Front Cover
 Thirty-ninth annual meeting - Florida...
 Editor's page
 Investigations at Hontoon Island...
 Sampling and excavation strategies...
 Appendix - Wetsite excavation strategies...
 Refuse disposal and midden formation...
 Hontoon Island, Florida (8-VO-202)...
 Preliminary study of the animal...
 Analysis of botanical remains from...
 The conservation of wooden remains...
 Book reviews, current research,...
 Join/rejoin the Florida Anthropological...
 Back issues order form
 Information for authors
 Back Cover

Group Title: Florida anthropologist
Title: The Florida anthropologist
Full Citation
Permanent Link: http://ufdc.ufl.edu/UF00027829/00036
 Material Information
Title: The Florida anthropologist
Abbreviated Title: Fla. anthropol.
Physical Description: v. : ill. ; 24 cm.
Language: English
Creator: Florida Anthropological Society
Conference on Historic Site Archaeology
Publisher: Florida Anthropological Society.
Place of Publication: Gainesville
Frequency: quarterly[]
two no. a year[ former 1948-]
Subject: Indians of North America -- Antiquities -- Periodicals -- Florida   ( lcsh )
Antiquities -- Periodicals -- Florida   ( lcsh )
Genre: periodical   ( marcgt )
Summary: Contains papers of the Annual Conference on Historic Site Archeology.
Dates or Sequential Designation: v. 1- May 1948-
 Record Information
Bibliographic ID: UF00027829
Volume ID: VID00036
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
    Front Cover
        Front Cover 1
        Front Cover 2
    Thirty-ninth annual meeting - Florida Anthropological Society
        Unnumbered ( 4 )
        Unnumbered ( 5 )
        Page 1
    Editor's page
        Page 2
        Page 3
    Investigations at Hontoon Island (8-VO-202), an archaeological wetsite in Volusia county, Florida - An overview and chronology - Barbara A. Purdy
        Page 4
        Page 5
        Page 6
        Page 7
        Page 8
        Page 9
        Page 10
        Page 11
        Page 12
    Sampling and excavation strategies at Hontoon Island (8-VO-202) - Ray M. McGee and Bruce K. Nodine
        Page 13
        Page 14
        Page 15
    Appendix - Wetsite excavation strategies at Hontoon Island (8-VO-202), Florida - Bruce K. Nodine
        Page 16
        Page 17
        Page 18
    Refuse disposal and midden formation at Hontoon Island (8-VO-202), Florida - Bruce K. Nodine
        Page 19
        Page 20
        Page 21
        Page 22
        Page 23
        Page 24
        Page 25
        Page 26
    Hontoon Island, Florida (8-VO-202) artifacts - Barbara A. Purdy
        Page 27
        Page 28
        Page 29
        Page 30
        Page 31
        Page 32
        Page 33
        Page 34
        Page 35
        Page 36
        Page 37
        Page 38
        Page 39
    Preliminary study of the animal remains excavated from the Hontoon Island site - Elizabeth S. Wing and Laurie McKean
        Page 40
        Page 41
        Page 42
        Page 43
        Page 44
        Page 45
        Page 46
    Analysis of botanical remains from Hontoon Island (8-VO-202), Florida - 1980-1985 excavations - Lee A. Newsom
        Page 47
        Page 48
        Page 49
        Page 50
        Page 51
        Page 52
        Page 53
        Page 54
        Page 55
        Page 56
        Page 57
        Page 58
        Page 59
        Page 60
        Page 61
        Page 62
        Page 63
        Page 64
        Page 65
        Page 66
        Page 67
        Page 68
        Page 69
        Page 70
        Page 71
        Page 72
        Page 73
        Page 74
        Page 75
        Page 76
        Page 77
        Page 78
        Page 79
        Page 80
        Page 81
        Page 82
        Page 83
        Page 84
    The conservation of wooden remains from archaeological wetsites - Elise V. LeCompte
        Page 85
        Page 86
        Page 87
        Page 88
        Page 89
    Book reviews, current research, comments and events
        Page 90
        Page 91
        Page 92
        Page 93
        Page 94
        Page 95
        Page 96
        Page 97
        Page 98
        Page 99
        Page 100
        Page 101
        Page 102
        Page 103
        Page 104
        Page 105
        Page 106
    Join/rejoin the Florida Anthropological Society - Membership application
        Page 107
    Back issues order form
        Page 108
        Page 109
        Page 110
    Information for authors
        Page 111
    Back Cover
        Page 112
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.



~1~~ -.

MARCH 1987


Investigations at Hontoon
An Archaeological Wetsite
County, Florida

Island (8Vo202),
in Volusia


*~~ 7 2

7` n, r f1

THE FLORIDA ANTHROPOLOGIST is published by the Florida Anthropological Society
Inc., P.O. Box 1013, Tallahassee, Florida 32302. Subscription is by membership
in the Society. Membership is not restricted to residents of the State of
Florida nor to the United States of America. Membership is for the calendar
year, January 1st through December 31st. Except as otherwise noted in
periodic special offers, membership may be received for any current year by
remitting dues for that year ON OR BEFORE SEPTEMBER 30TH of that year. Dues
postmarked or hand delivered on October 1st or later will be applied to
membership in the following calendar year. Annual dues are $12 (Individual),
$18 (Family), $15 (Institutional), $25 or more (Sustaining), $100 or more
(Pafron) and $200 (life). Foreign subscriptions are an additional $5 US to
cover added postage and handling costs for individual, family or institutional
membership categories. Except for back issues purchases, copies of the
journal will only be sent to members with current paid dues.

Requests for information on the Society, membership application forms, notifi-
cations of changes of address, notices of non-receipt or damaged issues, and
information on and orders for back issues should be sent to the Membership
Secretary. Donations should be sent to the Treasurer or may be routed through
the Editor to facilitate acknowledgement in subsequent issues of the journal.
Requests for copies of the Editorial Policy and Style Guide (see FA 37(1)) and
submissions of manuscripts should be sent to the Editor. Newsletter items
should be sent to the President. Address changes should be made AT LEAST 30
days prior to the mailing of the next issue. The Post Office will not forward
bulk rate mail. We publish the journal quarterly in March, June, September
and December of each year. Special issues, the Florida Anthropological Society
Publications series, are published as enlarged issues of regular volume
numbers as funding and available material permit.


Karen Malesky
5312 Bay State Road
Palmetto, FL 33561

Elizabeth Horvath
7404 12th Street N
Tampa, FL 33604

(Three Years):
Ralph Gozlin
7347 Hennessey Road
Jacksonville, FL

Harold Cardwell
1343 Woodbine Street
Daytona Beach, FL 32014

John F. Scarry
P.O. Box 1013
Tallahassee, FL 32302


(Two Years):
Jeffrey Mitchem
Florida State Museum
University of Florida
Gainesville, FL 32611

Marlene Levy
Dept. of Anthropology
Florida Atlantic Univ.
Boca Raton, FL 33431

AGENT: Wallace Spears
109 Blais Trail
Daytona Beach, FL 32018

(One Year):
William Goza
Florida State Museum
University of Florida
Gainesville, FL 33611


Louis D. Tesar
P.O. Box 1013
Tallahassee, FL

Joan Deming
308 6th St. NE
Largo, FL 33540

George Luer
3222 Old Oak Drive
Sarasota, FL 33579


James J. Miller
Div. of Historical Resources
Department of State
The Capitol
Tallahassee, FL 32399-0250

J kn W. Griffin
Route 5 Box 19
St. Augustine, FL 32084

William H. Marquardt
Florida State Museum
University of Florida
Gainesville, FL 32611

Glen Doran
Dept. of Anthropology
G-24 Bellamy
Florida State University
Tallahassee, FL 32306

Gandy Printers, Inc.
1800 S Monroe St.
Tallahassee, FL

Morgan H. Crook
Dept. of Sociology
and Anthropology
West Georgia College
Carrollton, GA 30118

NOIE: In addition to the above Editorial Board members,
others knowledgable in a manuscripts subject matter are
our peer review process.

the review comments of
solicited as part of


2 | OF F Ig Y

S The 1987 Annual Meeting of the Florida Anthropological Society
will be held on May 8-10 at the Clearwater Beach Hilton,
Clearwater Beach, Florida. This event will be hosted by the
Central Gulf Coast Archaeological Society, Twin Cities Chapter
of the FAS.

Registration tables will be set up at 2 PM Friday afternoon. A
welcoming reception, compliments of the CGCAS, will kick off the
weekend at 7 PM Friday. After a full day of paper presentations
S on Saturday, there will be a Western cookout on the Hilton pool-
S side patio, followed by a special guest speaker. Three concur-
rent Sunday morning workshops have also been scheduled, from 9
AM till noon.
o Details concerning accommodations, papers, etc. will be published
in a forthcoming FAS Newsletter. If you are not on our mailing
list, and would like more information, please write or call Joan
Deming, 308 6th Street NE, Largo, FL 33540 (Telephone 813 586-
2347). Special convention rates of $55/night waterfront, $50
streetside are being honored at the Hilton. To make reserva-
tions, call the Clearwater Beach Hilton at (813) 447-9566,
making sure to tell them you are coming for the FAS meeting.
A one nights deposit will be requested to hold your reservation.

Tell your friends to mark their calendars for this event. Feel
free to duplicate the registration form which follows. Regis-
tration will be $10 in advance; $12 at the door. The Western
buffet type banquet will be $15 per person, which includes tax
and gratuities. Reservations MUST be made in advance.
S Hope you can join us!

(Detach here)

(This form may be copied)

o Address

S Registration @ $10 each

Banquet @ $15 each


Please make checks payable to the Florida Anthropological
Society. Mail completed form and check to Joan Deming, 308
6th St. NE, Largo, Florida 33540.






MARCH 1987


Investigations at Hontoon Island (8Vo202),

An Archaeological Wetsite in Volusia

County, Florida

Editor's Page ....... ..................... 2
Acknowledgements by Barbara A. Purdy .. ........ ......... 3
Investigations at Hontoon Island (8-VO-202), An Archaeological Wetsite in
Volusia County, Florida: An Overview and Chronology by Barbara A. Purdy 4
Sampling and Excavation Strategies at Hontoon Island (8-VO-202)
by Ray M. McGee and Bruce K. Nodine .. . .. .. 13
APPENDIX: Wetsite Excavation Strategies at Hontoon Island (8-VO-202),
Florida by Bruce K. Nodine . . . . . .. 16
Refuse Disposal and Midden Formation at Hontoon Island (8-VO-202),
Florida by Bruce K. Nodine . . . .. . 19
Hontoon Island, Florida (8-VO-202) Artifacts by Barbara A. Purdy . 27
Preliminary Study of the Animal Remains Excavated from the Hontoon Island
Site by Elizabeth S. Wing and Laurie McKean . . ... .. 40
Analysis of Botanical Remains from Hontoon Island (8-VO-202), Florida:
1980-1985 Excavations by Lee A. Newsom . . ... .. .. 47
The Conservation of Wooden Remains from Archaeological Wetsites
by Elise V. LeCompte . . . . . . 85
Florida Archaeology, Number 2: four translations of Spanish Documents
by Dr. John H. Hann. Reviewed by W.S. Eubanks, Jr. . ... 90
Archaeological Geology, edited by George Rapp, Jr. and John A. Gifford;
and, Geoarchaeology quarterly journal Reviewed by Kevin McCartney 94
BOOK REVIEW: The North American Indians (Continued from last issue) 95
A Note on Undelivered FAS Journals by Louis D. Tesar . 98
Artifact Recording Methods by Phillip M. Pollock . ... .. 99
CURRENT RESEARCH: Tatham Mound by Jeffrey M. Mitchem . ... 103
ASVNET/ESAFNET Computer Network Announcement ..... . .. ..104
Announcement of the 50th Anniversary of Dr. Joffre Coe's Initial Archae-
ological Work at Town Creek Indian Mound, North Carolina . ... 105
Find New Members: Earn Back Issues Credits Notice ..... .. 107
Join/Rejoin the Florida Anthropological Society Membership Application 107
Back Issues Order Form ....................... 108


This is the first issue of my fourth year
as the Editor of The Florida Anthropologist,
the journal of the Florida Anthropological
Society. With this issue we enter our 39th
year of publication. While in past issues
several changes have been introduced to im-
prove the journal's format and broaden its
content, this issue introduces yet another
change to simplify citation of our journal
and regularize its quarterly publication.
This change is to correct the complicated
numbering system which has occurred over the
years. The problem rests with two aspects
of our publication -- both concern funding.

The first occurred as a result of editorial
budget constraints which limited the number
of pages which could be published to monies
available from annual membership dues. Thus,
if the Editor received material which he or
she believed should be presented in a single
publication, rather than divided between two
or more quarterly issues, to operate within
budget the material was presented as combin-
ed issues, i.e., FA 22(1-4), 36(1-2) and 39
(1-2). When issues have been combined in
this manner, it has been my experience (and
presumably that of my predecessors) that
some of our institutional subscribers auto-
matically send the Editor notice of non-
receipt for the second quarter when the ex-
pected quarterly issue does not arrive.

The second occurs when our occasional ser-
ies, Florida Anthropological Society Publi-
cations (FASP), has been published. Fund-
ing for this series is deposited in our mon-
ograph account and is obtained through do-
nations, grants and back issues sales. While
the first five issues of this series occur-
red as an independent monograph series, be-
ginning with FASP No. 6/FA 25(2 Pt. 2) in
1972 the monograph series has had a dual
designation to indicate that it is a suppli-
ment to our regular quarterly series, The
Florida Anthropologist. In such combined
issues, while their special funding status
has been acknowledged through their FASP
designation, the pagination has continued
that of the regular FA series.

A possible solution to the above problem
would be to drop the FASP joint designation
AND to indicate the "Special Issue" status
of FA issues in which monograph funds are
used to distinguish them from regular FA


issues. (I will make this suggestion as a
formal motion to our Board of Directors at
our next Annual Meeting on May 8-10, 1987.
If you have any opinions on this matter,
please let be know them at or before that

A change which can be made effective this is-
sue is the elimination of the part one/part
two subdesignation when combined FASP/FA is-
sues are published. Furthermore, when such
combined issues are published, the unexpended
regular issue funding will be used for the
remaining regular issues, which may thereby
be enlarged in content. Thus, all issues,
regardless of size or funding source, will be
issued quarterly and designated as single
numbered issues of The Florida Anthropologist.
This should not cause any problems since it
does not affect membership fees and pricing
for back issues will be based on the relative
cost of producing each issue.

It is with pleasure that I publish Florida
Anthropological Society Publications Number
13/The Florida Anthropologist Volume 40 Num-
ber 1. This special issue contains reports
on the results of work conducted at the Hon-
toon Island (8Vo202) site in Volusia County,
Florida. The preservation of normally per-
ishable organic remains at this wetsite gives
us a unique view of Native American culture.

