) ware.
A site radiocarbon dated at
about 2900 B.P., registered by
John G. Beriault in 1979 as
8CR199 and called the Fire-
a deer long bone. break site, contained well-made
fiber-tempered-plain sherds in
several levels. Beriault notes
that its midden contents extended to a meter below the
surface. It was located on a sand dune rising some 2 m above
sea level and about 4 km from the present Gulf beach,
immediately south of Vanderbilt Beach in north Naples. The
site was adjacent to another, the Bayhead Camp site,
8CR198, where cultural deposits extended to a depth of 66
cm, containing evidence that there had been a sea-level rise
since the site was occupied. No ceramics were found at this




THE 16N nm. AIrinnfl g 1995 yin.. 51(4

site, which was radiocarbon dated at ca. Table 5. Ver
2300 B.P. number of ind
Numerous occupation sites have been
found that contain ceramics assignable to Common Na
the Glades I Late Period. Most are near
sheltered waterways feeding into the Gulf,
although a number are located farther in- Snake
land, mostly alongside wet areas. They Turtle
include sites like Mulberry Midden (Lee et Alligator
al. 1993), located to the north of Heineken Gopher tortoi
Hammock and also showing evidence of Amphibian
having been used for provisioning a coastal
community and possibly as a "corridor" site ite-tai
on a route to an interior settlement. A simi- Raccoon
lar site is 8CR726 located 9.8 km southeast Opossum
of Heineken Hammock and containing Otter
sherds of a Fort Drum Incised variant. Rodent

Ceramics Catfish
All test pits at Heineken Hammock,
except Nos. 2, 5, 7, and 10, had sand- Drum
tempered-plain potsherds in the lowest of Jackfish
Sea trout
their three 10-cm levels, just above the Uidentifid
limestone bedrock. Except for a single bony fish
sherd of Sanibel Incised in Test Pit 9 that
presumably represents an intrusion, deco- Shark
rated sherds were found at higher levels
(Table 2). The early radiocarbon dates from Crab
shell collected from the same levels that Bird
contained the sand-tempered-plain pottery Unidentified
(cal. 2915-2560 B.C., cal. 2875-2455 vertebrates
B.C., and cal. 2155-1745 B.C.) exceed the Totals
age ascribed to the appearance of sand-
tempered-plain pottery in this area by Wid-
mer (1988:75) at about 500 B.C., during the
Glades I Early Period (Table 7). Widmer (1988:Table 5) and
Marquardt (1992:Table 2) show 2000 B.C. as the cut-off
point between the pre-ceramic Middle Archaic Period and
the Late Archaic Period, with its Orange Plain and Incised
and Perico Plain pottery. Although Marquardt based his
discussion on the Caloosahatchee region, he places its
southern boundary only 16 km north of the Heineken
Hammock site.
Discussing the Late Archaic or Pre-Glades Period (4000 to
2700 B.P.), Widmer (1988:72) regards a date of 2545 100
B.P. in Big Cypress Swamp, to the east of Heineken Ham-
mock, as providing a terminal date for a "transitional" sub-
period in which sand and fiber were used together as pottery
temper. He describes another date, 3110 + 80 B.P. from
shell at a different Big Cypress site, as being "about 150
years too early for the Transitional period but probably not
that out of line given the curatorial nature of marine shell
tools in an interior site."
Ehrenhard et al. (1978) note that a period of transition
from fiber-tempered potteryto sand-tempered pottery appears
to have occurred between 1000 B.C. and possibly 500 B.C..

tebrate species by weight, number of fragments, and minimum
lividuals (MNI).

me Scientific Name N Weight MNI
in gm

Serpentes 159 94.2 2
Testudines 1553 837.0 2
Alligator mississippiensis 6 17.6 2
ise Gopherus polyphemus 1 4.1 1
Amphibia sp. 40 30.6 2
Odocoileus virginiana 289 709.2 3

Procyon lotor 63 49.4 2
Didelpis virginiana 4 22.5 1
Lutra caanadensis 4 9.6 1
Rodentia 11 8.8 1
Bony Fish
Siluriformes 11 9.5 3
Archosargus 2 4.4 1
Sciaenops ocellatus 2 .5 1
Caranx sp. 1 1.0 1
Cynoscion sp. 1 .5 1

49 4.6 -
Cartilagenous Fish
Lamniformes 1 .2 1
Decapoda 1 1.3 1
Aves 60 18.9 2

1241 421.6 -
3499 2245.5 28

They list four sites within Big Cypress National Preserve as
possibly dating to this period, although they also note the
need for additional radiocarbon tests to date those early sites
A single Late Archaic site containing ceramics was found
at Horr's Island, 30 km to the south of Heineken Hammock,
by McMichael (1982:73). The sherds were fiber-tempered
but they did contain sand, considered to be a natural compo-
nent of the clay. Later, Russo and Cordell (Russo et al.
1991:Table 9.8) found one sand-tempered-plain sherd and
one fiber-tempered-plain sherd in the same level and a single
fiber-tempered-plain sherd at the level just beneath them in
a period they called "terminal late archaic/transitional."
On a dune in the Pelican Bay area north of the City of
Naples, the Firebreak site contained fiber-tempered sherds of
good quality from the second (10-20 cm) to seventh (60-70
cm) levels. The site was about 2 m above the Gulf of Mexico
water level. The records of John Beriault, who explored the
site, show that it was radiocarbon dated to ca. 2900 B.P.
Milanich (1994:301), referring to the Glades cultural
sequence, notes the Glades 1 Early Period, 500 B.C. A.D.

9Iqq Vo- l51(41

Tm njmnI A Im luil lDDI nrolf '


500, as marking the first appearance ofsand-tempered-plain
ware. At Fort Center, Sears (1982:192-193) saw the shift
from fiber to sand tempering during what he calls the first of
that site's four periods of occupation, running "from roughly
1000 B.C. to perhaps A.D. 250."
Still farther north, the Bullens (1976:9,13) found sand-
tempered sherds in the Hill Cottage Midden in the Sarasota
area in association with semi-fiber-tempered plain and
Orange Incised at a level radiocarbon dated at about 1400
B.C., which they ascribed to the Florida Transitional Period.
A very tenuous relationship with other early sand-tempered-
plain pottery is provided by Mikell (1997:86, 90) who notes
the association of such sherds with Norwood fiber-tempered
ceramics at the Reddick Bluff site in Walton County. His
Table I shows that Norwood Plain and sand-tempered-plain
sherds were both recovered from Levels 4 and 5 ofthat site's

Table 6. Invertebrate species by shell weight and minimum numb

Common Name Scientific Name ig
in gm

Block 1. He writes that although he does not have radiocar-
bon dates for that association "several dates for northwest
Florida Gulf Formational and Deptford sites place it between
ca. 2000 B.C. and 600 B.C." (Mikell 1997:90).
The upper levels of Heineken Hammock produced pottery
that can be ascribed to the Glades I Late (A.D. 500-700)
Period (Griffin 1988:Figure 6.3). Level 1 ofa column sample
adjacent to the west wall of Test Pit 1 held 16 sherds of Fort
Drum (or Sanibel?) Ticked ware. Twenty-one sherds of this
type also were discovered in the second level of Test Pit 9
(Table 8). Of the same general time period were sherds of
what appeared to be a single Sanibel Incised pot, found in the
first two levels of Test Pit 9. Adjacent Test Pit 4 had a single
Sanibel Incised body sherd on its surface. As to where the
smashed pots came from, there are numerous recorded sites
in the Naples coastal area dating to periods which would
accommodate the Sanibel
Incised and Fort Drum (or
er of individuals (MI). Sanibel?) Ticked pottery,
as well as sand-tempered
Pct. MNI Pt. plain.