Funding for this issue is derived from member-
ship dues, back issues sales, a $400 donation
from the Apalachee Anthropological Society of
Tallahassee, Florida, and a $1000 donation re-
ceived from the Friends of Anthropology Account,
University of Florida Foundation as a result of
efforts by Dr. Barbara Purdy. Dr. Purdy is also
to be thanked for arranging the preparation of
the galley texts for five of the Hontoon Island
articles; Dr. Wing and Lee Newsom prepared their
final galley texts. Dr. Purdy's efforts are
greatfully acknowledged. Phil Pollock and Kevin
McCartney are thanked for preparing the final
galleys of their articles. I prepared the re-
maining galleys and the final paste-up.

I hope that you enjoy this issue.

Louis D. Tesar
March 5, 1987

Volume 40 Number 1


March, 1987


. I'


The investigations reported in this volume were funded
by the National Endowment for the Humanities, the
National Geographic Society, Wenner-Gren Foundation
for Anthropological Research, and the American
Association of State and Local History. The Graduate
School at the University of Florida provided three
graduate student assistantships to support work related
to the Hontoon Island archaeological project. The authors
are grateful to these organizations and to the Department
of Anthropology and the Florida State Museum at the
University of Florida, and the State Department of
Natural Resources, specifically the rangers at Hontoon
Island and Blue Springs State Park, for furnishing
equipment, facilities, and services. We appreciate the
cooperation of the State of Florida Division of Historical
Resources in granting permission, on several occasions,
to excavate on state-owned land. We wish to acknowledge
also the guidance and contributions of many professional
colleagues, particularly Dr. J.J. Stipp, Department of
Geology, University of Miami and his students for
providing chronometric dates. The success of our project
was largely a result of the enthusiasm of volunteers and
citizens of Volusia County, especially H. Deane Smith.
Partially funding for the publication of this volume was
donated by the Friends of Anthropology, University of
Florida Foundation. Elise V. LeCompte achieved a high
level of excellence in her role as editorial assistant and
we commend Louis D. Tesar, Editor of The Florida
Anthropologist, for his patience and his helpful
suggestions during the preparation of the manuscripts.

Barbara A. Purdy, Professor
Department of Anthropology
University of Florida
Gainesville, Florida 32611

Volume 40 Number 1 THE FLORIDA ANTHROPOLOGIST March, 1987



Barbara A. Purdy

The excellent preservation of all manner of organic materials at water-saturated sites
provides a unique opportunity to study environment, diet, technologies, and artistic expressions
in bone, wood, and other usually perishable substances. The water-saturated site at Hontoon
Island, Volusia County, Florida can be used to demonstrate the kinds of information that can be
derived from excavating permanently saturated zones below the watertable. At Hontoon Island,
for example, we found very uniform deposits suggesting cultural stability from approximately
A.D. 0 to A.D. 1500, followed by an early 16th century cultural contact zone that would not have
been noticed had we restricted our investigations only to the terrestrial portion of the site.


Lacking the opportunity to observe or participate
personally in the cultures they excavate,
archaeologists would agree that the next best situation
is to discover sites where all industries are preserved
so that the interrelationships among material items
and the representativeness of each to the total way of
life can be studied. Such sites occur but, only in areas
of continuous dryness or wetness and few have been
investigated thoroughly. Analyses of the contents of
wetsites have revealed that less than 10% of material
culture consists of nonperishable items, i.e., the
typical pottery and stone excavated from terrestrial
sites. Without the organic fraction, previous views of
the past may be distorted or erroneous and they are
missing information about the environment, diet, and
technologies in wood, fiber, and bone. Even the elusive
functions of stone and shell implements can often be
determined by studying manufacturing techniques on
wood and bone preserved in wetland deposits.

Wetsites include the famous Swiss Lake Dwellings (see
Ruoff 1981), the Maglemosian site at Star Carr (Clark et
al. 1954), the Somerset Levels in southern England
where many kilometers of wooden trackways and other
wooden remains have been excavated (Coles 1982), the
Ozette site on the Olympic Peninsula where a mudslide
entombed a village preserving its contents in pristine
condition (Daugherty 1980), the 2500-year-old fishing
camp on the Hoko River (Croes and Blenman 1976), the
12,000-year-old Monte Verde site in Chile (Dillehay
1984:1) and others such as the Bronze and Iron Age bog
sites in Denmark and England and ancient sites
containing wooden artifacts in Australia.

In Florida, the Court of the Pile Dwellers at Key Marco
with its wooden masks, figurines, and utilitarian
objects was excavated in the 1890s (Cushing 1897;
Gilliland 1975). The Fort Center site on Fisheating
Creek near Lake Okeechobee with its large wooden
sculptures recovered from a charnel pond was
investigated throughout the late 1960s and early 1970s
(Sears 1982). The slough at the Little Salt Spring site,

estimated to contain one thousand human burials with
accompanying grave goods, has been excavated by
Clausen et al. (1979). The recently discovered deep peat
deposit at Titusville contains a full range of
archaeological materials, including wooden artifacts,
and 7000-8000-year-old human skeletons with brains
still remaining in the crania (Doran, personal
communication 1984). There are many other sites
where wooden artifacts and human skeletal material
have been found (MacDonald and Purdy 1982).

Because of the growing awareness that archaeological
wetsite components offer a more comprehensive view
of past cultural activities and because a number of
impressive wooden artifacts have been recovered from
Florida's wetlands, usually as a result of development
projects, it seems timely to report the results of
excavations and analyses from an archaeological
wetsite in Florida that have been ongoing since 1980.

State-owned Hontoon Island is located in Volusia
County, Florida midway along the 350-mile (560-km)
course of the St. Johns River (Figure I). It is managed by
the Florida Department of Natural Resources, Division
of Recreation and Parks. In December 1980, based on
information gathered during a statewide survey of
wet- sites, an archaeological research permit was
obtained from the Florida Department of State, Division
of Historical Resources and a 3-meter square was
excavated below the water table in a shallow lagoon
area adjacent to a large shell midden on the island. A
combination of hydraulic excavation and evacuation
techniques was used as a primary recovery method for
wholly submerged deposits. The unit was excavated to
a depth greater than 1.5 m below ground surface, I m of
which was cultural material and the remaining was
culturally sterile overburden. The screening of all the
samples with water and sluicing was an effective way
to recover even the most fragile materials. The
organic fraction had survived in excellent condition.
Numerous bone pins and awls, adzed wood, and items
suggestive of trade were found. Remains of 48 species
of animals and 28 species of plants were identified.
From the single 3-meter square, 5,777 pieces of bone,

Volume 40 Number 1


March, 1987









Figure 1. Hontoon Island, Florida (8-VO-202) (prepared by L.A. Newsom).

3,624 pieces of wood, and 3,265 pottery sherds were
recovered from a matrix of peat and shell. The actual
dynamics of site formation at Hontoon Island are still
not understood completely but judging from the vast
accumulation of wood chips and a broken canoe paddle,
it is virtually certain that canoe making and mooring
were taking place; the recovery of numerous cut,
shaved, and charred posts suggests that simple pile
structures were built in or near the water edge. Three
large wooden carvings were found near Hontoon Island
in the past, an owl (Bullen 1955), pelican, and otter
(Figures 2-4). The owl is now located at Fort Caroline
and the latter two figures are on display at the Museum
of Florida History, Tallahassee.

In February 1982, a second archaeological research
permit was obtained and a 2-m square was opened at
Hontoon Island, approximately 10 m from the previous
one. The purpose of the excavation was to retrieve
wood from precise locations in the profile so that they
could be radiocarbon dated to determine if the peat
matrix was contaminating the wood samples thereby
yielding inaccurate dates. After culturally and
biologically sterile sand was reached at a depth of 1.5
m, the profile was straightened and a column sample
divided into 30x30xl0-cm segments was cut from the
face extending from the ground surface to 140 cm
below surface. Volumetric samples furnish an exact
representation of each species and, thus, a unique
opportunity to gain a fuller understanding of the
prehistoric economy and the natural environment. With
such information it is possible to assess the
importance of the use of plants relative to the use of
animals by prehistoric people.

Since the processing of the column sample produced
some startling results, a third archaeological research
permit was obtained and a 2-m x 26-m trench with a
6-m lateral extension adjacent to the trench (Figure 5)
was opened during a University of Florida
archaeological field school, Spring 1984 to (I) verify
finds of previous tests including evidence of extreme
modifications that occurred following European contact
in the 16th century, (2) expose the shell mound
stratigraphy to culturally sterile sand to determine the
original configuration of the island, (3) document the
loss of information as one moves from the permanently
saturated zone to the dry part of the midden, (4) obtain
large quantities of wood to study woodworking
techniques and preservation requirements, (5) recover
wooden carvings similar to those found previously near
Hontoon, (6) obtain and identify floral remains to build
a comparative collection such as has been completed
for faunal remains at the Florida State Museum, and (7)
procure volumetric samples from the profile of the
trench and continuous core samples adjacent to the
trench to confirm that the previous column sample,
taken in 1982, represents a pattern and not just a
fortuitous occurrence from an arbitrarily chosen
location. In May 1985, when the water table was much
lower than during the excavation of the trench in 1984,
it was possible to excavate partially two additional
sections of the trench and trench extension.

Examination of the materials from the test squares,
the first column sample, and observations of the trench
deposits reveal that the major food items utilized at
Hontoon Island, as expected, were aquatic species. Of
great interest, especially because America will soon be
celebrating the quincentenary of Columbus' voyage, is
the discovery of abrupt changes in diet, woodworking,
and artifacts around A.D. 1500 including (I) complete
abandonment of shellfish as the food staple, (2)
indication that wood may have been cut with metal
instead of aboriginal marine shell or stone, (3) decline
in numbers of marine shell tools but increase in other
artifact classes, (4) quantum increase in squash
remains and the addition of new varieties, (5)
increased use of turtle and channel catfishes and
decreased diversity of faunal species utilized, (6)
change in design on some bone pins, (7) appearance of
Spanish objects including a copper coin and Majolica
dating to the early 16th century, and (8) European
influence on the shapes of some pottery vessels.

Stratiaraphv and Chronology

The primary reasons for excavating a 3-m square in
1980 were to expose an area large enough to view the
stratigraphy and to retrieve a representative sample of
the fragile artifacts, flora, and fauna for analysis.
Zone I consisted of approximately 50 cm of culturally
sterile organic overburden followed by 10 cm (Zone 2)
of midden containing copious amounts of pottery, bone,
wood, etc., but no shell. Zones 3 and 4 contained all
manner of artifacts and ecofacts in a matrix of shells
of freshwater species. Zone 5, a sand deposit,
apparently predated human occupation in the area
because artifacts were found only in the Doorly defined
transition area between Zones 4 and 5 (Figures 6 and 7).
This stratigraphic sequence has been observed in all
subsequent excavations at Hontoon Island except that
Zone I is thicker farther out in the lagoon and. on the
terrestrial portion of the site, Zones I and 2 are missing,
Zone 3 is truncated, and the shell midden proper (Zone
4) is much thicker.

Dr. J. J. Stipp, Director of Alpha Analytic, Inc. and Beta
Analytic, Inc., Coral Gables, Florida and his students at
the University of Miami, Craig Barker and Patricia
Johnson, generously provided free of charge all of the
chronometric dates received for Hontoon Island
beginning with the 2-m square excavated in 1982.
Table I lists the radiocarbon dates received for
Hontoon Island in 1980,1982, and 1984. No dates are
available yet for the 1985 excavations.
Thermoluminescence dates also are shown for 1984.
From the results of radiocarbon analysis of wood
samples from the 1980 excavation, we had a general
idea of the time period of the site. In 1982 we dated
samples taken at the contacts between Zones I and 2
(60 cm below surface), Zones 2 and 3 (74 cm below
surface), and Zones 4 and 5 (122-140 cm below
surface). When the 1984 trench was dug at Hontoon
Island, we made arrangements with Dr. Stipp to date
samples by both radiocarbon and thermoluminescence

Figure 2. Owl Totem Found in the St. Johns River Adjacent to Hontoon Island,
Florida in 1955. The Carved Portion is 2 M Long and the Entire Specimen is 4 M


Figure 3. Pelican Carving and Replica Found in the St. Johns River
Adjacent to Hontoon Island, Florida in 1977. The Carved Portion is
Approximately .70 M (see Schwehm 1983).

Figure 4. Otter Carving and Replica Found in the St. Johns River Adjacent to
Hontoon Island, Florida in 1977. The Carved Portion is Approximately .70 M (see
Schwehm 1983).

24 25
4 j1 6

0 N

0 2 4
I I t

Figure 5. Plan View of 1984 Trench at Hontoon Island, Florida (8-VO-202)
(prepared by L.A. Newsom).


Zone I

50 cm b. s.
Zone 2 60 cm b.s.

Zone 3 74 cm b.s.

Zone 4

140 cm b.s.
Zone 5 150 cm b.s.
Figure 6. Stylized Drawing of the Profile of the 1980 Test Pit at the Hontoon
Island Site (8-VO-202), Florida (prepared by L.A. Newsom).

Figure 7. Photo of Stratigraphy in the 1980 Test Pit at the Hontoon Island Site
(8-VO-202), Florida. From top to bottom, deposits consist of approximately (1) 50
cm of culturally sterile organic overburden, (2)10 cm of a shell-less organic
deposit containing abundant cultural remains, (3)14 cm of a deposit dominated by
freshwater clam (Elliptio sp.), (4)70+ cm of a deposit containing freshwater snail
(Viviparus sp.) and freshwater clam (Elliptio sp.), and (5) culturally sterile sand.


Lab Sample Depth Age Calendar Age
(cmbs) (years BP) (A.D.)

Beta-2518 wood 100 55060 1400
Beta-2519 wood 130 76070 1190
Dicarb 1994 wood zone 2 10060 1850
Dicarb 1992 wood zone 3 22045 1730
Dicarb1993 wood zone 4 73050 1220
Dicarb 1991 wood zone 5 72055 1230
Dicarb 2011 wood zone 5 640135 1310


UM Sample Depth Uncorrected Age Corrected Age Calendar Age
(cm bs) (years BP) (years BP) (AD)

2602 peat 60 19050 17050 1770
2604 untreated charcoal 74 25050 26050 1680
2605 treated charcoal 74 35055 40070 1540
2637 wood 90-100 48090 1470
2606 treated charcoal 122 87055 87060 1050
2607 untreated charcoal 122 86045 86555 1060
2608 wood 122 104050 100060 920
2609 Mercenaria 122 114045 111070 810
2610 Busycon 125-130 111080 112095 800
2611 wood 140 102045 930



Sq. Depth Sample Age Calendar Sample Age Calendar UM
(cm) (BP) Age (BP) Age

Viviparus 317050

Viviparus 309070
wood 60040

Marine shell


1220 BC 3003

1140 BC 3005
AD1350 3009

170 BC
AD 680
1830 BC
70 BC
130 BC


1000 BC 3002

lModified from Barker (1982:22).

Table I. Hontoon Island Chronology


420 90
1920 70

AD 1510
AD 1530
AD 1550
AD 64
AD 690
AD 750
90 BC
40 BC
AD 30
740 BC
650 BC
550 BC
AD 100
AD 200
AD 250

analyses. We were interested (1) in receiving
additional verification of the date for the contact
between Zone 2 (shell-less cultural zone) and Zone 3
(shell midden), (2) in determining the earliest
occupation of the midden within the area of the trench,
and the differences in the dates depending upon the
horizontal position in the trench (i.e., terrestrial or
lagoonward), (3) in comparing the results of
radiocarbon with thermoluminescence, and (4) in
documenting the age of the water-saturated zone, i.e.,
where wood preservation occurs both vertically and
horizontally. The data are revealing although
additional studies are needed because all of the
objectives were not met and some of the results were

Thermoluminescence samples taken in square 35 at 45
cm below surface yielded an average date of A.D. 1530.
The dates obtained from this depth are significant
because they relate to the historic contact period. The
organic fraction from the same location was
insufficient for analysis. Thermoluminescence
analysis of pottery from square 34,105 cm below
surface, provided an average date of A.D. 700 for the
location where wood and other plant materials first
appeared both vertically and horizontally in the profile.
Unfortunately, radiocarbon analysis of Viviparus sp.
from this level, as with all of the freshwater shell
analyzed, produced a date far older than expected.
Pottery and charcoal collected in square 33 at the
contact with the sand horizon, 201 cm below surface,
produced an average thermoluminescence date of 650
B.C. and a radiocarbon date of 170 B.C. The dates
received on wood and freshwater snail are not
considered accurate (see Table I). From square 26 of
the terrestrial portion of the midden, 130 cm below
surface, thermoluminescence dates averaged around
A.D. 175, radiocarbon analysis of marine shell was 70
B.C. and charcoal was 130 B.C. Again, snail shell
(Viviparus sp.) was considered inaccurate.

The thermoluminescence dates provide a nice sequence
and dovetail well with European contact, stone points,
and St. Johns ceramics (Milanich and Fairbanks
1980:148). The radiocarbon dates on wood from the
1984 trench are disappointing and suggest that a source
of contamination exists. The freshwater snail shell
provides dates consistently older than expected, by at
least a thousand years. The best correspondence in
dates between thermoluminescence and radiocarbon are
those from charcoal and marine shell.

Until the chronology at Hontoon Island is better
defined, we conclude that the portion of the shell
midden at Hontoon Island, investigated during our
excavations, ranges in time from approximately A.D. 0
to A.D. 1770. We further conclude that there is an early
European contact zone at the site that is manifested by
the presence of some Spanish artifacts and all manner
of changes that occurred about A.D. 1500. The site was
probably abandoned around 1763 when the British took
over Florida and the remaining Timucuan Indians left
with the Spanish. This statement is collaborated by
the fact that no domesticated animals of European
origin have been found at Hontoon Island. These

imported species would be expected it tere nad been
more recent European influence in the area. Other
changes that occurred in the contact zone are described
in other papers about Hontoon Island (this issue).
Evidence of an earlier Preceramic Archaic occupation
at Hontoon Island was found in a 2-m square excavated
40 m west of the trench, and ;n I;m;tPA test;in f
large shell midden on the south end of the island.

References Cited

Barker, Craig
1982 Radiocarbon and Thermoluminescent Analyses of Specimens
from Hontoon Island. Work conducted under the direction of Dr.
J.J. Stipp, Professor of Geology, University of Miami. Ms. on file,
Purdy's office, Department of Anthropology, University of Florida.

Bullen, Ripley P.
1955 Carved Owl Totem, Deland, Florida. The Florida Anthropoloqist
8: 61-73.
Clark, J.G.D.
1954 Excavations at Star Carr. University of Cambridge Press.
Clausen, C.J., A.D. Cohen, Cesare Emiliani, J.A. Holman, and J.J. Stipp
1979 Little Salt Spring, Florida: A Unique Underwater Site. Science
Coles, John M.
1982 The Somerset Levels: A Waterlogged Landscape. Proceedings of
ICOM Waterlogged Woodworking Conference, Ottawa, pp. 129-141.
Published by International Council of Museums, Ottawa.
Croes, D.R. and Eric Blenman, Editors
1980 Hoko River: A 2,500-year-old Fishing Camp on the Northwest
Coast of North America. Laboratory of Anthropology, Reports of
Investigations, No. 58, Washington State Univrsity, Pullman.
Cushing, Frank Hamilton
1897 A Preliminary Report on the Exploration of the Ancient
Key Dweller Remains on the Gulf Coast of Florida. Proceedings of
the American Philosophical Society 35(153).
Daugherty, Richard D.
1980 Wetlands Reseasrch at the Ozette Site. In Florida Maritime
Heritage. Proceedings of a conference held in Tampa, March 1980.
Edited by Barbara A. Purdy.
Dillehay, Tom D.
1984 A Late Ice-age Settlement in Southern Chile. Scientific
American 251(4):106-117.
Doran, Glen H.
1984 Personal communication. Dr. Doran is the direction of the
Windover Site project and is a faculty member in the Department
of Anthropology, Florida State University, Tallahassee.
Gilliland, Marion Spjut
1975 The Material Culture of Key Marco, Florida. The
University Presses of Florida, Gainesville.
MacDonald, George F. and Barbara A. Purdy
1982 Florida's Wet Sites: Where the Fragile Past Survives.
Early Man 4(4):6-12.
Ruoff, Ulrich
1981 Die Entwicklung der Unterwasserarchaologie im Kanton
Zurich. Helvetia Archaeoloqica 45/48-62-70.
Sears, William H.
1982 Fort Center: An Archaeological Site in the Lake
Okeechobee Basin. University Presses of Florida, Gainesville.

Barbara A. Purdy, Professor
Department of Anthropology
University of Florida
Gainesville, Florida 32611

Ray M. McGee and Bruce K. Nodine

The techniques utilized at Hontoon Island (8-VO-202)
to ensure that the fraction of material recovered and
analyzed is representative of the site included survey,
sampling, excavation, microstratigraphy, screening,
processing, recording, preservation, and analyses. Some
of these procedures are discussed elsewhere.

The 1980 and 1982 test excavations at Hontoon Island
were conducted below the watertable and a
combination of hydraulic excavation and evacuation
techniques, using pumps, was the primary recovery
method. The 20x26-m trench with 6-m lateral
extension dug during the 1984 field session, and the
two additional units excavated in 1985 (Figure I),
extended from the wholly submerged lagoon edge to the
top of the terrestrial portion of the shell mound.

The location of the trench was selected as a result of
the 1980 and 1982 test excavations and a three-month
survey of Hontoon Island carried out in 1983 by Nodine.
Although the entire 660-ha island was tested at the
time of the survey, the primary intent was to identify
water-saturated areas where organic material was
preserved, to ascertain the depth and extent of the
preservation, and to delimit (and thus avoid) areas that
had been disturbed by shell mining in the 1930s and the
construction of a picnic area in the 1970s. The results
of the survey revealed that there are a number of other
archaeological sites on the island, including large shell
middens, but the deposits containing the superb
preservation of organic materials occur only on the
island's northern perimeter. The survey then
concentrated on this area and 204 soil samples were
collected using a bucket auger. From this population,
20 samples representing two transects from the lagoon
up the shell midden, perpendicular to the contour lines
of the midden, were chosen for analysis. Analysis
revealed that organic materials were preserved along
the entire lagoon and extended under the dry part of the
midden at an elevation of 2.75 m below site datum.

In January 1984, using standard procedure, the trench
was laid out perpendicular to the contours of the
midden as far north as possible without encountering
the area of recent disturbance. During excavation of
the trench, all elevations and distances were
calibrated to a datum that had been established at the
time of the 1983 survey. The datum was located at the
foot of a fire tower. Each 2-m section of the trench
was taken down in arbitrary 25-cm units. Water
excavation was used to recover material from the
submerged deposits, and shovels and trowels were used
to excavate the portions of the trench that lay above
the permanently saturated zones. The recovery of

material below the present watertable without
resorting to diving gear, while not original at Hontoon,
involves techniques not usually utilized by
archaeologists. After logistical problems were solved,
excavation and recovery proceeded more efficiently.
(See Appendix for a description of pumps and problems
associated with wetsite excavations.)

A system to water screen the excavated material was
built by Dr. Paul Gleeson, an archaeologist from the
state of Washington, who had worked for 15 years at
the Ozette Village wetsite on the Olympic Peninsula.
The excavated material was dumped from a
wheelbarrow onto a hopper, flushed through a 2.5-cm
grid which collected large objects, then onto a 1.25-cm
mesh screen where the soil matrix was rinsed away,
and removed to a table where it was sorted into
categories (Figure 2). This proved to be such an
efficient method that a second water screen was
erected so that two squares could be excavated
simultaneously. The spoil from the water screen was
dumped into the mined area of the midden west of the
trench to prevent contamination of the site.

The use of 1.25-cm screen can be defended for the
Hontoon Island site. From several levels of one 2-m
section of the trench, .625-cm mesh was used for
one-half of the square and 1.25-cm mesh for the other
half. Upon comparing the results of these two screen
sizes, we learned that we had lost very little
information by 1.25-cm screening because we
recovered, primarily, only smaller fragments of the
same materials in the .625-cm screen. In addition,
recovery was slowed to a standstill using the smaller
mesh. The use of 1.25-cm screen can be justified also
by noting that any size mesh will result in the loss of
some information and that is why we opted to take
column samples every 2 meters along the trench to
obtain a controlled 100% sample for flotation. Figure 3
shows the location of the column samples and Figure 4
shows the depth of excavation in each square of the
trench. In addition, we took continuous core samples
from several locations adjacent to the trench and into
the lagoon, on line with but beyond the trench. The
continuous cores permit sampling far deeper than is
possible with excavation. Furthermore they can be
preserved and can furnish a stratigraphic record for
future reference. Portions of the cores were used to
obtain samples for radiocarbon dating and for pollen

Microstratigraphy of each 2-m section was drawn by
Robin Teas, a crew member, for the south and west
-valls of the entire 26-meter trench and extension.

Volume 40 Number 1


March, 1987


Hontoon Island
I 984 Trench and Test Pit
1985 Excavation Units

N -~

Figure I. Hontoon Island Site, Florida (8-VO-202) Showing the
1984 Trench and Trench Extension. Squares 39 and
59 were Excavated in 1985.

Figure 2. Method of Water Screening (Top) at the Hontoon Island Site
(8-VO-202), Florida (1930s mined area can be seen in the
background). Screened Material (Bottom) is Moved to a
Table and Sorted into Categories.





Foot Path




39 59

Figure 3. Location of Column Samples (Top) Taken
from Profile of 1984 Trench at the Hontoon
Island Site (8-VO-202), Florida. Bottom
Photo Shows a Closeup of the Profile
Where One of the Column Samples was

Hontoon Island
I984 Trench and Test Pit
1985 Excavation Units

Foot Path



Figure 4. Depth of Excavation in Each Square of
the Trench and Trench Extension at
the Hontoon Island Site (8-VO-202), Florida.

Many postholess," fire hearths, and shell concentrations
were noted. The exact vertical and horizontal location
of preserved plant remains were recorded on the
drawings. The plant material would not have survived
if the watertable had fallen below this point. We
secured wood from this area for radiocarbon analysis
and pottery for thermoluminescence dating (Purdy, this

On the east end of the trench (toward the lagoon),
organic material was preserved at 2.5 m below site
datum, as predicted from the auger samples, and
continued until culturally sterile sand was reached at
3.7 m below datum. An auger sample was taken in the
westernmost unit of the trench to determine if plant
remains had been preserved and to define the base of
the midden. Organic preservation was encountered at
2.75 m below datum and sterile sand at 3.5 meters
below datum. Later, fragments of preserved plant
material were recovered in the western unit (square 25
of the trench; see Figure I for location) during
excavations at 2.5 to 2.75 m below datum. This unit
was not completed because the trench walls collapsed
after several days of heavy rain and there was not
enough time remaining to reexcavate the area. With the
knowledge that organic material was preserved at the
west end of the trench, a test unit 42 m west of, but on
line with, the trench, across the mined area, was
opened to gain additional information about site
development and organic preservation. No organic
preservation was found in this square. Culturally
sterile sand was encountered at 2.5 m below datum.
All levels in the trench contained abundant ceramics,
but no ceramics were recovered in this square. The
only artifacts from this unit were two bone points.

The organic deposit at the bottom of the midden in the
western portion of the trench reveals that this part of
the midden has been continuously wet since the
material was deposited. Since the test unit on the
west side of the mined area has sand below the shell
midden, rather than organic material, the people living
there must have deposited shell, gradually building up
the dry sand rise, and eventually expanding the midden
toward the shallow lagoon area.

While still in the field, the materials recovered
from the trench underwent preliminary sorting,
conservation, and study. The pottery was dried,
placed in major style categories, counted, weighed,
and recorded. The bone was washed, dipped into a
weak solution of polyvinyl acetate (Elmer's glue), and
the species identified and recorded. The wood and
other plant remains were identified and sorted into
categories (nuts, berries, pinecones, fiber, fungi,
etc.), counted, weighed, charted, and stored either
waterlogged or in a preservative to prevent
degradation (polyethylene glycol [PEG] for the wood
and formalin acetic acid [FAA] for other plant parts).
The woods, especially the softwoods, have preserved
well using 60-80% PEG 540 Blend. The wood
recovered from the test pit in 1980, for example, was
removed from PEG in 1981 and appears to be in stable
condition. Experiments are now underway using PEG
along with freeze drying in an effort to solve
problems that sometimes exist with preserving the
hardwoods. This topic is discussed in another paper
(LeCompte, this issue) as is the flotation method
used to separate the material contained in the
column samples (Newsom, this issue).
Ray M. McGee
Bruce K. Nodine
Department of Anthropology
University of Florida
Gainesville, Florida 32611

Bruce K. Nodine

This is a revision of a paper presented in 1984 at the Southeastern Archaeological Conference
in Pensacola, Florida. It was prepared for inclusion in this issue.

The benefits of conducting research at
water-saturated archaeological sites are becoming
increasingly apparent. Associated with these benefits,
however, are problems of excavating below the water
table and of processing large volumes of degraded
organic material.