Florida crown conch
Sunray venus
Lightning whelk
Eastern oyster
Quahog clam

True tulip
Florida fighting conch
Atlantic bay scallop
Cross-barred venus
Pygmy venus
Horse conch
Florida cerith
Transverse ark
Thick lucine
Banded tulip
Rough lima
Atlantic rupellaria
Top shell
Pointed venus

Pear whelk
Atlantic cockle
Slipper shell
False angel wing
Channeled whelk
Unidentified gastropods

Cannibal snail
Florida apple snail

Melongena corona
Macrocallista nimbosa
Busycon contrarium
Crassostrea virginica
Fasciolaria tulipa
Strombus alatus
Argopecten irridians
Chione cancellata
Chione sp.
Pleuroploca gigantea
Anadara transversa
Lucina pictinata
Fasciolaria hunteria
Lima scabra
Rupellaria typical
Tugula sp.
Busycon spiratum
Dinacardium robustum
Crepidula plan
Petricola pholadiformis
Busycon canaliculatum




Land and Fresh-water Snails
Polygyra uvulifera 114.6
Planorbidae 24.1
Euglandia rose 24.4
Pomacea paludosa 9.05









In evaluating its dates,
the mechanics of this site
must be kept in mind. This
is a tightly-compressed
site; its limestone bedrock
dates to the Pleistocene
and in all the time that has
elapsed since it was ex-
posed to the air only some
30 cm of sediments have
accumulated. From all
evidence, prehistoric occu-
pation of the site was light
and sporadic. The sparse
use of shellfish failed to
give it the volume which
in some sites makes it eas-
ier to differentiate the pe-
riods of occupation. Also,
in such an ancient and
shallow deposit, the possi-
ble effects of root action,
including the uprooting of
trees, must be kept in
mind, although with one
previously noted exception
(Test Pit 7) a quite com-
mon pattern of soil texture
and coloring prevailed.


Artifacts found at the
site were compatible with
its apparent history ofhav-
ing served as an occasion-


1F 1[ H o' 1



I ...O i1 V 5

Table 7. Heineken Hammock radiocarbon dates in perspective.
Date" Event and/or Provenience Reference

2500 B.P. Start of Glades Tradition Widmer 1988:73b
2900 B.P. Fiber-tempered pottery in Naples area. Beriault 1979
2950 B.P. Start of transition to use of sand-temp- Widmer 1988:Table 5
ered plain
3000 B.P. Suggested earliest date for Fort Center Sears 1982:192
occupation; fiber-tempered sherds
3005 B.P. 8CR697; sand-tempered plain Lee et al. 1993:46
3110 B.P. Big Cypress shell early for Transi- Widmer 1988:72
tional Period
3400 B.P. Sand-tempered plain with semi-fiber- Bullen and Bullen
tempered and Orange Incised in 1976:9, 13
Sarasota area
3800 B.P. End ofpre-Glades I, start ofpre- Widmer 1988:Table 5
Glades II with Orange Plain, St. Johns
Plain, Perico Plain
4000 B.P. Fiber-tempered plain ceramics on Lee et al. 1997:11
8CR766, Marco Island
4050 + 125 B.P. Earliest Orange ceramics at Palmer Bullen and Bullen
site 1976:Table 2
4155-3745 B.P. 8CR231, Heineken Hammock This paper
4500 B.P. Start ofpre-Glades I Late; Orange Widmer 1988:Table 5
4600-5200 B.P. 30 dates for occupation of Horr's Russo 1996:194
4855-4555 B.P. 8CR231, Heineken Hammock This paper
4870-4650 B.P. 8CR231, Heineken Hammock This paper
5000 B.P. Start ofpre-Glades I Early; ceramic Widmer 1988:207
occupation of Marco Island

shell tools large enough for conventional
radiocarbon testing were spared, mostly be-
cause of the possibility that curation by the
site's inhabitants may have compromised their
age but also because of their value in educa-
tion. (The SWFAS laboratory has a modest
display window and routinely provides materi-
als to the Collier County Museum.) It should
be noted that the site did not contain obvious
debitage from the manufacture of tools from
shell, as is commonly found in more perma-
nent, coastal habitation sites, indicating that
the tools had been manufactured before being
brought to the site.

Vertebrate Use

Ties to coastal waters are demonstrated by
the presence of marine shellfish remains and
are further evidenced by at least token amounts
of salt-water fish remains that were found in
all but two of the test pits. Although bone
attests to the presence of at least three deer,
none of the major parts of the animals were
present, lending weight to the image of the site
as a provisioning point for a settlement located
elsewhere. Except for Test Pit 10, all excava-
tion units contained deer bone. Foraging for
immediate use while camping at the site is
indicated by the near-ubiquitous presence of
turtle and snake bone, as well as the remains of
modest amounts of the bones of smaller mam-
mals, amphibians, and birds.

ally occupied base for provisioning. One, a broken bone
projectile point from Level 1 of Test Pit 5, is suggestive of
prehistoric hunting, and two others carried the hunt into
more recent times: a ball for a muzzle-loaded firearm from
Test Pit 7 and the spent shell of a modern shotgun from Test
Pit 1. The tentative Seminole War-period connection invoked
by the musket or pistol ball was reinforced by the possibility
that the glass flake, found in the same area, could have
resulted from tool making or repair. In a survey of Pine
Island Ridge in Broward County, a glass fragment showing
signs of edge flaking was found in a Seminole context
(Robert S. Carr, personal communication 1998).
The site's shell tools attest to its having been used for
domestic purposes, although the thin deposits of ash and
charcoal indicate only intermittent stays. There are imple-
ments for wood working (e.g., adze/celts, columella cutting-
edged tools) and for opening bones for marrow (e.g., anvils
and a columella hammer), a dipper for eating or drinking,
and a knife/scraper which could have been used in hide-
working or food preparation. The adze/celt made of a
lightning whelk whorl had a beautifully sharpened edge
which could have been employed in a variety of uses. Those

Invertebrate Use

Despite evidence that the site was intermittently occupied
over an apparent spread of more than 3000 years (from the
earliest radiocarbon date of 2815 B.C. to the end of the
Glades I Late period at ca. A.D. 750 indicated by the pres-
ence of Fort Drum [or Sanibel?] Ticked and Sanibel Incised
pottery), the relative utilization of invertebrate species
appears quite consistent from test pit to test pit and from
level to level, although it must be recognized that their
numbers were so small that mathematical comparison
becomes chancy. It is striking that the site's invertebrate
inventory included only 22 Eastern oysters (Crassostrea
virginica), although they are usually the most heavily
represented shellfish in midden deposits in this area (Table
6). Another "corridor" site only 16 km distant, Mulberry
Midden (8CR697) whose earliest occupancy was at ca. 985
B.C., had 206 oysters versus 29 quahog clam (Mercenaria
campechiensis), 44 lightning whelk (Busycon contrarium),
and 17 sunray venus (Macrocallista nimbosa) (Lee et al.
The evident preference for crown conch (161 MNI) at