Each wetsite is unique and so is the problems to be
solved. At Hontoon Island we devised a system, using

three types of pumps, to evacuate water from a trench
whose depth in places was more than two meters below
the water table. (An attempt to use well points in
excavating the 1980 test pit failed and was not tried
again for the 1984 trench although we are aware that
well points have been used successfully at other

The largest of the pumps used at Hontoon Island was a

Volume 40 Number


March, 1987


Pace brand centrifuge pump with a 7.5-cm hose
attached to an 8-hp Briggs and Straton engine. Its
special characteristics make it well suited for high
volume evacuation of water. It has nylon impellers
which are inexpensive to change if there is too much
grit in the water. It is self-priming and can run
without water. This feature is important because the
pump can be run faster than necessary and, when the
water is down to the desired level, it will idle until the
level rises. Most pumps will burn out if they are
allowed to run without water.

A second type of pump used at Hontoon Island was a
mud hog which is a moderately high-volume pump with
7.5-cm hoses and a 5-hp engine. It is a diaphragm
pump commonly used to clean septic tanks. It can pump
mud or anything else that will fit in the intake.

A third type of pump used at Hontoon Island was a
Sears centrifuge pump having a 5-cm hose and 1.5 hp
engine. Several of these pumps were used for
excavation and for screening.

All pumps need routine maintenance. Oil must be
changed after every 20 to 40 hours depending on the
type of pump and the conditions of use. If pumps are
used at different times, it is desirable to keep a log
book to record when pumps are started and shut off.
When possible, oil should be changed when the engine is
warm. Spark plugs for the various kinds of pumps
should be kept on hand to avoid excavation delays. Oil
may get into the cylinder and gum the plug if the pump
is worn. The spark plug is a good place to begin to look
for the source of a problem if the pump is hard to start.
A wire brush and a plug gapper can be used to clean
spark plugs. If the area is dusty, it is necessary to
have some detergent around to clean the filters and
recoat them with a little oil. We did not have a
problem with dust at Hontoon Island.

Pumps may operate for several hundred hours without
much maintenance. The Sears pumps lasted until the
last week of the four-month excavation period. It is
wise to have back up pumps or access to a good

It cost about fifty dollars a week to operate the
pumping system. We found that the pumps used 3.8
liters of gasoline every 4 to 6 hours depending upon the
pump size and the speed of operation. We used about 91
liters of gasoline a week. Since water would fill the
trench rapidly when the pumps shut down, we learned
to distinguish by tone when they were about to run out
of gasoline and we were ready to fill them
immediately. One soon discovers the most efficient
and economical speed at which to run the various

The intake hoses for the pumps should be rigid in order
not to collapse under the suction from the pumps, and
they should be as short as possible. It is difficult for
the pumps to draw water up over a long distance. It is
better for the pumps to take water in easily and then
push it out over whatever distance is required. The
intake lines should have only enough screening on the
ends to keep out rocks, weeds, and other solids.

The Pace pump can pump about 1710 liters/min. It was
the main drainer of the trench. Its intake was put in a
basket that was placed in the bottom of the lowest
unit. The intake hose had a .625-cm mesh around it.
The output for the main drainer should be short and
wide open so the pump will have no output pressure.

The mudhog was used for solids that accumulated in
the bottom of the sump area. As with the Pace pump,
the intake of the mudhog should be short and the output
wide open.

The Sears pumps were not allowed to pump to capacity.
Hose trees were put on the outputs of these pumps so
that the water could be pressurized into garden hoses.
The inputs were put near a supply of clean water. We
found it easiest to repump the water from the output of
the Pace pump and, therefore, we had to be careful that
the Sears pump did not run dry when the Pace pump shut
off. It is best to run some of the hoses from each
small pump to the excavation area and to the screen
area because if one pump runs out of gas or breaks
down, it is possible to run on other pumps.

The effective channeling of water to a pump that can
handle it is an important factor in excavating below the
water table. The strategy was to use square 33 of the
trench as a sump unit and pumping station, and drain
all other units to it. This square was excavated, by
conventional means, to the water table. Using augers,
shovels, and posthole diggers, we then dug a 50-cm
diameter hole in one corner to approximately 75 cm
below the excavated floor. The mud hog intake was put
in the basket in the hole, the unit was drained, and we
continued excavating with trowel, skim shovel, and

There were problems with the stability of the walls as
the sump unit got deeper. From this experience, we
concluded that the sump unit should be taken down only
as far as necessary to drain the adjacent units. We
learned also that, at Hontoon Island, it was more
important to prevent the matrix from eroding than to
attempt to stop the water.

In the beginning we used sand bags but the water
eroded around them and made the situation worse. Next
we used hardware cloth and metal poles but the

hardware cloth was not rigid enough to permit us to
lower it along the wall when the excavation unit got Bruce K. Nodine
deeper. Stainless steel grids used in conjunction with Department of Anthropology
the metal poles worked well as they could be put up in University of Florida
sections and could be pounded down with a maul as Gainesville, Florida 32611
excavation proceeded. Walls that were of permanent
height were terraced using the grids lined with plastic.
This method slowed the water without stopping it.

Since the major problem was water running over the
matrix below its point of entry, we concentrated on
collecting and channeling it directly to the pump's
intake. The hardware cloth was formed into a trough
and lined with plastic to make channelers. Most of the
damage caused by erosion took place when a unit was
draining but this was prevented if the channelers were
in place before the pumps were started. Once a unit
was drained, the walls remained quite stable if the
channelers were used.

Excavation was carried out in a unit higher in elevation
than the sump area. We found that a 10-cm gradient
over a 2-m distance was sufficient. We excavated a
"step" along the entire edge of the excavation unit
closest to the sump. The step extended 10 cm into the
unit and was cut to the bottom of the level to be
excavated. Here we installed a metal pole with a
plastic sheet attached which spanned the length of the
excavation unit and kept the runoff from the excavation
separate from other water. This method controlled the
erosion to levels below the excavation level and
prevented contamination of adjacent units. Excavation
with hoses began when the runoff from the excavation
unit was isolated. Water excavation reduced the risk of
damage to the fragile organic components. Beginning at
the low end of the square, the matrix was rinsed into a
screen "dust pan." Artifacts seen during this process
were measured in situ. The rest of the material was
put in buckets with mesh bottoms and sent to the
screening station. The silt laden water that
accumulated on the plastic sheet was run through a
.625-cm screen and then through a .3125-cm screen.
These screens were cleaned often and the material
from them was also sent to the screening station.
Thus, the water flowing out of the funneling system
contained material less than .3125-cm and this was
channeled directly to the intake of the mud hog.
Occasionally some material got past the mud hog but it
would settle in the bottom of the sump unit since
almost all of the turbulence in the sump was avoided
by directly channeling the clean water to the Pace
pump. When excavation was completed for each level,
the mud hog was used to "vacuum" the sediment in the
sump unit to prevent it from filling.

Using the system described above, all of the excavation
runoff was kept separate. The trench was kept dry by
channeling the inflowing water directly to the high
volume Pace pump and the smaller centrifuge pumps
were used to pump water for excavation and screening.

(8-VO-202), FLORIDA

Bruce K. Nodine

This paper is a modification of Chapter 3 of Society and the Universal Problem of Refuse
Disposal: An Anthropological Analysis (Nodine 1986). It has been prepared for inclusion in this

Archaeologists are usually concerned exclusively with
people's garbage but rarely address refuse disposal as
a primary focus of investigation. Studies of hunters
and gatherers indicate that these people simply move
away from their refuse before accumulation reaches
excess. In contrast, there are many studies of modern
urban centers that attest to the vast problems
associated with present-day disposal. These bodies of
literature represent the extremes in refuse behavior,
i.e., moving habitation away from refuse and moving
refuse away from habitation. The most fruitful arena
for producing and testing models of refuse disposal
behavior lies between these extremes but there have
been few investigations of discard behavior among
nonindustrial sedentary people or semi-sedentary

In writing about human residues, Gould (1980)
addresses and formalizes the quest for predictive
statements concerning a people's discards:

At the simplest or most basic level... whenn and
under what conditions do people go to the trouble
of removing their rubbish and other discards instead
of simply moving away when these residues
accumulate to excess? Perhaps the answer... lies
in the degree of nucleation and permanence of
settlement of a community, or perhaps it has more to
do with the nature of resources beina exploited
(Gould 1980:114).

Schiffer (1972:162) postulates that "with increasing
site population (or perhaps site size) and increasing
intensity of occupation, there will be a decreasing
correspondence between use and discard locations for
all elements used in activities and discarded at the

Investigations specifically oriented to refuse disposal
(e.g., Hayden and Cannon 1983, Murray 1980) attest to
the complexity of refuse disposal processes. For
example, samples obtained from the Human Relations
Area Files (HRAF) indicate the association between
refuse disposal complexity and settlement permanence
is statistically significant (Nodine 1986).
Furthermore, the distribution produced by this
association shows that innovations occur first in the
settlement sphere and are usually followed by

innovations in refuse disposal practices. The cases
particularly important to investigating refuse disposal
processes are those in which the disposal techniques
do not follow changes in settlement pattern,
suggesting that variables other than settlement, such
as the type of resource exploited or social traditions,
may have a primary influence upon disposal.

The prehistoric shell midden located at Hontoon Island
may be considered such an anomalous or "lag behind"
case because the disposal practices appear to be
techniques most often used by nomadic people (i.e.,
kitchen midden or convenience dumping) while the
settlement at Hontoon appears to be sedentary or
semi-sedentary. The inhabitants at Hontoon Island
exploited a wide variety of resources in their fresh
water riverine environment including the fresh water
snail Viviparus georgianus.
Findings indicate that the discard of massive volumes
of this snail's shell and other materials may have
actually produced the land on which the Indians lived.
Furthermore, it is probable that at Hontoon Island the
disposal practices do not reflect merely the local
settlement pattern. The need to dispose of incredible
volumes of non-biodegradable refuse may also have had
a primary influence on places available for habitation,
thus influencing the settlement pattern.

The above assertions were facilitated by using a model
of midden formation produced by May (1982). I altered
May's model for use on the Hontoon Island midden by
adding information supplied by Meehan (1982) who lived
with coastal Australians for extended periods during
1972 and 1973.

May began his study by searching the HRAF for cultures
which exploit shellfish. From the HRAF he obtained a
sample of twenty cultures which he used to produce a
model for midden formation. He then tested his model
using a trend surface analysis. May's strategy was to
predict (from his model) what a processing area would
look like, making it possible to predict what a discard
area would look like (May 1982:246). May was able to
identify areas, or excavation units which were
respectively discard and habitation areas
(1982:237-238) but he was not able to delimit
individual habitation episodes (1982:241). This
situation is understandable for, as the model to be

Volume 40 Number 1


March, 1987


presented below predicts, the processes leading to
midden formation are very complex, leaving, of course,
a very complicated archaeological record.

The dynamics of midden formation can be best
understood through ethnoarchaeology. Betty Meehan
studied the collecting and discard practices of the
Gidjingali of coastal Australia. From her
investigations she produced an excellent book (Meehan

Using new information from Meehan, I altered May's
model for a test on the midden at Hontoon Island. From
excavations and core samples, it is evident that the
base of the shell midden at Hontoon Island was first
deposited on a flat area with a small rise compared to
the surrounding swamp. Starting with this
information, I now present my understanding (to be
tested below) of what people living at Hontoon Island
for over 1000 years did about volumes of garbage.

When first inhabiting Hontoon Island, the Indians would
discard shell or other refuse anywhere around them.
This would produce a pattern much like the map
presented by Meehan (1982:116) of a home base shell
fishing settlement. Once there was some shell
deposition, this area would be even more elevated and
desirable than other areas nearby. Soon a small midden
would develop.

To emphasize this point, I made a transparency of the
map presented by Meehan. By rotating the
transparency, I began to achieve the effect of the
accumulation of repeated depositional patterns. Figure
I is the result of one accumulation. Figure 2 is the
result of two accumulations combined. Figure 3 is the
result of four accumulations. Here, we can see the
emerging pattern: a midden is beginning to accrue with
most of the hearths clustering near the center. Figures
4 and 5 represent seven and nine accumulations,
respectively. If Figure 5 could be produced on the
ground, I believe most shell midden archaeologists
would call it a mound, not a shell scatter, but it was
produced by an accumulation of shell scatters. As this
idealized midden shows, there are more hearths near
the center and more freely associated refuse near the
edge. It is interesting to note that in the late
nineteenth century, Walker (1880, 1883, 1885) proposed
a similar evolutionary framework for the development
of mounds.

Perhaps the Meehan configuration is not so closely
followed as the midden accumulates. Perhaps the
inhabitants would begin to throw their refuse down the
slope more often because they could get the garbage
further away with the same or less energy than before,
thus adding to convenience or economy of effort of
disposal (Hayden and Cannon 1983). In this case the
contrast between the central "hearth" area and the
outer "refuse" area would be even more pronounced. As
one side or area of a midden became too cluttered, the
people would clear it, leveling the undulating piles of
refuse (May 1982:133; Waselkov 1981:59) or move to

the other side of the mound and level the "new" area
(May 1982:133-134). The above descriptions from May
and Waselkov are from pat ethnographic sources.

Meehan presents firsthand information:

There is a predictably complicated pattern for
disposing of food debris in home base camps ....
Various parts of base camps are continually being
moved, distances ranging from a few metres to
several hundred metres. Debris continues to be
deposited on the periphery of each hearth complex.
... At regular intervals, every week or two, the
entire camp area is cleaned up with rakes, sticks, or
feet. The rubbish is dumped in various places around
the periphery of the hearth complex, usually in areas
that are unimportant for use and access. Eventually,
quite large banks of debris form, and these are
occasionally burnt if a lot of grass is growing on
them or if they are attracting large numbers of flies
(Meehan 1982:114).

This activity, although systematic, produces a very
complicated archaeological record. With the Indians
moving around on the mound, leveling old piles of
refuse, burning sections and living on top of their
garbage, the identification of individual habitation
episodes is virtually impossible.

May also had problems identifying individual habitation
episodes (1982:241). He states, however, that
"interpretation of the distribution of ubiquitous midden
variables resulted in the discovery of habitation and
activity areas in the central part of the midden and
discard areas at the periphery" (1982:250).

This is the crux of the model: the processes presented
will produce substantially more refuse on the
periphery and more hearths (activity areas) near the
center. This statement is preceded by May's
presentation of this configuration, Meehan's discussion
of it in her book, and the exercise utilizing
transparencies from Meehan's map. The next step is to
see how well this configuration is represented in the
ground at Hontoon Island.

According to the model, the relative locations of
associated hearth and refuse areas could be in any
configuration at any given time. For this reason, the
most effective way to test for the modeled processes
is to use information from deep intensive excavations.
This will not bring to light the daily or even yearly
process but will illuminate the result of over 1500
years of depositional processes.

This test will complement May's work which utilized
extensive surface information. As mentioned above,
May looked for activity areas which helped identify
discard areas. I will look for discard areas which will
help define activity areas. Whereas May used two types
of shell and sandstone fragments, I will use ceramics,
bone, and features as variables to test for the
predicted configuarion.

Hearths are shown thuS
The rubbish heaps surrounding Pthe i ing areas are ,ndicated thus ''
and here the middens were dense and up to 0 5 m thick Elsewhe'e the
mdder's became thin, with sparse scatters of shells and other food
remains, indicated thus .-


-'As'-`'~" ~~

~ ~ ` '* .
q .-

' 2 t ', .n

r o..
.f .\ *.-

< :* ?/




Figure I. Result of One Accumulation of Shell Disposal

(from Meehan 1982).

Figure 2. Result of Two Combined Accumulations of Shell Disposal

(achieved by rotating a transparency of Meehan's [1982]


. C

p- .

' tr





* $- 1


~ ,,


-I ,...


*?. *

*^ 0 I -** "

"' "^

'~ "'~

.: ..... ... --:^ ^- s^.*:;:,^ +/' "" "
,r:. :-^, _---.^ "^ "- .-
.... .,I .....

A: I C .
"* .H i *'
.- .~ .-

6o,. -. .,- ,.--C _

I- w

Figure 3. Result of Four Combined Accumulations of Shell Disposal Figure 4. Result of Seven Combined Accumulations of Shell Disposal
"-. -.d ... .

+j-7 +,~ CA

Figure 5. Result of Nine Combined Accumulations of Shell Disposal (achieved by
rotating a transparency of Meehan's [1982] map).

Square Number of Pottery
Levels1 (kg)





Pottery Density


Bone Density


1 Each level was excavated as a 2-m x 2-m x .25-m unit, hence each level equals 1 m3. The
squares which have half levels (e.g., 6.5) represent squares which were on an incline, thus the
first level was not a full cubic meter. The slope was such that it generally cut the first level's
volume in half on units affected.

Table 1. Bone and Pottery Weights and Densities by Square

Figure 6. Features by Square and Elevation in the 1984 Trench Excavated at the
Hontoon Island Site (8-VO-202), Florida.

Figure 7. Features by Square in the 1984 Trench Excavated at the Hontoon Island
Site (8-VO-202), Florida.

The deep trench excavated at Hontoon Island provides
nearly continuous information of a cross section of the
midden. Even more important to the testing strategy is
the nature of the deposits at the lower end of the
trench: here the shell and other deposited debris is
mixed with peat. Because of the nature of peat
formation (i.e., it must remain constantly saturated and
virtually undisturbed) these units have not been dry or
disturbed since the deposit began accumulating about
2000 years ago (Purdy and Newsom 1985:567).

This point is significant because it helps set up the
differences between the activity areas and the refuse
areas at Hontoon Island. Activity areas could be
anywhere on the dry part of the midden. Areas above
the water table should have attributes of both activity
and refuse while areas which have been constantly
saturated should have attributes mostly of refuse

I will use three variables to test the predications
resulting from the model presented above: density of
pottery, density of bone, and presence/absence of
features. I will first use the pottery and bone to define
the peripheral refuse area and then bring in features to
show the contrast of activity areas. Moreover, if the
model is accurate, the relative location of feature
areas should move generally away (up and out) from the
core of the midden as it accumulates. This type of
movement may be necessary for two reasons: (I)
vertical movement is necessary to stay on top of the
accumulating refuse, and (2) horizontal movement is
necessary to stay in close proximity to the water as
the midden "grows" out.

The density of pottery in each unit along the trench is
Calculated by dividing the total dry weight of sherds
from each unit by that unit's volume. The density of
bone is similarly calculated for each of the eight units.
(Note that the bone used is shattered bone and bone
debitage, not bone artifacts). I believe that the areas
where the density of pottery and bone are highest were
more often considered refuse areas. It does not matter
Sif these materials represent primary or secondary
refuse. A broken pot or a splinter of bone may have
been used as convenience tool and may have lain around
on an activity area for a while. Eventually, procedures
were sometimes taken to move these materials to the
periphery (Table I).

SUsing the pottery density as the dependent variable and
the distance from the top of the trench as the
independent variable, pottery increases exponentially
with r2 = .73, corr. coef. = .85 with 6 degrees of

SThere are several factors which may cause this type of
configuration other than the processes mentioned. The
first is differential preservation. If the bone were to
decay more rapidly on the terrestrial part of the
midden, bone density would increase down the trench.
In fact, the bone in all areas of the trench was in
excellent condition. Also, the pottery is not subject to
the decay processes which may act on bone. The notion

that differential preservation is not a factor here is 25
further supported by the high correlation of the bone
density to the pottery density: r2 = .94, corr. coef. =
.97 with 6 degrees of freedom.

The second factor is slumping, or the oozing effect
which may occur on inclined sites (Rick 1976; Rowlett
and Robbins 1982). Some downslope movement
probably occurred but the stratigraphy is intact and
discrete. This can be seen easily when considering the
quantity and condition of the features encountered
during the excavation. Ruling out differential
preservation and gross post-depositional movement,
the exponential increase of pottery and bone density
when moving towards the lagoon presents a good case
for the identification of the periphery of the midden as
the place most often used as refuse areas.
Features (hearths, post molds, pits) represent a
discrete location of previous activity (Schiffer
1976:115). According to the model, these items should
be far more prevalent on the upper part of the midden.
Figure 6 shows the location and types of features
encountered along the trench. This figure depicts the
features in their actual elevational locations. The
lowest location of hearths occurs at or near the same
elevation in several units. This elevation is about I m
above the elevation previously mentioned as the lowest
level of the water for at least the last 1500 years.
Surely, the water level fluctuated within certain
bounds and the activity areas were moved accordingly;
it appears, however, that there are no activity areas
below the low water mark. Following the model, the
features at the periphery of the midden should be
younger than those in the core of the midden even if
they are at the same elevation. This should be true if,
for no other reason, hundreds of years of deposition is
necessary on the periphery before it is dry land on
which an activity area could be placed.

This process can be demonstrated by depicting the
features as distance below surface instead of actual
elevation (Figure 7). This is the "up and out" pattern
discussed earlier. The interpretation of pottery style
changes and radiometric dates indicate that the
distance below surface on this section of the midden is
highly representative of the age of the level. This
diagram, then, represents the features by level
coevally. It is apparent that the older features are
more to the "core" of the midden. As the model
predicted, while the midden accumulated vertically and
horizontally, the features (activity areas) followed a
similar pattern.


The results from the Hontoon Island study support the
presence of the predicted configuration: refuse areas
on the periphery and activity areas clustering near the
center. The presence of this configuration in turn
increases the validity of the model of midden
formation. The processes which the model describes
were confirmed by Meehan using ethnographic
information, tested for and confirmed by May using
surface extensive information, and tested and

confirmed on the Hontoon Island midden using deep
intensive excavations.

Although the model accounts for the midden's general
configuration, there are some aspects concerning the
site for which the model offers no prediction. There is no
conclusive finding to indicate how many people were
living at Hontoon Island. This is an important point in
assessing the "intensity" aspect of the refuse
situation. Furthermore, the above confirmed
configuration could have been produced by several
types of settlements on top of the midden; e.g., (I) the
people could have lived on one half of the midden and
thrown their refuse on the other side, or (2) the people
could have been spread evenly across the midden.
Although the model does not specifically account for
this difference, from observation of the matrix changes
at Hontoon Island, I believe the latter to be the case. If
the former had been the cause of the configuration, one
would expect more consistent layers in the
stratigraphy (i.e., thick layers of refuse above and
below thick layers of feature areas).

The verified configuration also does not confirm or
refute the general settlement pattern. (By general
settlement pattern, I mean the year-by-year
movements of the population.) The statement that the
population was permanently settled at Hontoon Island
(Purdy and Newsom 1985) comes from faunal (Wing
1984) and floral (Newsom 1984) information. The fact
that the population was permanently settled is not
integral to the above model: it does not matter if the
described occurrences took place every month
throughout the year, or just annually when the
population returned for a season. The disposal strategy
remains the same.

The fact that the population was settled permanently
at Hontoon Island becomes important when discussing
the social dimensions of the refuse strategy. In
addition, the population's permanence makes the
Hontoon Island situation interesting from a refuse
point of view, i.e., permanent population utilizing
convenience and kitchen midden techniques. This is
when "intensity" comes into the picture. Melosi
(1981:3) states: "it is the modern industrial society,
not the ancient primitive society, that has experienced
the most intense refuse problem." Melosi also makes
the point that one of the major factors which helps
deal with refuse problems is that the problem must be
perceived by those affected by it (1981:3). If people
are acutely aware of their refuse production and
disposal strategy, the refuse problem will not be
intense. I examine this issue at greater length in
Nodine (1986) and discuss models used by Portnoy
(1981) and Oetelaar (1985). At Hontoon Island there
must have been social problems pertaining to refuse
disposal that were solved eventually by fixed rules.

The above discussion not only indicates how social
models can be used in conjunction with the
archaeological record, but helps clarify many of the
aspects of refuse disposal. In the Hontoon Island case,

people are permanently settled so their general
settlement pattern is established. They should, at
some point, begin to use a discrete village dump.
However, the attributes of the refuse itself along with
social variables intercede to dictate a disposal
practice different from the expected. The seemingly
simple process of throwing away refuse from an
evening meal becomes a complicated issue.

This case study approach to refuse disposal at Hontoon
Island yields information about the interaction of
refuse variables and settlement variables. On the level
of archaeological method and theory, this approach
brings to light how modeling for past processes can
help understand past behavior. And on the level of
shell midden archaeology, we gain another example of
how a shell midden is produced and interpreted.

References Cited
Gould, R A
S1980 ivi hac v. Cambridge University Press. Cambridge.
Hayden, Brian and Aubrey Cannon
198" Where the Garbage Goes Refuse Disposal In the Maya Highlands
Journal tAnlthropo Arae 2 117-163.
May, J Alan
1982 Midden Formation Modeling Using Ehnographic and Archaeological
Data: A Trend Surface Anahysis of Midden Deposits at the
Cadston Annis Sile (15BI5), Kentucky. Ph.D. dissertation.
Department of Anthropology, University of Missouri, Columbia.
University Microfilms Internalional, Ann Arbor.
Meehan, Betty
1982 Shell Bed to Shell Midden. Australian nslilute of Aboriginal
Stud.ie. Canberra.
Melosi. Main V (ed.)
1981 Garbage In the Cities: Refuse Relorm and the Envnmnt
1880-1980 Texas A&M Universlty Press. College Staion
Murray, Priscilla
1980 Discard Locailon, The Ethnographic Data. Amercan Antiquily
45 490-502
Newsom. Lee Ann
1984 Archaeological Plant Remains from Honloon Island Paper
presented in the symposium. Hontoon Island, an Archaeological
Welsile. Soulheaslern Archaeological Conference. Pensacola.
Nodne. Bruce K.
1985 Society and the universal Oroblem of refuse disposal an
anthropological analysis M A thesis, Department of Anthropology,
University of Florida. Gainesville
Oerelaar, G.
1985 Settlement Plans, Environmenta Constraints, and Refuse
Patterns. Paper presented at the 50lh Annual Meeting of the
Society for American Archaeology, Denver.
Ponnoy. Alice W
1981 A Microarchaeologcal View of Human Settlement Space
and Function. In Modern Maleral Cullure R.A. Gould and M B.
Schlffer (eds ), p. 213-224 Academic Press. New York.
Purdy, Barbara A and Lee Ann Newsom
1985 Significance of Archaeological Wetstes: a Florida Example.
National Geographic Research 1(4) 564-569.
Rick, John W
1976 Downslope Movement and Archaeological Intrasile Spatial
Analysis. American Antquity 41.133-144.
Rowlet, Ralph M. and Michael C. Robbins
1982 Estimating Onginal Assemblage Content to Adjust the
Postdepositional Vertical Artifact Movement World Archaeology
Schifer, M chael B
1972 Archaeological Context and Systemic Context. Amencan
Aniquity 37 156-165
1976 Behavioral Archaeology Academic Press, New York
Walker. S.T.
180_ Report on the Shell Heaps c i T ., i-i Annual
Report of the Smithsonian ,i: i 1-422
1883 The Aborigines of Florida Annual Report of the Smithsonian
Institution. 1881. pp 677-680
1885 Mounds and Shel Heaps on the West Coast f Florida Aua
Report of the Smithsonian Instilulion, 1883. pp 854-868
Waselov. Gregory
1982 Shellfish Gathering and Shell Midden Archaeology Ph D.
dissertation. University of North Carolina. Chapel Hill
Wing ES
1984 Faunal Remains from Hontoon Island In Symposium on
Hontoon Island, an Archaeological Wetsite Southeastern
Archaeological Conference. Pensacola.
Bruce K. Nodine
Department of Anthropology
Univeristy of Florida
Gainesville, Florida 32611


Barbara A. Purdy
Artifact Categories


The typical and atypical artifacts recovered during
excavations at site 8-VO-202 on Hontoon Island,
Florida and changes that occurred through time in the
artifact assemblage, are described and discussed
below. From all excavated units, totalling
approximately 132 m- (McGee and Nodine, this issue),
we recovered a total of 87 marine shell tools, 119
pieces of stone (most of which were utilized), 17 shark
teeth, 113 bone tools plus hundreds of bone fragments
that could have been utilized, 48 wooden artifacts
plus thousands of adzed wood chips and naturally
deposited wood debitage, 37 historic period artifacts,
34 pieces of red or yellow ochre, and more than 70,000
pottery sherds. The overwhelming majority of the
pottery, >99%, is St. Johns paste. The decoration
consists of predominantly St. Johns Check Stamped in
the upper levels. Plain sherds occur in various
proportions throughout but are dominant in the lower
levels (St. Johns 1) (Goggin 1952; Milanich and
Fairbanks 1980).

Bone artifacts were made primarily from deer and
varied considerably in size. Some of the bone artifacts
were very delicate while others were sturdier and
might have functioned as weapons. We also recovered
two bone beads and three bone pendants, two of which
were manufactured from deer scapula. Shell
implements, made from marine clams or conchs,
included scrapers, a drinking cup, and a variety of other
tools and utensils. Wooden artifacts consisted of a
fire starter, an atlatl or notched stick, a wedge, a
number of burned "posts," small pointed objects
perhaps pegs or projectiles, and other objects
unidentifiable as to function. The most common stone
point was Pinellas which fits in well with the pottery
chronology and the radiocarbon dates received for the
site. A Duval point was recovered in Level 10 of one
2-m section of the 1984 trench which verifies the
increasing antiquity with depth. Duval should predate
Pinellas according to already established
classifications (Bullen 1975:13). We found a stone
plummet or net weight in a permanently saturated
level of the trench. It still retains pitch or resin
permitting an examination of the kind of cordage that
bound it at the neck. The plummet, a metate made of
coquina, a mano, and several broken or fragmentary
stone celts were all manufactured of nonlocal stone.
There were numerous pieces of sandstone recovered,
many of which had grooves or otherwise showed signs
of use. Two pieces of sandstone may have been
"sculpted" into art forms although their shapes may be

A simplified "lumper" approach to the classification of
the artifacts excavated at Hontoon Island is presented
in Table I. All numerical and other data pertaining to
each artifact has been compiled at the University of
Florida and is available for inspection and study by
interested scholars.