1998 VOL. 51(4)



EE HENEE AImc 8C23 23

Heineken Hammock echoes the content of a Marco Island
site, Satin Leaf(8CR766) some 27.4 km distant and of a time
period comparable to the 2155-1745 B.C. age of Heineken
Hammock's Test Pit 1. There workers at a shell-tool work-
shop consumed 1171 individual crown conch versus 357
oysters (Lee et al. 1997). Unlike the Satin Leaf workshop
conch, whose whorls were reduced to small fragments, the
Heineken Hammock conch were relatively intact.
Whether the meager use of oyster can be profitably brought
to bear on speculation as to the degree that rising Gulf waters
had created a productive estuarine system at the time periods
involved is questionable, since its shell was present in all
levels, although in greater numbers at the 10-20 cm depth.
The presence of established estuarine sea-grass meadows is
suggested by the use of Atlantic bay scallop (Argopecten
irridians), lightning whelk, and quahog clam (cf. Walker
1992:270-271), although Florida fighting conch (Strombus
alatus) and cross-barred venus (Chione cancellata) were not
present in the lowest levels. While the crown conch was
present throughout all levels, in Levels I and 3 80% was of
a type usually found in estuarine sea-grass meadows. In
Level 2 there was an approximately equal amount of a form
which commonly frequents oyster beds. Florida fighting
conch was present only in Level 2 of Test Pit 6, which also
contained sand-tempered-plain pottery in both Levels 2 and


Noting that the high dunes of Useppa and Horr's Islands
have protected evidence of late Archaic occupation of coastal
Southwest Florida from rising Gulf waters, Milanich
(1994:101) forecast "More late Archaic coastal sites probably
exist in as yet unsurveyed dune settings or in locales back
from the coast." The discovery at Heineken Hammock of
marine shell dating to 4500 B.P. in a temporary campsite
well back from the gulf shoreline suggests the presence of a
stable community within a practical hiking or canoeing
distance at that time. The postulated existence of a populated
shore area conforms to a pattern discerned by Widmer
(1988:69) of early coastal sites during that general time
period, which he believed to be non-ceramic. That also is a
reasonable assumption for later occupations at Heineken
Hammock, as demonstrated by the presence in the site's
upper levels of Glades I Late-period Sanibel Incised and Fort
Drum (or Sanibel?) Ticked sherds, as well as sand-tempered
plain, which also was found in the earliest and lowest levels
in company with bones of deer and other animals suggestive
of provisioning a more permanent settlement.
Exploration of Horr's Island, some 27 straight-line
kilometers to the south, has revealed year-around occupation
coincident with the earliest date from Heineken Hammock.
Thirty dates at a village site there fall between 4600 and
5200 B.P. (Russo 1996:194). Russo et al. (1991:373-374)
note that "...nonessential subsistence items such as freshwa-
ter fauna, deer, and plants [the remains of which were found
in village middens] may have been procured from interior

sites. The small size and paucity of known Archaic sites east
and interior from the southwest Florida coast suggests that if
such moves were undertaken, they were logistical, short term
and probably more common north of the Ten Thousand
Islands area."
Land elevations and existing swamps and creeks in the
area south and west of the Heineken Hammock site suggest
a drainage pattern favorable to canoe travel, at least during
periods of abundant rainfall, to both the greater Naples area
and to waterways leading to Marco and Horr's islands
(Figure 15). Leighty et al. (1954) classify the soils to the
south and southwest of Heineken Hammock as being subject
to slow or ponded surface runoff. They describe the area
thus: "Following heavy rains and during the rainy season in
summer and early fall, the cypress strands, sloughs, wet
prairies and islands within the Big Cypress Swamp may be
covered by a few inches to several feet of water."
There are known mounds in the Naples area, but the bases
of any area occupied 4500 years ago would be well below the
present sea level. Widmer (1988:Table 14) shows a sea level
at 4500 B.P. of 4 m below present and a rate of sea-level rise
per 100 years of.213 m. He (personal communication, 1998)
believes it conceivable that the early Heineken Hammock
occupation date could correspond to a higher sea stand and
occupation similar to that at Horr's Island.
On a broader regional scale, the middle-late Holocene
transition, within which the first dates for Heineken Ham-
mock fall, is portrayed by Gunn (1997:145, Figure 5) as a
period during which dramatic changes of sea level occurred,
with stands ranging from near present to almost 2 m below
current levels. For most regions of the southeastern United
States, he regards it as "an unusually moist and populous
period." Other investigators see a somewhat higher stand at
4500 B.P. For example, a curve drawn by Stapor et al.
(1991) shows sea level at 1 m below present while Tanner
(1991) pictures a still higher level for the period at 40 cm
below present.
The earliest radiocarbon dates from Heineken Hammock,
ranging from 2915 to 1745 B.C., are not far outside the
limits commonly ascribed to the beginning of the Late
Archaic Period (ca. 2000-1200 B.C.), and marked by the
introduction of ceramics (Marquardt 1992; Widmer 1988).
Pottery found in the Florida west coast region and dated to
this introductory ceramic period commonly is fiber-tempered
and identified as "Orange Plain" or "Orange Incised," and
was followed later by semi-fiber-tempered and a change to
sand-tempered-plain during a "transition" stage.
This full pottery sequence was not found in the Heineken
Hammock deposits. At the earliest levels of several test pits,
a few centimeters above the limestone formation on which
the site stood, sherds of sand-tempered-plain ware were
discovered, but there were no sherds of the fiber-tempered or
semi-fiber-tempered ceramics. Higher deposits held Fort
Drum (or Sanibel?) Ticked and a good part of a Sanibel
Incised bowl of the Glades I Late Period (A.D. 500-750).
From the evidence available at Heineken Hammock, one
can conclude that:



I L'-r I

8 991 Voo 51 4)

2,38 lea V01- 5aor

1) The site was occupied at a very early date, probably by
people who did not have pottery and who came from estab-
lished communities on the coast. In this connection it has
been established that Horr's Island had a year-round popula-
tion during the same time that Heineken Hammock was
occupied and the people who lived there apparently did
import inland provisions. The absence of pottery at the
Bayhead Camp site north of Naples, although its dates are
more recent than Heineken's, suggests the presence of a pre-
ceramic occupation within reach of the Heineken Hammock
camp site.
2) The first potsherds left at Heineken Hammock were
sand-tempered plain, which is generally regarded as originat-
ing during the terminal Archaic or Florida Transitional
Period, ca. 1000 B.C. The close association of marine shell
dated well before that period with sand-tempered-plain
pottery can be explained by the thinness of the site's deposits
and possible resultant commingling of artifacts from differ-
ent occupations.
3) Later visitors using Fort Drum (or Sanibel?) Ticked and
Sanibel Incised pottery camped in the area.
Shellfish consumed at the site attest to the presence of
established estuarine sea-grass meadows in this part of the
Gulf coast and it can be noted with interest that its occupiers,
like workers at a shell-tool workshop of comparable age on
south Marco Island (Lee et al. 1997), consumed a preponder-
ance of Florida crown conch, Melongena corona, a denizen
of such sea-grass habitats, and used only a relatively small
number of oysters, a species dominant in later sites of the
area. Use of canoes by those who frequented the area is
strongly implied by the site's location, a relatively high place
in what has been historically a swampy area, as well as the
geographic suggestions of waterways leading to Naples and
Rookery bays, both providing sheltered waters to coastal or
island points.
Heineken Hammock's deposits are thin and vulnerable to
perturbation, as noted before; however, the near-juxtaposi-
tion in one portion of the site of evidence of a 4500-year-old
occupation and the presence of people using sand-tempered-
plain pottery, which did not come into general use in the area
until nearly 2000 years later, provides fertile ground for
speculation. The most parsimonious explanation is that the
grounds simply were not occupied in the period between the
early deposition of sea shell and their use by people using
sand-tempered-plain ware. That long hiatus could be
ascribed to sea-level changes that forced the removal of
parent coastal villages to more distant areas now drowned;
climatic changes which could have created conditions
making canoe or other travel in that area less feasible; or
social factors which could have inhibited use of the site.
Widmer (personal communication, 1998) has speculated that
there could have been episodes of rise and fall of the sea
during the period in question, which could have interrupted
occupation of lower coastal areas. There is scope here for
future research.
Looking back in time from the Heineken Hammock
occupation, one notes the marked contrast between the

maritime culture presented by the contents of the Heineken
Hammock site and those of the Bay West site some 16 km
north, whose radiocarbon dates range from ca. 7550 to ca.
5885 B.P. Artifacts from this early site portray an economy
completely oriented to the uplands and show none of the
dependency on the sea exhibited by later settlers (Beriault et
al. 1981). Conceivably, this shift in life styles can be attrib-
uted to a change in climate from the arid conditions of the
early Holocene to those more like today's. Widmer
(1988:165) notes that by 5500 B.P. the sea level, and the
commensurate ground-water table, had risen to a point where
there was discharge from springs and surface storage of
water which gave rise to more rainfall.


Since John Beriault and Gail and Charlie Strader discovered the green bottles
and potsherds that ultimately became 8CR23 I, many people have contributed
to deciphering its clues. Some spent a few hours over a trowel, others devoted
weeks to identifying, measuring, and recording the objects that were found.
In addition to the co-authors ofthis report, who spent many hours analyzing
the site's data, major contributions of lab time were made by Ella May Ablahat,
John Dante, Lynn Foster, Janet A. Gooding, Lynn Lee, Donald G. Mielke, and
Annette L. Snapp. Joe Long provided plane table locations. In addition to
working in the lab and the site, Lois M. Polewka and her husband Stan paid for
one of the radiocarbon dates. Laboratory facilities were provided by the Collier
County Museum, for which lab personnel work as volunteers; its director, Ron
Jamro, has been helpful in many ways. The observations of George Luer,
Michael Russo, and Randolph Widmer have added greatly to the accuracy and
depth of this report.
Diggers on the first exploration were Beriault, Walt and Mary Buschelman,
Tom and Joan Clark, Guy Fischer, Art Lee, Joe and Freda Long, Leo Ruble,
James Ward, and Mary Ruth Winchell. The crew for the second dig included
Gloria Andrews, Linda Ballou, Jean Belknap, Beriault, the Buschelmans,
Patrick Devins, William Gauger, Robert Frenzan, Gooding, Mitchell Hope,
Wayne and Shirley House, the Lees, Lois M. Polewka, P. W. Quails, Ray
Seguin, Snapp, Charlie Strader, Don Taggert, and Jack Thompson.
Special thanks are due the Berkshire Village Association, Charley Farley,
President, and Newell Property Management Corporation, William A. Newell,
Manager, for granting permission to explore the site, and for the protection it
proudly affords its "Indian camp ground." Diggers remember fondly the
neighborhood lady who served them an elegant afternoon tea.
Personnel of the Beta Analytic Radiocarbon Dating Laboratory were
unfailingly helpful.
SWFAS has suffered the loss of two of its members who were active in the
Heineken Hammock exploration. Mitchell E. Hope ofNokomis, an enthusiastic
and meticulous field worker whose archaeological and paleontological work
had earned him wide-spread recognition, died in January 1996 following a heart
operation. Joseph True Long of Naples, a professional surveyor who played a
major role in gaining SWFAS its present stature, succumbed to cancer in
November 1996.

References Cited

Beriault, John G.
1979 Bayhead Camp, 8CR198, 199. Florida Archaeological Survey Site
Form. On file, Florida Division of Historical Resources, Tallahassee.
Beriault, John, Robert Carr, Jerry Stipp, Richard Johnson, and Jack Meeder
1981 The Archaeological Salvage of the Bay West Site, Collier County,
Florida. The Florida Anthropologist 34:39-58
Bullen, Ripley P., and Adelaide K. Bullen
1976 The Palmer Site. Florida Anthropological Society Publications
Number 8, Gainesville.
Cockrell, Wilburn A.
1970 Glades I and Pre-Glades Settlement and Subsistence Patterns on
Marco Island, Collier County, Florida. M.A. thesis, Department of
Anthropology, Florida State University, Tallahassee.
Cordell, Ann S.
1992 Technological Investigation of Pottery Variability in Southwest