In addition to the 113 bone artifacts listed in Table I,
there were tens of thousands of unmodified or
splintered food bones recovered at Hontoon Island.
Much of this discarded scrap, especially the fractured
bone, may have served briefly as convenience tools
leaving little or no evidence of utilization. Thus, the
bone artifact count would be increased measurably if
these fragments, lacking formal shapes and sustained
usage, had been detected. The bone artifacts have not
been studied minutely for methods of manufacture but
a few comments can be made.

The bone artifacts classified as pins were highly
polished, symmetrical, and often had some carving or
incising at the proximal end. The utilitarian objects,
on the other hand, were dull and asymmetrical with no
decoration. The five ulna awls are a good example.
Polish on the decorative objects could have been
achieved by rubbing with sandstone, shark skin, or
through use. Some of the pins are so highly polished
that no other manufacturing marks are still visible.
Most of the objects, however, retain some striations
possibly made with a shark tooth, a flint flake, or a
sandstone abrader.

Two methods appear to have been used at Hontoon
Island to manufacture bone pins, and both were utilized
throughout the time represented in our excavations,
that is, from about A.D. 0 to A.D. 1760. In one technique
the bone was split lengthwise, then the outer surfaces
were tapered and polished but the channel was not
removed. These pins are very uniform in size and differ
primarily from one another in that one has a hole
drilled in the middle of the top of the head (possibly to
attach something such as a feather) and another is
slightly incised (Figure I). One of the pins in this
category is fossilized and might have been an old
artifact that was found and brought to the site. These
pins are highly polished but the polish on some of them
has eroded partially leaving blotchy dull areas. Upon
examination of these artifacts, it is very apparent that
this situation did not result from differential
polishing. These are called expanding head pins in the
literature (Goggin 1952).

Volume 40 Number 1


March, 1987


Stone 12 3 4 5 6 7 8 9 10 11 12 13 14 15 No

Point' I 13 13 7 2 1 1 1 2
Utilized flake 4 6 6 3 4 I I
Waste flake II
Tool 1 1 2
Heat exploded 2
Sandstone abrder3 1 6 2 1 1 2
Celt frags & hammers2 3 3 1 1
Coqulna metate I
Limestone plummet2
Ochre 3 10 11 6 2 1

Bone 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 No

Pins 4 11 7 8 8 1 3 5 4 1
Beads 2
Pendants 3
Utilitarian 3 10 6 6 6 3 2 3 I
Exotic/fossil I 2 I
Miscellaneous 1 2 2 2 I
Shark teeth
Drilled I 1 1 1
No alteration detected 5 3 1 2 1

Shell 1 2 3 4 5 5 7 8 9 10 11 12 13 14 15 No

Marine clam
Scraper4 3 3 2 1
Unknown function4 I 4 1 1
Unused5 2
Adze 3 1 4 1 2 2 2
Hammer I
Gouge 2 1 2 3
Unknown (unction 2 2 1 5 3 1 1 2
Fragment 1 2
Oyster (unused) I
Freshwater mussel
Species & function not
known 1 1 4 2

Historic 1 2 3 4 5 6 7 8 9 10 1 12 13 14 15 No

Olive jar
Glass beod

Wood 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 No

Cypress ( Txodium sp )
Firestarter I
Plank fragment
Impressed design
Point 1 1 2
Post 1 1
Unknown function I 2 1
Pine(ELnu sp )
Lighterknot 3
AtlatI (9)
Post I 2
Willow (Sali dcaolinrane)
Pe. cut
Ash (Frax tus sp )
Bay (Eecrmsp )
Staggerbush (LVona sp )
Plug (0)
Function unknown
Oak (Quercus sp.)
Bowl preform
Elm (Ulmus americna)
Button bush (Cephalanthus
Cedar (Julperus sp )
Canoe paddle
Persimmon (Diospyrs
Wedge 1
Unidentified hardwood with
incised lines
Species & function unknown 4 1 2 3

I Two points made of silicifiedcoral were found, one in level 3 and the other in level 4 Both waste
flakes were silicified coral
2Non-Florida stone Ounflint listed alsowith historic artifacts.
3The sandstone abrader found in level 7 is very smooth and the shape resembles an animal
Some are DomSni dicus
5 Includes very small whole marine shells like Cerithlum. abuneum that are sometimes found with a
drill hole at sites in Florida

S 2
8 13

5 1

Table 1. Nonceramic Artifacts from Hontoon Island

The other method was to cut out a section of the long
bone and then shape it until it was perfectly round in
mid-cross section with no channel remaining. As an
alternative, after the bone was split, the channel may
have been removed by extensive sanding or by some
other method. We have only one artifact that shows
how the long bone was cut to remove a portion of it for
a pin. The proximal ends of these pins tended to have
more incising than the first type and often they have
peg tops of various shapes (Figure 2) probably for the
attachment of decorations. Some of the incising could
have been functional to provide purchase for a line tied
around the circumference.

A third method, of which there is only one example,
was used to make a "dagger," which was classified as a
utilitarian object (Figure 3). Here the leg bone has not
been split but has been modified into a very effective

A common technique of manufacture was observed on
many utilitarian bone artifacts, especially those that
probably functioned as awls, fid, perforators, or
gravers. Durable but very sharp distal ends were
formed by cutting at a dihedral angle (Figure 4). The
result is nearly identical to what has been described
for stone tools shaped by directing a burin blow to
either side of the tip. This same technique was used
also in shaping the pointed ends of many of the bone
pins but evidence of it was largely obliterated by
subsequent finishing.

The ends of some small bone artifacts are flat and
rounded like little spatulas. One of the bone pins has
uniform rotary marks encircling the proximal end for a
distance of several millimeters.

Bone artifacts range in weight from 0.20 g to 36.65 g,
in length from 1.30 cm to 27.10 cm (many are broken),
in width from 0.49 cm to 6.50 cm, and in thickness or
diameter from 0.18 to 4.92 cm. The primary species
was deer including deer scapula which was used for
two of the pendants. Bird bone was used for a bead and
for one of the elaborately incised objects. The leg bone
of a crane, Grus canadensis or, perhaps, whooping
crane, Grus americanus was used to manufacture the
artifact shown in Figure 5, and one of the pins was
made of alligator bone.

Fifty-nine bone artifacts are listed as decorative in
Table I. Thus, the decorative items, unexpectedly,
exceed slightly the utilitarian category making even
more convincing the suggestion that unmodified
fractured bone scrap might have served as utilitarian
objects. The bone pins were probably used as hair and
clothing ornaments. There is, of course, a value
judgment involved in classifying objects as bone pins
since some of these might have served other functions
and might, in fact, have been manufactured for
utilitarian purposes. But I believe the same principle
can be applied to bone as has been mentioned
previously (Purdy 1981) for stone implements: if the
finished product is the objective, e.g., a bone pin or a
stone spearhead, then greater care and precision will

be exerted in its manufacture. If, however, the bone or
stone object is a tool to make a tool, e.g., an awl or a
scraper, then the main concern of its manufacture will
be its ability to perform a specified function.

Bone artifacts were recovered from every excavation
unit and to a depth greater than 3 m. As shown in
Table I, significantly larger numbers of bone artifacts
occur in the upper levels. This situation might be
explained by an increase in population, a change in
activity, or both. In the last section of this paper, I
explore these possibilities further.


Forty-one of the 81 chert artifacts found at Hontoon
Island were points, and 33 of the remaining 40 chert
artifacts were utilized flakes or tools. This situation
suggests that (I) chert tools were not being
manufactured at the site, (2) chert was not readily
available, and (3) materials other than chert, such as
bone or marine shell, performed tasks that usually
required chert. A far-fetched additional explanation is
that few tasks were performed at the site that
required sharp, durable materials. That chert was hard
to come by is demonstrated by the fact that in nearly
all levels, some stone had been reworked because old
weathered surface had been partially removed by
reflaking. Of the remaining 60 pieces of stone, ochre
and sandstone represent the largest categories,
totalling 48. The only metate recovered from the site
was made of coquina.

The greenstone celt fragments, the coquina metate, the
plummet made of non-Florida limestone, as well as the
chert are interesting because they suggest that at
least occasionally Hontoon Islanders had some contact
with outsiders. From the distribution of the stone in
the deposit, it is apparent that this contact was
accelerated after historic contact (see Table I and
Discussion section).

Thirty-four of the 41 stone points were recovered from
the historic levels of the site. Twelve of the 41 chert
points and several stone tools belong to the
Preceramic Archaic period and thus they are out of
place chronologically. It is interesting that all of the
Archaic Period spearheads and two of the tools were
recovered in the Historic Period levels. This situation
suggests that (I) the Indians may have uncovered old
chert points when they cleared land for the newly
introduced idea of cultivating crops, or that (2) the
source of chert was cut off as a result of changing
relationships after historic contact and this forced the
Indians to become relic hunters in order to increase the
number of chert points needed for defense or for
hunting (see Discussion section). One observation that
verifies that these are really old points and not
reproductions is the fact that none of the Pinellas
points, believed to be in place chronologically (Bullen
1975 ; Purdy 1981), are as weathered as the Archaic
Period points.

Twenty-seven of the remaining points are Pinellas

(some nicely made, some crude). A Duval point was
recovered in Level 10 of one 2-m section of the 1984
trench and is in correct chronological position based on
radiocarbon analysis and on ceramic styles. The other
point is an extremely small, unclassified, stemmed

There is quite a variety in the color and texture of the
points recovered. This may suggest that the Indians
were obtaining the stone from several locations.
Although this statement should be accepted with
caution because of the great diversity that may occur
even within a single quarry source, it is somewhat
substantiated by the fact that four artifacts of
silicified coral were recovered at the site, two from
historic and two from prehistoric levels. Outcrops of
silicified coral occur in the Tampa Bay area and the
Suwannee River as well as a few additional locations.
Other kinds of chert also occur in these areas and in
numerous quarries in the central highlands.

The stone artifacts range in weight from 0.35 g to
830.00 g, in length from 1.20 cm to 15.02 cm, in width
from 1.00 cm to 12.00 cm, and in thickness or diameter
from 0.12 cm to 4.25 cm.


Tools and ornaments were produced from various
species of marine clams and conchs including Dosinia
discus, Mercenaria mercenaria, Dinocardium robustum,
Chione cancellata, Busycon sp. and Melongena sp. Nearly
three times as many tools were made from conchs than
were produced from marine clams. Two oyster shells
(Oystrea sp.) and one Murex fulvescens were recovered
also. These probably were fossils although the Murex
had a ragged hole in it that might have been manmade.
A shell hook (Figure 6) was made of Elliptio sp., the
freshwater clam. It was the only shell artifact made
from a freshwater species recovered at Hontoon Island
during our excavations.

The technology involved in producing implements and
ornaments from marine shell species is not understood
thoroughly (Prentice, personal communication 1986).
According to Prentice, in areas where large numbers of
specimens were available, the Indians probably found it
expedient to bash the conchs in order to fracture them.
If the columella was to be used as a tool or a gorget,
the outer portion of the shell would be hit until the
column could be removed easily. If the wall of the
conch was to be used, the desired sized piece of the
fractured shell would be chosen and then ground to a
working edge.

Prentice conducted an experiment to determine if it
was possible to cut a knobbed whelk (Busycon carica
[Gmelin]) with flint tools in order to manufacture a
shell cup. After 386 minutes (nearly 6-1/2 hours),
exhausting 60 flakes in the process, Prentice was able
to detach a large triangular section from the body of
the shell (Prentice 1983). While this method is
considerably more time, energy, and flint consuming

than bashing, it is more accurate, aesthetically
pleasing, and conserves the shell. In examining the
shell tools recovered at Hontoon Island, it appears that
both bashing and cutting methods were used. I have
concluded, however, that little or no local production
of shell implements occurred at Hontoon Island, at
least in the areas excavated, because large amounts of
shell debitage should have been found if the first
technique were utilized and one would expect
quantities of flint debitage if the second technique
were utilized.

Prentice also conducted experiments on the same whelk
to determine the length of time and amount of flint
needed to remove the columnella and to cut small
sections in order to produce beads. "After 1391 minutes
(slightly more than 23 hours) of cutting time and the
use of 191 flakes, the Busycon had been reduced to one
cup, three columnella pieces, and 17 bead blanks"
(Prentice 1983). He then proceeded to drill holes in the
bead blanks with a flint microdrill and discovered that
heat-altered blades produced from Florida chert were
superior to unheated blades for drilling. After the
holes are drilled, the blanks are strung on a line and run
through a grooved sandstone abrader in order to make
them round and uniform in size (Prentice 1983;
personal communication 1986). For other useful and
comprehensive descriptions of the manufacture and use
of shell tools, see Holmes (1883), and Phillips and
Brown (1978).

The functions of the shell tools have been broadly
assigned to adze, scraper, or gouge categories (Table
I). One shell cup, made from Busycon sp., was
recovered. No celts or hoes were found. Many of the
implements placed in the unknown category were
fragments and, thus, their function could not be
determined. It is very probable that some of the shell
tools were multipurpose. The major problem, of
course, is to determine the actual use of an implement.
It is one thing to note that a gouging, scraping, or
adzing motion was involved and quite another to
specify what material was being worked or what kinds
of objects were being produced. We are virtually
certain that wood was processed with shell tools and
we have even been able to match up the marks on some
of the wooden artifacts with the working end of the
shell implements. Comparative studies and replicative
experiments, similar to those conducted by Prentice,
desperately need to be carried out in Florida to solve
the problem of use wear on many classes of artifacts.

In contrast to the abundant decorative bone artifacts,
only three ornaments were made of shell: a bead and a
pendant were manufactured from marine species and
the shell hook from Elliptio sp. No shell gorgets were
Significantly larger percentages of shell tools were
found in the squares excavated near the lagoon than in
more terrestrial units. This situation suggests that
activities, like woodworking, occurred on the periphery
of the site. As an alternative, it could also indicate


Figure I. Bone Pin of the Expanding Head Variety Recovered at the Hontoon Island
Site (8-VO-202), Florida.