T F __ A --. -



Florida. In Culture and Environment in the Domain of the Calusa,
edited by William H. Marquardt, pp. 105-189. Monograph Number
1, Institute ofArchaeology and Paleoenvironmental Studies, University
of Florida, Gainesville.
Ehrenhard, J. E., R. S. Carr, and R. C. Taylor
1978 The Archaeological Survey of the Big Cypress National Preserve,
Phase I. Southeastern Archaeological Center, National Park Service,
U. S. Department of the Interior, Tallahassee.
Goggin, John M.
1939 A Ceramic Sequence in South Florida. The New Mexico Anthropolo-
gist 3:35-40.
Griffin, John W.
1988 The Archaeology of Everglades National Park: A Synthesis.
Southeast Archaeological Center, National Park Service, U. S.
Department of the Interior, Tallahassee.
Gunn, Joel D.
1997 A Framework for the Middle-Late Holocene Transition, Astronomical
and GeophysicalConditions. Southeastern Archaeology 16:134-151.
Henderson, John A.
1876 Survey of Collier County, Florida. Townships 49,50, and 51 South,
Ranges 25 and 26 East. Surveyor's General's Office, Tallahassee.
Lee, Arthur R., John G. Beriault, Walter Buschelman, and Jean Belknap
1993 A Small Site Mulberry Midden, 8CR697 Contributes to Knowl-
edge of Transitional Period. The Florida Anthropologist 46:43-52.
Lee, Arthur R., John G. Beriault, Jean Belknap, Walter M. Buschelman,
Annette L. Snapp, and John W. Thompson
1997 Salvage Excavation of an Archaic-Period Special-Purpose Site in
Collier County. The Florida Anthropologist 50:11-24.
Leighty, Ralph G., M. B. Marco, G. A. Swenson, R. E. Caldwell, J. R.
Henderson, OlafC. Nelson, and J. R. Henderson
1954 Soil survey of Collier County, Florida. United States Department of
Agriculture in cooperation with the Florida Agricultural Experiment
Station, U. S. Government Printing Office, Washington, D. C.
McMichael, Alan E.
1982 A Cultural Resource Assessment of Horr's Island, Collier County,
Florida. Florida State Museum, Department of Anthropology,
Miscellaneous Project Reports Series Number 15, Gainesville.
Marquardt, William H.
1992 Recent Archaeological and Paleoenvironmental Investigations in
Southwest Florida. In Culture and Environment in the Domain ofthe
Calusa, edited by William H. Marquardt, pp. 9-57. Monograph
Number I, Institute of Archaeology and Paleoenvironmental Studies,
University of Florida, Gainesville.
Mikell, Gregory A.
1997 A Case of Direct Association Between Fiber-Tempered Pottery, Late
Archaic Stemmed Points, and Santa Fe Points at the Reddick Bluff
Site, Walton County. The Florida Anthropologist 50:83-93.
Milanich, Jerald T.
1994 Archaeology ofPrecolumbian Florida. University Press of Florida,
Missimer, Thomas M.
1984 TheGeology ofSouth Florida: A Summary. In Environments ofSouth
Florida: Present and Past II, edited by Patrick J. Gleason, pp.385-
404. Miami Geological Society. Coral Gables.
Munsell Color
1975 Munsell Soil Color Charts. Munsell Color. Baltimore.
Russo, Michael
1996 SoutheasternMid-HoloceneCoastal Settlement. InArchaeologyofthe
Mid-Holocene Southeast, edited by Kenneth E. Sassaman and David
G. Anderson, pp. 177-199. University Press of Florida, Gainesville.
Russo, Michael, Ann S. Cordell, Lee Newsom, and Sylvia Scudder
1991 Final Report on Horr's Island: The Archaeology of Archaic and
GladesSetlementandSubsistence Patterns. Partll. Florida Museum
of Natural History, Department of Anthropology, Gainesville.
Sears, William H.
1982 Fort Center: An Archaeological Site in the Lake Okeechobee Basin.
University Presses of Florida, Gainesville.
Stapor, Frank W., Jr., Thomas D. Matthews, and Fonda E. Lindfors-Kearns
1991 Barrier-Island Progradation and Holocene Sea-Level History in
Southwest Florida. Journal of Coastal Research 7:815-838.
Tanner, William F.
1991 The "Gulf of Mexico" Late Holocene Sea Level Curve and River
Delta History. Gulf Coast Association of Geological Societies.
Transactions 41:583-589.

Walker, Karen Jo
1992 The Zoology of Charlotte Harbor's Prehistoric Maritime Adaptation:
Spatial and Temporal Perspectives. In Culture and Environment in
the Domain ofthe Calusa, edited by William H. Marquardt, pp. 265-
366. Monograph Number 1, Institute of Archaeology and Paleo-
environmental Studies, University of Florida, Gainesville.
Widmer Randolph J.
1974 A Survey and Assessment of Archaeological Resources on Marco
Island, Collier County, Florida. Miscellaneous Projects Report Series
19, Florida Division ofArchives, History, and Records Management,
1988 The Evolution of the Calusa: A Non-Agricultural Chiefdom on the
Southwest Florida Coast The University of Alabama Press, Tuska-
Yamataki, H.
1990 Collier County Interim Soil Survey and Narrative Description.
Collier Soil and Water Conservation District, Naples.

I A01

HUEwakrEE HAunrw1r lE1T711


Donate a minimum of $100 to The
Florida Anthropologist Endowment and
receive a signed, limited edition print
by renowned artist, Dean Quigley,

Choose from one of the following
prints in Quigley's Lost Florida series.

HOLOPAW depicts a middle Archaic campsite in north
Florida. Hunters are returning fro %C cessful bear hunt; a
woman grinds tubers to r; a small child is in awe
of a giant alligator; d an converses with traders from a
distant place. g e foreground, a flintknapper works on
finishing his finely made projetlcpoint.

THE HUNTERS jJlitet a Paleoindian hunting party in
combat witt dl enle mastodon.

THE KILL depicts two Archaic period hunters preparing to
Spear a large alligator in the depths of a southern cypress

The Kill CLOSING IN depicts a group of Paleoindian hunters closing
in on a mastodon and her calf.

CALUSA depicts the large village site at Pineland, Florida ca.
A.D. 900-1100. Featured on the cover of Florida's First

TATAMAHO depicts an Indian gigging a garfish while his son
playfully loads the catch in a basket. Nearby, a panther waits
for a free meal and a raccoon eagerly awaits leftovers.

THE RETURN depicts two tattooed Akans returning from a
day of fishing. Their canoe i ~ ith fish and shellfish.
In the distance, smrJl~ls tom a campsite hidden amidst
the mangrove ~hated by cedar trees.

BOSKITA depicts the Green Corn Ceremony, one of the most
important rituals of many southeastern Indian cultures. Boskita

SOLITUDE is based on t l s of early European Also available: Florida Indian Artifacts, a 16 x 20 inch poster
explorers and de ps it hunter returning to his canoe showing the wide range of artifacts made and used by the
loaded with fix % native peoples of Florida.

Mail yourdonation to: The Florida Andhropologist. P.O. Box 2818, Riverview, FL 33568. Only a united number of prins are available. Please specify first and
second choices.


Status of Florida's Isolated Finds Program

2809 N. W 161 Court, Gainesville, FL 32609

A little over a year ago, I published a short article in The
Florida Anthropologist regarding Florida's Isolated Finds
Program (Knight 1997). Here I provide an update on the
status of that program.
The Isolated Finds Program was developed to allow river
divers to search for, collect, and report artifacts from
Florida's rivers. However, as antiquity laws were strength-
ened from the 1960s through the 1980s to provide greater
legal protection for artifacts present on state-owned land,
collection of isolated artifacts in rivers became an illegal
activity with possible serious punishment. Although there
was no effort to systematically enforce these laws in the
rivers, the anticipation of enforcement and oft-repeated
anecdotes about enforcement (with no substantiation) led to
a growing breakdown in communication between many river
divers and the state's Division of Historical Resources. In
recognition of this eroding situation, and of the minimal
archaeological significance of most river artifacts present in
shifting bed sediments, the Division's Bureau of Archaeolog-
ical Research instituted the Isolated Finds Program in June,
1997. The program allows river divers the right to keep
artifacts they find in exchange for a report that provides
details of the find, including a location map, a photocopy of
the artifact, and any additional information that might be
relevant. Following submission of a report to the Bureau, a
response letter is sent to the collector that indicates the
beginning of a 90-day period of review in which the bureau
may elect to request more information about the artifact, or
even request the artifact for further study.
During the first year of the Isolated Finds Program a total
of 141 reports have been received by the Bureau form a total
of 29 separate divers. These reports describe artifacts
collected from 17 rivers, including, in order of the number of
separate reports: the Santa Fe, St. Johns, Aucilla, Wacissa,
Rainbow, Suwannee, Oklawaha, Steinhatchee, St. Marks,
Chipola, Waccasassa, Withlacoochee, Lake George, Peace,
Itchetucknee, and Ochlockonee. A total of 1014 prehistoric
artifacts have been reported, including 714 stone tools, 203
ceramic pieces, and 187 worked bone tools. Of the stone
tools that were "diagnostic," the relative occurrence was:
Archaic Stemmed 19%, Bolen 18%, general Paleoindian
- 12%, scrapers 10%, Pinellas 5%, Newnan 5%,
Suwannee 5%, Hernando 4%, and Simpson 2%. In
addition to these projectile-point types, a variety of unique

Florida artifacts have been reported including bola stones,
bone fishhooks, barbed and drilled bone points, Waller
knives, and a carved alligator-jaw knife. None of these
artifacts have been requested by the Bureau, and all are now
the legal property of the collectors who found and reported
Florida's Isolated Finds Program is an excellent demon-
stration project for collectors throughout the country. As
most archaeologist are acutely aware, artifacts are eroding
out of sites and being permanently lost in lakes, reservoirs,
and along rivers throughout the United States. Many of these
artifacts are on state or federal land and their collection is
generally forbidden by law. The mind set that it is better for
these artifacts to be lost for perpetuity than to allow their
collection and curation by interested avocational archaeolo-
gists is naive and easily exposed as a greater evil. Artifact
collectors need to be allowed to collect artifacts in a responsi-
ble fashion on public land. The Isolated Finds Program is
one effective formula for encouraging responsible collecting.
Collecting artifacts in Florida's rivers is a privilege that
entails the obligation to share the information with the
public. With this privilege comes the responsibility to avoid
collecting from intact archaeological sites and to document
and report all significant finds and sites. Florida's divers
need to fully embrace this program and report all covered
finds to the Bureau of Archaeological Research (1-850-487-
2299). We need to help other state and federal resource
protection agencies take similar productive steps to benefit
from the volunteer efforts of amateur archaeologists to better
document the prehistoric past.

Reference Cited

Knight, Robert L.
1997 Professional/Amateur Cooperation in Florida: The Isolated Finds
Program is a Good Start. The Florida Anthropologist 50:31-37.


8 991 Vol 51 4)


In addition to the members of our Editorial Review Board, several persons have
generously reviewed manuscripts submitted to The Florida Anthropologist during the past
year. It is with pleasure that we acknowledge the important service that these individuals
have provided to the Society.

Susan Anton
George Ballo
Robert Edic
Richard W. Estabrook
Kathleen Hoffinan
George Luer
William H. Marquardt
Barbara A. Purdy
Douglas T. Peck

Donna Ruhl
Michael Russo
Kenneth E. Sassaman
Marion F. Smith
Anne V. Stokes
Louis D. Tesar
Karen Jo Walker
Ryan Wheeler
Curtis W. Wienker

Join the 1999 Allendale Paleoindian

Expedition in the beautiful Savannah River Valley of

Allendale County, South Carolina, May 4 29, 1999.

Calling for volunteers from the public, no experience necessary, to sign up for a week or
more to help excavate ancient Early Man sites associated with prehistoric chert quarries.
Sites include possible pre-Clovis, Clovis, Dalton, and Early Archaic occupations.
Volunteers learn excavation techniques and artifact identification. The expedition also
provides a good excavation experience for undergraduate and graduate students. The
cost is $366.00 per week ($300.00 is tax deductible).

o free camping with hot showers
o lunch and evening meals provided
o evening lectures and programs
o Paleoindian book and T-shirt
o motels within 30 minutes

Contact Dr. Al Goodyear, SC Institute of
Archaeology, USC, 1321 Pendleton St.,
Columbia, SC 29208
PH: 803-777-8170
L Pji~ aia fl ii ^ 11 :7FRE1 R F ,FMRFI1

A-. 37-1m mnro

To F ~ Amr*lmM~


Archaeology of Northern Florida, A.D. 200-900: The
McKeithen Weeden Island Culture. Jerald T. Milanich, Ann
S. Cordell, Vernon J. Knight, Jr., Timothy A. Kohler,
Brenda J. Sigler-Lavelle, University Press of Florida,
Gainesville, 1997. xviii + 222 pp., figures, tables, refer-
ences, index, $29.95 (paper).

DepartmentofSociology andAnthropology, State University
of West Georgia, Carrollton, Georgia 30118

In 1975 Jerry Milanich and Bill Sears piled into my brown
Pinto and we drove from Gainesville to the McKeithen site
in Columbia County, Florida. With Lex McKeithen as a
guide we toured this multi-mound Weeden Island site. The
spectacular effigy pots that McKeithen had shown to
Milanich demonstrated that this site should be contempora-
neous with the Kolomoki site in southwest Georgia. Soon
after this visit Milanich, with support from the Wentworth
Foundation and later the National Science Foundation, began
work at the first Weeden Island civic center to be dug since
Sears' work at Kolomoki in the late 1940s and early 1950s.
The results of this work were presented at numerous meet-
ings, in a book published by Academic Press, and in a set of
slides from Pictures of Record. This book is a reprint of the
1984 Academic Press volume with a new Preface by
McKeithen is an interesting site because, like Kolomoki,
it should not be there. Conventional models tell us that
multi-mound civic centers with organized villages and flat-
topped, truncated mounds belong in the Mississippian
Period, not the Middle Woodland. Complex multi-mound
sites are generally thought to have been the center of chiefly
activities, a form of political organization that is seldom
thought to have been present in pre-Mississippian cultures.
Milanich and his co-authors present an extremely readable
narrative of their work. Lacking most of the jargon of "The
New Archaeology" that was so much in vogue at the time
this work was done, this book presents a tour of the total
project. Chapter 1, entitled "Behind the Scenes," is a
wonderful introduction to how the project began and devel-
oped. It should be read and enjoyed.
The concepts) of Weeden Island and a discussion of
Weeden Island settlement patterns in North Florida are
discussed in Chapters 2 and 3. Much of these discussions
come from Milanich's interpretations of the very uneven
survey coverage of the Florida/Georgia/Alabama area and
Brenda Sigler-Lavelle's year-long survey of the area sur-
rounding the McKeithen site. Sigler-Lavelle identified 11
Weeden Island mound sites in Columbia County, seven of

which were tested to identify the village areas associated with
them. It was found that six of these had middens located to
the east. The authors consider and reject the idea that there
was a settlement hierarchy and suggest a pattern of fissioning
similar to that proposed for Early Neolithic communities in
the Middle East, where there are civic centers and what
appear to be outlying communities.
The discussions of the midden and mound excavations in
Chapters 4 and 5 are clearly presented and provide some of
the most detailed and useful descriptions of ways to approach
these kinds of projects. Especially interesting and useful is
Tim Kohler's use of the concept of "Elite Pottery" and his
ability to define elite areas in the midden. It is unfortunate
that although portions of what appear to be two structures
were excavated, we are told very little about McKeithen
houses. The excavations of the three mounds are discussed in
the chapter entitled "Charnel Knowledge." Aside from a
wealth of information on mound construction and the
wonderful Weeden Island ceremonial vessels, we find that
the three mounds were probably jointly constructed and
capped over a 75-year (or shorter) period from A.D. 350-475.
This overlaps with the Kolomoki site in time but is clearly a
much different pattern. It shows that the observed ceremonial
aspect at McKeithen was a short-lived phenomenon and that
for the bulk of it's 700-year occupation, the outward trap-
pings of what Milanich appropriately calls a protochiefdom
are absent.
Detailed discussions of the ceramics from the mounds,
both their physical properties and possible symbolic mean-
ings, are presented in the next two chapters. Ann Cordell's
analysis of the ceramics from the mound is well done and
clearly presented. She points out that the pottery recovered
from Mound A is a utilitarian assemblage and is very similar
to the midden pottery while the Mound B and C pots are
exotic in form. It is suggested, although not demonstrated,
that the Mound A pots might have been used in bone-
cleaning activities and were left on the charnel mound when
it was capped. Jim Knight's discussion of the possible
symbolic meaning of the Mound C pots is interesting,
provocative, and takes Weeden Island studies to a new level
of abstraction. I am very sympathetic to this approach, but I
would feel a lot more comfortable with it if he had tried to
link his interpretations to known animal symbols for various
Southeastern Indian cultures.
The final chapter in this book is entitled "Interpreting
Weeden Island." Here Milanich and his co-authors take
what they learned from the McKeithen site and apply it to
other Weeden Island sites, especially Kolomoki. The success
of this approach is contingent on the quality of data for the
sites it is being applied to. In the case of Kolomoki, I do not
think that the authors are very successful, but it is not for a