Figure 2. Bone Pin of the Pegged Top Variety Recovered at the Hontoon Island Site
(8-VO-202), Florida (specimen is approximately 20 cm long).


Figure 3. Bone Dagger (?) Recovered at the Hontoon Island Site (8-VO-202),

Figure 4. Method of Shaping Distal End of Bone to Form a Cutting or Graving Tool.
Many specimens of this type were recovered at the Hontoon Island Site (8-VO-202),

Figure 5. Bone Artifact Manufactured of Crane (Grus sp.) Recovered at the Hontoon
Island Site (8-VO-202), Florida.

Figure 6. Shell Artifact Manufactured of

Freshwater Clam (Elliptio sp.) Recovered
at the Hontoon Island Site (8-VO-202),
Florida. It is uncertain whether this
artifact was intended to be decorative or











Figure 7. Broken Canoe Paddle Manufactured from Red Cedar (Juniperus sp.)
Recovered at the Hontoon Island Site (8-VO-202), Florida. Note adzing and shark
tooth finishing marks.

Figure 8. Illustration of Gold Ornament Recovered from the Historic Zone at the
Hontoon Island Site (8-VO-202), Florida.

Figure 9. An Early 16th Century Copper
Maravedi Coin Recovered from the
Historic Zone at the Hontoon Island Site
(8-VO-202), Florida.

Figure 10. Limestone Plummet Manufactured of Non-Florida Stone
Recovered at the Hontoon Island Site (8-VO-202), Florida. Note the
cordage and resin still evident in the illustration at the right.


Figure 11. Bone Pin and Closeup (Right) Recovered from the Historic Zone at the
Hontoon Island Site (8-VO-202), Florida. This type of decoration did not occur on
bone artifacts recovered from prehistoric levels.

, I


that broken or exhausted shell implements were tossed
toward the lagoon rather than being discarded onto the
main habitation area (see Nodine 1986).
Shell artifacts range in weight from .35 g to 347.06 g,
in length from 1.80 cm to 18.10 cm, in width from 1.30
to 13.85 cm, and in thickness or circumference from
0.20 to 8.90 cm.


Along with the other botanical remains from Hontoon
Island, the wooden artifacts, debitage, and ecofacts
supply information about environment and culture that
is seldom preserved. Large carvings like the owl, found
in 1955 (Bullen 1955), and the pelican and otter, found
in 1977, were not recovered during our excavations
since 1980 but the 48 wooden artifacts listed in Table I
and thousands of pieces of debitage provide clues to
the wood species available and selected by the
aboriginal inhabitants.

Cypress and pine appear to be the predominant woods
utilized but many other species were used also
suggesting that, in some cases or for some
woodworking projects, availability of a species might
have been just as important as wood properties. One
example pertains to the artifacts classified as posts.
These were made of cypress, pine, elm, ash, and bay.
Long, straight pieces about 20 cm or more in
circumference were selected and were sharpened on
one end by using a shell or stone adze. Fire may have
been used to facilitate the shaping process. Only
portions of posts driven into the ground were
recovered, because the tops had been burned off or
eroded away.

Tool marks and manufacturing techniques can be
identified on many of the wooden artifacts. Future
experiments are expected to specify the types of
implements and processes utilized. Shark teeth marks,
for example, can be seen on the canoe paddle made of
red cedar (Figure 7), stone or shell adze marks can be
seen also on this specimen as well as on many others,
and a shaft or pole of pine exhibits cross grain cutting.
Most of the wooden artifacts have some charring that
indicates the use of fire either in the manufacturing
process or through use. Charring is present on many of
the thousands of flat debitage pieces and may have
occurred from utilizing a fire hollowing technique in
the production of canoes. Differential burned areas on
portions of a single piece of wood contribute
significantly to problems of preservation.

Determination of wood species and studies of wood
properties, technology, and artifacts are in progress at
the University of Florida. The species listed in Table I
were identified by Lee Ann Newsom.

The wooden artifacts range in weight from 1.30 g to
491.10 g, in length from 3.30 cm to 48.90 cm, in width
from .75 cm to 23.70 cm, and in thickness or
circumference from .60 cm to 9.30 cm.

Historic Artifacts

The position of the European artifacts (Figure 8) in the
top three levels of the deposit (Table \) suppOd5 our
belief that the stratigraphy in the location of our
excavation units at Hontoon Island has not been
disturbed by activities occurring in the twentieth
century. The three historic period artifacts that were
recovered from Level 4 were from the most lagoonward
excavated squares (39 and 59) where greater organic
deposition was present. It is important to point out
that all but two of the 37 historic period artifacts
were recovered from squares fringing the lagoon where
less trampling and erosion occurred after objects were
discarded or lost.

The copper coin (Figure 9), majolica and olive jar
sherds, probably the glass bead, and one iron artifact,
an estradiot hilt, support our conclusion based on other
evidence that the historic artifacts were introduced
fairly early in the contact period. The other evidence
consists of limited radiocarbon and
thermoluminescence analyses, stratigraphic position,
and the fact that no European animal bones or plants
were recovered. Another bit of evidence that supports
the conclusion that the European artifacts were
introduced at Hontoon Island early in the contact period
is Rouse's (1951:216) statement that "only in the earlier
periods ... did the Spaniards have any real difficulty
with looting" of shipwrecks because during the later
periods, "the Spaniards had only to send one or two
soldiers from their outposts further north to recover
the loot. ." Yet, contrary to the conclusion reached by
Rouse (1951), Goggin (1952) and others, that "the coming
of the Spaniards caused no appreciable changes in the
Indian cultures of the area" (Rouse 1951:257), the
extreme changes that took place at Hontoon Island
suggest that the European artifacts represent more
than the addition of a few exotic baubles to the

I am grateful to Jonathon Leader for his identification,
description, and classification of the historic period
artifacts from Hontoon Island. Leader (1985,1986)
presents a comprehensive discussion of metal artifacts
found in Florida.


From the 1984 Trench (Squares 24-36, Table 2), a total
of 66,164 pottery sherds was recovered. They weigh
397,237 g, averaging 6 g per sherd. In 1980, 3153
sherds were recovered from the 3-meter test pit and
143 sherds were recovered from the 1982 column
sample whose dimensions were 30 x 30 x 140 cm. The
sherds from the 2-meter test pit excavated in 1982
were not counted and the sherds from the 1984 column
samples, trench extension, and the 1985 excavations
have not been counted yet. Greater than 99% of the
pottery is St. Johns Plain and Check Stamped. An
analysis of the pottery from Hontoon Island will be the
subject of a master's thesis (McGee, in preparation).

Square 241
Level No. Wt. Wt.
(9) (g)

1 87 662 7.61
2 118 1509 12.79
3 93 1132 12.17
4 72 986 13.69
5 386 1767 4.58
6 332 3997 12.03
7 389 2910 7.48
8 430 3357 7.80
9 540 2529 4.68

Square 25
No. Wt. Wt.
(g) (g)


Square 26
No. Wt. Wt.
(g) (g)




Square 29
No. Wt. Wt.
(g) (g)

467 1241 2.66
1207 5088 4.22
708 3977 5.62
358 2032 5.68
445 2512 5.64
278 1208 4.35
241 1126 4.67

Square 30 Square 31
Avg. Avg.
No. Wt. Wt. No. Wt. Wt.
(g) (g) (g) (g)

1113 4038 3.63
1541 7818 5.07
596 3664 6.15
804 4252 5.29
320 1601 5.00
238 2026 8.51

1831 6976 3.81
951 4777 5.02
655 3126 4.77
887 4082 4.60
709 4423 6.28
321 1756 5.47
134 1441 10.75

Total 2445 18849 7.70 6932 37764 5.45

Square 32
Level No. Wt. Wt.
(g) (g)

I 115 576 5.00
2 1011 4452 4.40
3 1643 8759 5.33
4 595 4040 6.79
5 473 4104 8.68
6 215 2414 11.23

Total 4052 24345 6.01

Square 33
No. Wt. Wt.
(g) (g)

17 65 3.82
485 4222 8.71
778 6629 8.52
923 4714 5.11
797 4678 5.87
704 3760 5.34
766 4357 5.69
1333 12044 9.03
670 6127 9.14
488 7320 15.00
488 7320 15.00
62 597 9.63

7449 61833 8.30

5542 31639 5.70

Square 34
No. Wt. Wt.
(g) (g)



3704 17184 4.64

Square 35
No. Wt. Wt.
(g) (g)

148 495 3.34
4974 18152 3.65
2240 10509 4.69
1117 7968 7.13
851 6191 7.27
357 2424 6.79
350 5697 16.28
697 14408 20.67

4612 23399 5.07 5488 26581 4.84

Square 36
No. Wt. Wt.
(g) (g)

2 4 2.00
466 2549 5.47
5010 20478 4.09
1572 6572 4.18

9592 66486 6.93 10734 65844 6.13 5635 23688 4.20

Grand Totals:
No. 66164
wt. 397237 g
Avg. Wt. 6.00 g

1Square 24 was disturbed because of shell mining in the 1930s; therefore
figures do not reflect the true situation in that area.

2Level 4 of Square 36 was only partially excavated.

Table 2.--Ceramic remains from the 1984 trench at Hontoon Island

The raw data distribution of the pottery is revealing
and supports our inference that through time the mound
grew toward the lagoon and actually increased the size
of the island. If one compares, for example, the number
of check stamped versus plain St. Johns sherds from
terrestrial square 26 of the 1984 trench (Williams
1984) with those of the 1980 test pit excavated near
the lagoon (Newsom 1981), it is apparent that the
cultural material from the bottom of the 1980 test pit
is comparable to only levels 3 to 4 of the more
terrestrial portion of the 1984 trench (Table 3). It also
demonstrates that the terrestrial or living portion of
the mound was eroded from trampling and other
activities and from the compaction that occurred when
the botanical materials decomposed.
One might conclude that the number of sherds versus
sherd weight may give an indication of the amount of
activity that has occurred on a site after a broken pot
has been discarded. All other factors being equal, for
example, if the average weight of sherds decreases or
increases, it may be correct to assume that there was
more or less activity on the site following deposition
but prior to burial of a living floor (see Table 2, square
25, level 4 and square 26, levels 6 and 7). It is
interesting to note that in all of the lagoonward
squares, beginning at approximately square 31, there is
often an increase in number of sherds and a significant
increase in average weight of sherds with depth. This
situation probably exists for two reasons: broken pots
may have been thrown toward the lagoon (Nodine 1986)

1980 Test Pit(3 sq. m)

Level Plain


% Check Stamped %



and once there, they were not stepped on and reduced to
smaller sizes. Another explanation for the large
increase in average weight per sherd with depth may be
related to the thicker walls of the vessels in the
earlier time period represented at the site (i.e., St.
Johns I). However, this does not seem to be borne out
by the average weight of sherds from the lower levels
of squares 25 and 26 which are correspondingly older
than the lagoonward squares. Also, even though all of
the ceramics were washed and dried before they were
counted and weighed, it is possible that the sherds
from the saturated levels of the trench remained
somewhat waterlogged and this may account partially
for the weight differential (Newsom, personal
communication, 1986).
In an effort to understand the meaning of 66,164 sherds
weighing 397,237 g I weighed and measured an Indian
pottery bowl from the collections in the Department of
Anthropology, University of Florida. The bowl weighs
1110 g, is 27 cm (approximately II inches) in diameter,
12 cm (approximately 5 inches) high, and has a wall
thickness of .6 cm which falls within the thickness
range for the Hontoon Island sherds (McGee, n.d.). If
this were an average sized vessel being produced at
Hontoon Island, then sherds from 358 vessels were
recovered from our excavations of the 1984 trench. A
total of ninety-three, 25-cm levels in eleven 2 x 2-m
squares was excavated (93 m3) in the area of the 1984
trench where the sherds were recovered. This averages
8.5 levels per square and represents at least 1500

1984, Sq. 26(2 sq. m)



% Check Stamped


1Accumulation of decomposed organic vegetation since site abandonment.
20rganically stained sand becoming lighter with depth.
3Some of these may have been eroded check stamped that were categorized incorrectly because it
seems logical that a larger percentage of check stamped sherds should be present in the higher
levels compared to plain sherds.

Table 3. Comparison of numbers of plain and check stamped sherds from a
water-saturated versus a terrestrial unit at Hontoon Island.

years of occupation at the site suggesting that only
forty-two vessels were produced for each 175 years of
occupation. Admittedly, I have taken liberties with
these figures, but I believe this calculation suggests
that what originally seemed to be immense quantities
of ceramics, turns out to be a very low density for the
time and area involved. It is possible that broken pots
were used for something away from the site thus
reducing the amount of ceramic debitage. There are
ethnographic descriptions of such a practice (Moseley,
personal communication 1986).

To gain additional support for the accuracy of my
calculations pertaining to the number of vessels
represented by 66,164 sherds, Ann Cordell,
Archaeologist at the Florida State Museum, weighed
and measured ten chalky paste vessels for me from the
museum collections that were recovered from burial
sites. These vessels were reasonably complete except
for the presence of kill holes in most of them. I
figured the weight of the sherd missing from the kill
hole of each pot, added it to the weight of each vessel,
and obtained an average of 837 g for each pot. If this
were the average size for vessels produced at Hontoon
Island, then 475 vessels are represented. However,
from preliminary evaluations of rim sherds from
Hontoon Island, McGee (personal communication, 1986)
tends to believe that vessels, larger than those from
the museum collections, are represented at the site.
Still, even if the 475 number is correct, only fifty-six
vessels were recovered for each 175 years of
occupation at Hontoon Island.

In addition to St. Johns Plain and Check Stamped, a few
Dunns Creek Red, Little Manatee Zoned Stamped, and
fiber-tempered sherds were recovered at Hontoon
Island. There were also three sherds that had a design
on the inside, a number of vessel feet and handles, and
the shapes of at least two sherds exhibit European

Summary and Discussion

All in all the artifacts are quite ordinary and not very
abundant when one considers the enormous quantities
of flora and fauna present. The recovery of the wood is
unique but most of it represents byproducts of
manufacture, perhaps for canoes or structures for the
top of the midden that have long since disappeared.
Certainly, we did not find carvings like the owl, pelican
, and otter (Purdy, Overview and Chronology, this issue)
that were discovered near Hontoon Island in the past.
There are a number of objects, such as a Clearfork
Gouge, Preceramic Archaic points, and several pieces of
fossilized bone (including Pleistocene horse and tapir
teeth) that lead us to conclude that these people were
relic collectors because these objects are out of place
chronologically. There are, however, other possible
explanations for their presence.