lack of trying. Aside from what Sears has written about the
site itself ,we know very little about Kolomoki culture.
However, they do present an intriguing discussion of the
geometry of the site and, I think, erroneously state "...the
importance of Kolomoki as a center declined as the village
population apparently rose to a level from which new
villages budded off' (p. 190). Indeed, I think it can be
demonstrated that the population of Kolomoki stayed fixed
at the site as the influence of the political elite declined. If
there was any fissioning it certainly did not occur in the
interior around the Kolomoki site.
My brown Pinto was traded in for a green Rabbit long ago.
After a succession of non-animal cars I now drive a blue
Taurus (I wonder how Jim Knight would interpret that). The
life of the McKeithen excavations have survived these
symbolic changes in my transportation and will, I believe,
continue to inspire new ways of looking at Woodland
cultures and to provoke many ideas. It is fortunate for those
of us interested in Southeastern archaeology in general and
the Woodland Period specifically that the University of
Florida Press has re-issued this excellent book. The original
Academic Press volume was too expensive for many of us to
purchase. However, this reissue is modestly priced and of
high quality and really should be read by anyone interested
in the archaeology of the Southeast and in the archaeology of
political organization.

Cahokia and the Archaeology of Power. Thomas E. Emer-
son, University of Alabama Press, Tuscaloosa, 1997. xv +
317 pp., figures, tables, references, index, $29.95 (paper).

Anthropology Department, University of North Carolina,
Chapel Hill, North Carolina 27599-3115.

Archaeologists are becoming increasingly more sophisti-
cated in the questions we ask about the people we study. We
are also becoming more sophisticated in the perspectives we
bring to the material evidence we recover and the interpreta-
tions of the patterns we observe in our data. Thomas Emer-
son's Cahokia and the Archaeology of Power is a testimony
to that increasing sophistication. It asks questions, and
employs perspectives, that would not have been seen 25 years
Cahokia and the Archaeology of Power is an important
book. I think it makes a substantive contribution to our
understanding of the development of the Cahokia polity. It
is also a fine example of the productive use in archaeology of
the concepts of cultural reproductiono, structuration,
ideology and hegemony. In sum, I believe that it is a worthy
integration of archaeological data (settlement patterns,
material culture) and anthropological theory.
Emerson's study outlines the theoretical basis for his
approach to the archaeology of the Cahokia chiefdom. The
study is based on several important theoretical concepts. The
first of these is that cultural patterns are the products of the

actions of individuals. Emerson (like many other anthropol-
ogists today) borrows the idea of structuration (the relation-
ships between the practice of individuals and the system that
those actions create) from the sociologist Anthony Giddens.
Integral to this idea is the dual nature of culture and of social
structure. The social system and social structure are the
medium through which individuals act and the product of
those actions. This view links the patterns of behavior
created by shared culture and the possibility of change. It
also allows the possibility of change without intention. Here,
Emerson cites Michael Foucault "people know what they do,
they frequently know why they do what they do; but what
they do not know is what what they do does."
Emerson also uses the concept of hegemony derived from
the writings of Antonio Gramsci. Gramsci saw hegemony as
a concept linking the coercive and persuasive aspects of
power. He clearly linked hegemony to dominance, but he
also pointed to resistance to a dominant ideology by subordi-
nate groups. Hegemony for Emerson is the exercise of power
by a dominant or subordinate group through coercion and
consent based on ideological grounds.
The ideological component of domination often has
material manifestations. Emerson points to ceremonies
associated with creation of sacred landscapes, symbols of
individual power (exotic ritual objects that symbolize social
distance and esoteric power), and symbols of warfare, and
notes that these can be seen in the archaeological record. He
notes that hierarchically organized societies typically have
sumptuary rules that regulate the distribution of some
ideologically (religiously or politically) charged goods and
items of wealth. Further, elites use architecture and artifacts
to signify their political and religious power.
Emerson proposes that such architectural forms and
artifacts can be recognized in the archaeological record and
that, in the American Bottom, their "context, appearance,
ubiquity, and disappearance can be used to trace the rise and
fall of centralized Cahokian power." Thus, he argues that
changes in the power of the elite should be visible in changes
in the presence and absence of the elite's symbols in the rural
In Chapter 3, Emerson provides an overview of the
archaeology of the American Bottom, focusing on the
material culture and chronological framework. This is a good
review of the recent work in the area and of the current
models of the evolution of the Cahokia chiefdom. These
models see Cahokia as a short-lived complex chiefdom
(certainly not a state) with considerably less regional power
and influence than many of the previous models. There is
less agreement, however, on the degree to which the Cahokia
polity was centralized. Some see Cahokia as a loosely
integrated system, while others (including Emerson) see an
extremely centralized polity (at least for a brief time in the
11th and 12th centuries).
In Chapter 4, Emerson presents a broad overview of Miss-
issippian settlement patterns across the eastern United States.
He notes the widespread appearance of dispersed settlements
during the Mississippian period. (For instance, the Apala-

Tax Frowns ANTamo n

8 991 voL sl(4)




chee appear to have had a pattern of dispersed communities
rather than nucleated villages over much of their history.)
He goes on to define two patterns of integration of rural
settlements into the larger Mississippian societies, patterns
he calls direct and sequential. In Emerson's direct mode,
farmsteads and hamlets were basically food-producing units
directly tied to a mound center. In his sequential mode,
lower-level sites were functionally differentiated and may
have included several different types (e.g., farmsteads, civic
nodal sites, and ceremonial nodal sites). Farmsteads were
integrated through incorporation into dispersed communities
centered on specialized nodal sites. Together, a nodal site
and its associated farmsteads could be thought of socially as
a community and spatially as a contiguous group
Emerson argues that in the Cahokia system, rural settle-
ments were functionally differentiated. Further, he sees this
functional differentiation as a specific case of sequential
integration of rural populations.
In Chapter 5, Emerson describes the archaeological
evidence from rural sites within the Cahokian polity. He
focuses on evidence for "architectures and artifacts ofpower"
and the creation of sacred landscapes related to the identity
of the Cahokian elite. Looking at small rural sites, he sees
several distinct types: farmsteads and specialized political,
religious, and mortuary sites. He argues that the appearance
and disappearance of these specialized sites is a reflection of
political and ideological changes at the center of the Cahokia
In Chapter 6, Emerson argues that patterns seen in
settlement organization, architecture, and material culture of
rural sites reflects powerful hegemonic control exerted from
the major political centers rather than an imitative reflection
created by independent rural people. He sees the elite of
Cahokia maintaining their control of the rural population by
appointing rural functionaries who resided at the nodal
communities. He goes on to argue that the purpose of this
control was to intensify agricultural production to support the
central elite.
In Chapter 7, Emerson uses ethnohistorical and ethno-
graphic accounts of eastern US societies as guides to inter-
pret the symbolism and meanings underlying artifacts found
at Cahokia and surrounding sites. He suggests that Cahokian
religious beliefs were dominated by a fertility, world-renew-
al, mortuary complex (antecedent of historic Green Corn
ceremonialism and the "Busk").
In Chapter 8, he suggests that it is likely the Cahokian elite
manipulated this communally-based ideology to increase
their own power and to create a divine chieftainship at the
height of Cahokia's power.
In Chapter 9, Emerson paints a picture of Cahokia's
long-term history. During the Lohmann Phase (AD 1050 to
1100), the period during which the Cahokia chiefdom
emerged, Emerson argues that the rural population was
integrated by community and religious-based leadership.
During the Stirling Phase (AD 1100 to 1200) the Cahokia
elite consolidated political power and appropriated com-
munal ideologies. However, this consolidation was fragile

and the extremely centralized structure collapsed during the
Moorehead Phase (AD 1200 to 1275). Finally, the chiefdom
disintegrated during the Sand Prairie Phase (AD 1275 to
I am much in favor of using archaeological data (particu-
larly those from the eastern US) to examine fundamental
questions about people and their societies. I believe that it is
by addressing these kinds of questions that we can eventually
lay to rest the view that archaeology is not really relevant to
anthropology as a broader discipline. This is not to say that
we should address only questions that cultural anthropolo-
gists think are important. Our questions also are important.
We should be concerned with broad, long-term cultural
processes. We should be concerned with the relationships
between behavior and the material world. We also should be
concerned with local cultural histories, the reconstruction of
the structure and operation of societies, and the reconstruc-
tion of the lifeways of the people in those societies. It is to
Tom Emerson's credit that he strikes a balance between these
I am also in sympathy with an actor-based approach
focusing on the production of culture as a means of under-
standing the past. People do create culture by their actions.
People do live and act within systems created by previous
actions. Models that incorporate individual actors and an
awareness of the constraints imposed on such actors by the
cultures within which they operate are powerful tools for
examining and understanding the past.
However, I am not totally convinced by Emerson's model
of the structure of the Cahokian polity. I am not fully
convinced that the rural nodes he sees reflect control im-
posed by central authorities at Cahokia. My difficulty lies
not in the patterns Emerson sees but in his interpretation of
those patterns. I think that there may be other possible
interpretations (e.g., that the remains reflect the operation of
local communal cults and the distribution of elite goods to
local leaders). Such interpretations have been suggested to
account for "elite" goods found at rural sites in other Missis-
sippian systems.
Having said this, I must stress that Emerson's model does
accommodate the patterns in material remains that have been
recovered from the Cahokia area. Also in Emerson's
defense, there are ethnohistoric examples from the Eastern
Woodlands of chiefs imposing control on outlying communi-
ties (e.g., Powhatan). Clearly, his model of the evolution of
the Cahokia polity is a reasonable one. It should stimulate
much important work (as well as serving as a model for
much of that work).
Cahokia and the Archaeology of Power is an important
book. It belongs on the shelves of all archaeologists inter-
ested in the Mississippian societies of the eastern United
States or in the archaeology of intermediate societies (chief-
doms). It also belongs on the shelves of cultural anthropolo-
gists interested in inequality and the creation of social
identities (especially elite identities). I highly recommend it.


About the Authors:

Laura Kozuch was born in Illinois in 1961, and moved to Florida in 1980. She received her Ph.D. in May
1998, and has worked for the past year at the St. Louis District of the Army Corps of Engineers. Her research
interests include shark usage by humans, shell artifacts, fish and shellfish consumption, the Muspa tribe, and
fauna of all types.

Irvy R. Quitmyer is a graduate of the University ofFlorida. He holds ajoint appointment in the Environmental
Archaeology Laboratory and the Invertebrate Paleontology Laboratory at the Florida Museum of Natural
History. Human ecology of maritime people, fauna ofprecolumbian shell middens, and skeletal growth of
aquatic organisms are among his areas of study.

Lee A. Newsom is Assistant Professor and Curator for the Center for Archaeological Investigations at Southern
Illinois University at Carbondale, with joint appointments in the departments of Anthropology and Plant
Biology. She received her Ph.D. from the University of Florida in 1993 based on the paleoethnobotany of a
series of Caribbean archaeological sites. She continues to work with archaeobotanical assemblages in Florida
and the Caribbean islands.

Arthur R. Lee is an avocational archaeologist, director of the Southwest Florida Archaeological Society's
(SWFAS) Craighead Laboratory, and a past president of the Florida Anthropological Society.

John G. Beriault is the current and founding president of the Southwest Florida Archaeological Society and
a past president of the Florida Anthropological Society.

Jean Belknap is an avocational archaeologist and paleontologist, and long-time analyst at the Craighead
Laboratory. She is a member of both SWFAS and the Time Sifters Archaeological Society.

Walter M. Buschelman, an avocational archaeologist who has been a member of the Craighead Laboratory
staff for many years, has served on the directorate of SWFAS and is a member of the Florida Anthropological

John W Thompson is treasurer of both the Florida Anthropological Society and SWFAS, and is a staffmember
at the Craighead Laboratory. An avocational archaeologist since the early 1970s, he has served as trustee of
the Missouri Archaeological Society, president of that society's St. Louis chapter, and trustee of the Cahokia
Mounds Museum Society.

Carl B. Johnson, an avocational archaeologist, is a member of SWFAS and of the Craighead Laboratory staff.

Robert L. Knight is a consulting environmental scientist who lives in Gainesville, Florida. He received his
Ph.D. in systems ecology from the University of Florida in 1980. As a river diver in Florida, he actively
participates in the state's Isolated Finds Program.

Karl T. Steinen received a M.A. in anthropology from Florida Atlantic University in 1971 and a Ph.D. in
anthropology from the University of Florida in 1976. His main research interests are political and religious
dynamics during the Woodland Period on the Gulf Coastal Plain. He is a Professor of Anthropology at the
State University of West Georgia.

John Scarry is currently Research Associate Professor of Anthropology at the University of North Carolina
at Chapel Hill. Before moving to North Carolina he was the Director of Program of Cultural Resource
Assessment and Adjunct Professor of Anthropology at the University of Kentucky from 1990 to 1994. From
1976 to 1990 he worked in the Florida Bureau of Archaeological Research. His primary research interests
focus on the Mississippian societies of the Lower Southeast. He has served on the Board of the Florida
Anthropological Society (1983-1984) and was Membership Secretary of the Society from 1984 to 1987.

250 F2 P134 7
8B/17/99 347 0

TIhnLm Aurrar~~

8 991 V 51 4

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

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