Some materials were brought into the site because
they do not occur locally. There is, for example, an
occasional piece of pottery that is not typical of the
area. Flint rock does not outcrop along the St. Johns

River but is present within 25 miles of Hontoon Island.
The small size and evidence of use on most of the
chipped stone implements suggest that flint may not
have been easily obtainable. The material of the
coquina metate, the plummet (Figure 10) and celts were
not available locally but the distances they travelled to
get to the site varied considerably. The marine shell
also had to be brought in from the coast although we
have not ruled out the possibility that the Indians were
utilizing fossil shell beds that occur close by on the St.
Johns River. The shark teeth may also fall into this
category. It is unlikely that any of the imported
material was essential for the maintenance of the
group except, possibly, the marine shell which was
manufactured into implements to work wood and other

We have positive evidence of local manufacture of
wooden objects because of the quantities of debitage.
Bone and pottery industries probably were present also
at or near the site.

Except for the marine shell drinking cup and some large
pieces of wood, most of the artifacts are small. Many
artifacts are broken. We found more complete
specimens in the water-saturated squares. If people
were working along the edge of the water, they could
easily lose objects if they were dropped and settled
quickly into the organic deposit. There are many good
opportunities at Hontoon Island to construct theories
about site formation processes, and this situation
applies to the artifacts as well.

With very few exceptions, such as the nicely made and
decorated bone artifacts and a shell hook (Figure 6),
there is nothing spectacular to hint at the artistic
accomplishments we know existed prehistorically
along the St. Johns River. Perhaps all of the elaborate
carvings, other evidences of craftsmanship, and
expressions of the belief system were placed in graves
which we did not find at the site even though they were
mentioned by Wyman (1875:28).

Because we did not recover any European objects in the
1980 and 1982 squares or in the 1982 column sample,
we attempted to explain in other ways the abrupt
changes that occurred after A.D. 1500. The most
obvious change toward the top of the deposit is the
disappearance of freshwater shell. About 10 cm of
cultural deposit lies
above the shell and contains abundant remains of
pottery, bone, wood, and other items. The mystery is
why the Indians abandoned a major food resource,
freshwater snails and mussels in the early 1500s, but
continued to live at the site. We have tentatively ruled
out an invasion by another Indian group as an
explanation because the pottery style is identical in
the shell zone with that of the nonshell zone although
the shapes of some sherds in the nonshell zone suggest
an alien influence. There is a chance that a
catastrophe occurred that destroyed the shell beds for
we verified in the 1984 trench and again in 1985 an
observation made during the 1980 and 1982 excavations:

the uppermost portion of the shell deposit is
predominantly freshwater mussel (Elliptic sp.) that
seems to lie like a frosting over meters' thick
predominantly freshwater snail (Viviparus sp.) We
have no explanation yet for the change from snail to
mussel. It seems unlikely that either species was
affected by a climatic event because both can tolerate
a wide range of temperatures (Thompson, personal
communication 1984). The disappearance of the shell
may have occurred if the Indians simply did not desire
to eat it any more. This situation may have existed if
there had been a sudden emphasis on cultivated crops,
such as squash and corn. Other changes that took place
at this time in the flora and fauna are discussed in
other papers in this issue and in Purdy and Newsom

From the 1984 trench, we recovered two majolica ware
sherds, two pieces of silver, three pieces of copper,
and many iron objects. One of the copper items was a
Spanish maravedi coin (Figure 9) minted at the Santo
Domingo mint in the early to mid-1500s probably during
the reign of Carlos V and his mother Joanna (Deagan,
personal communication 1985; Lyon, personal
communication 1985). The blue on white majolica ware
also was manufactured during this period. The coin,
the majolica ware, the estradiot hilt, the absence of
any European plants or animals, plus supporting
radiocarbon and thermoluminescence dates verify that
the Spanish material predates St. Augustine and thus
may have been brought in as shipwreck material (Smith
and Gottlob 1978:12) traded inland from coastal Indians.
The changes we have noted, however, indicate that the
impact of the Europeans was greater than one would
expect from the introduction of a few exotic items.

One example pertains to the bone artifacts recovered at
Hontoon Island. These were undoubtedly used for many
purposes including awls, hairpins, and weapons. In the
prehistoric levels these are usually undecorated or
contain only a small amount of incising. We continued
to find bone artifacts in the historic zone but a
dramatic change occurs in decoration on some of the
bone pins and on a piece of deer scapula (Figure 11). It
is interesting to note also that all three of the bone
pendants were recovered in the historic levels. It is
possible that the new art style was introduced by
Indians from other areas as they were displaced by the
Europeans, or it may represent a traditional design that
could not be executed on bone until metal tools were
available. Perhaps, as an alternative, the new art style
was influenced by people from a different hemisphere
and it may have been accompanied by changes in
worldview and symbolism. The change in decoration on
the bone pins is the topic of another paper (Purdy, in

As mentioned above, we noted changes in the shape of
some pottery sherds, suggesting European influences,
and most of the unusual pottery was found in the
shell-less zone.

Another interesting observation is that there was an
increase in all artifact classes, except marine shell

tools, in the contact zone. Thirteen of the 17 shark
teeth and 34 of the 41 stone points, for example, came
from the upper horizon. These absolute numbers
become more significant when one considers that the
volume of material removed from the prehistoric
levels far exceeds that of the contact zone. The
increase in artifacts at the site during a time when one
would expect a decline in material culture resulting
from population decimation, may indicate that Hontoon
Island was a refuge area for displaced people because
of its favorable location along the St. Johns River. In
contrast only about 20 of the 87 marine shell
implements were found in the upper horizon suggesting,
as one possibility, that metal may have become

References Cited

Bullen, Ripley
1955 Carved Owl Totem, Deland, Florida. The Florida Anthropologist
1975 A Guide to the Identification of Florida Projectile Points.
Revised 'edition. Kendall Books, Gainesville.
Deagan, Kathleen
1985 Personal communication. Dr. Deagan is curator in archaeology
at the Florida State Museum. She is an historic archaeologist.
Goggin, John M.
1952 Space and Time Perspective in Northern St. Johns Archeology,
Florida. Yale University Publications in Anthropology, No. 47.
Holmes, William H.
1883 Art in Shell of the Ancient Americans. Bureau of American
Ethnology. 2nd Annual Report. 1880-1881.
Leader, Jonathan M.
1985 Metal Artifacts from Fort Center: Aboriginal Metal Working in
the Southeastern U.S. M.A. thesis, Department of Anthropology,
University of Florida.
1986 Metalwork Among the Florida Indians in the Early Contact Period.
Symposium on Spanish and Native American Encounters in Sixteenth
Century Florida. Archaeological Institute of America and the Institute
of Early Contact Studies. Florida Slate Museum, Gainesville.
Lyon, Eugene
1985 Personal communication. Dr. Lyon is an historian whose specialty
is the early contact period in Florida.
McGee, Ray M.
1986 Personal communication. Mr. McGee is in the master's program,
Department of Anthropology, University of Florida. He is
studying the pottery from Hontoon Island.
Milanich, J.T. and C.H. Fairbanks
1980 Florida Archaeology. Academic Press, New York.
Moseley, Michael
1986 Personal communication. Dr. Moseley is Professor of
Anthropology, Department of Anthropology, University of
Newsom, Lee Ann
1981 Preliminary Report on Ceramics from Hontoon Island. Manuscript
on file in Purdy's office, Department of Anthropology, University of
1986 Personal communication. Lee Ann Newsom is in the doctoral
program at the University of Florida. She has worked at the Hontoon
Island site since 1980 and has studied many of the materials: recovered,
particularly the botanical specimens.
Nodine, Bruce K.
1986 Society and the Universal Problem of Refuse Disposal: An
Anthropological Analysis. M.A. thesis, Department of Anthropology,
University of Florida.
Phillips, Philip and James A. Brown
1978 Pre-Columbian Shell Engravings from the Craig Mound at Spiro,
Oklahoma. Peabody Museum Press, Cambridge.

Prentice, Guy 39
1983 An Experiment in Shellworking. Manuscript prepared for Lithic
Technology. Ms. on file in Purdy's office, University of Florida.
1986 Personal communication. Mr. Prentice is in the doctoral program,
Department of Anthropology, University of Florida.
Purdy, Barbara A.
1981 Florida's Prehistoric Stone Technology. University Presses of
FlorfaT Gainesville.
n.d. Tracing a Design on Bone Pins. Manuscript in preparation.

Purdy, Barbara A. and Lee Ann Newsom
1985 Significance of Archaeological Wet Sites: A Florida Example.
National Geoaraphic Research 1(4):564-69.
Rouse, Irving
1951 A Survey of Indian River Archeology, Florida. Yale University
Publications in Anthropology, No. 44.
Schwehm, Alice Gates
1983 The Carved Wood Effigies of Fort Center: A Glimpse of South
Florida's Prehistoric Art. M.A. Thesis, Department of Fine Arts,
University of Florida, Gainesville.
Smith, Hale G. and Mark Gottlob
1978 Spanish-Indian Relationships: Synoptic History and Archeological
Evidence, 1500-1763, pp. 1-18. In: Tacachale. Edited by Jerald T.
Milanich and Samuel Proctor. University Presses of Florida,
Thompson, Fred
1984 Personal communication. Dr. Thompson is malacologist and
curator in Natural Sciences, Florida State Museum, University of
Williams, Joan
1984 Preliminary Ceramic Report on Square 26 from Hontoon Island,
Volusia County, Florida. Manuscript on file in B.A. Purdy's office,
Department of Anthropology, University of Florida.
Wyman, Jeffries
1875 Fresh-water Shell Mounds of the St. John's River, Florida.
Peabody Academy of Science, Salem, Massachusetts.

Barbara A. Purdy, Professor
Department of Anthropology
University of Florida
Gainesville, Florida 32611


Elizabeth S. Wing and Laurie McKean


Our knowledge of the past uses of plant and animal
resources is dependent upon the preservation of the
remains of these organic materials and their careful
excavation from archaeological sites. The Hontoon Island
site provides an extraordinarily enhanced view of a past
economy because of the remarkably good preservation of
organic remains from this site. The conditions that are
responsible for this excellent preservation are permanent
submergence in water. A portion of the accumulated
debris of the site is deposited in permanently saturated
peat and muck in the marsh and lagoon at the edge of the
site. Anaerobic conditions retard the natural processes of
deterioration which often eliminate or reduce the floral
and faunal component of an archaeological deposit.

The Hontoon Island site is not only exceptional for the
fine preservation of organic remains but also valuable as a
documentation of the changes in human uses of biological
resources. We have biological remains accumulated over
1800 years of occupation at this site. The time span of this
occupation extended from 0 A.D. to 1750 A.D. which
witnessed the changes to aboriginal life in Florida wrought
by the advent of Spanish exploration and settlement.

The Hontoon site is located on the northern end of
Hontoon Island which is midway along the St. Johns River.
The site is situated in mixed hardwood forest and at the
edge of a freshwater marsh. The interior of the small, 1650
acre, island is predominantly pine flatwoods.

Excavatations at Hontoon began in 1980 under the
direction of Dr. Barbara Purdy of the University of Florida.
This paper discusses the vertebrate animal remains
recovered from the first 3 m pit excavated at the site and
a portion of a 2 m pit excavated in 1985. All of these
materials were recovered using a 0.6 cm gauge screen.
Excavations in 1982, 1984, and 1985 included the removal
of a 30 cm by 30 cm or a 30 cm by 15 cm column sample in
conjunction with each excavated pit or trench section. The
results of the analyses of these excavation units are
reported elsewhere (Newsom 1986; Wing in press) and are
in part still under investigation.

Materials and Methods

Investigations at Hontoon required special
techniques to recover the material from submerged

sediments and to stabilize the organic remains once they
were exposed to the air. A description of the excavation
procedures can be found in the paper by Purdy and
Newsom (1985). The excavated bone deteriorates rapidly
once it is exposed to the air even when it is allowed to dry
slowly. Therefore, all bone was soaked in a dilute solution
of Elmers glue long enough for the glue to penetrate the
bone. It was subsequently air dried, sorted to class, and
identified to the lowest taxon possible.

All identifications were made by comparison to series of
reference specimens. Standard methods of quantification
were used. These include counts of identified specimens,
estimates of minimum numbers of individuals, and weights
of identified remains.

Identifications were made in the Zooarcheology
Department of the Florida State Museum in 1981 by
Laurie McKean and in 1986 by Tamara Markowitz and
Elizabeth Wing. The sample identified by McKean is from
the pit excavated in 1980, while sample numbers 135 and
145, both from the 1985 pit number 59, were identified by
Markowitz and Wing respectively. The two samples (135
and 145) from the 1985 pit differ very little from one
another and are therefore combined for this analysis. A
total of 7753 identified fragments representing 38 species
forms the basis of this analysis. This represents over 31
kilograms (glue impregnated dry weight) of vertebrate
remains. Details of the faunal composition and
quantification of the finds are found in Tables 6 through 8.
These data appear in summary form in Tables 1 through 5.

Five zones were defined from the first excavation at
Hontoon Island:

Zone 1 (0-60 cm) is culturally sterile.

Zone 2 (60-75 cm) consists of pottery of St. Johns II
period, adzed wood, and bone.

Zone 3 (75-85 cm) consists primarily of the freshwater
mussel (Elliptio), abundant bone, pottery, and
adzed wood.

Zone 4 (90-120 cm) has a shell matrix consisting of diverse
species but dominated by the aquatic snail
Viviparus georgianus. A few individuals of the
apple snail (Pomacea paludosa), a spiral snail
(Ganiobasisfloridensis), freshwater mussel
(Elliptio spp.), marine quahog (Mercenaria

Volume 40 Number


March, 1987


campechiensis), marine clam (Chione cancellata),
marine cockle (Dinocardia robusta), and marine
venerid clam (Macrocalista sp.) are also present.
All the marine shells showed some sign of having
been altered.

Zone 5 (120 cm unknown depth) was not completed
because of limited time. At the completion of
excavation a reduction in the amount of shell was
noted as well as the replacement of the fiberous
peat by organically stained sand.

Zones 2 and 3 became known as the Elliptio or historic
levels and Zones 4 and 5 are called the Viviparus or
prehistoric levels (Purdy and Newsom 1985). The 1980 pit
includes both prehistoric and historic components whereas
the 1985 pit includes only historic remains. The historic
sample is slightly more than three times the size of the
prehistoric sample in this analysis.

Studies of the biological remains from column samples
have shown how important the remains of small organisms,
lost in the 0.6 cm screening procedure, are to a more
complete understanding of the exploited fauna. We realize
a number of small species and smaller individuals of
species represented by larger specimens are absent from
the samples analyzed here. Despite this loss of information
from the samples, they can still provide insight into the
past uses of animals.

Results and Discussion

Faunal Composition

Eleven species predominate in these samples. They are the
two freshwater molluscs, mussel (Elliptio buckleyi) and
snail (Viviparus georgianus), and five freshwater
vertebrates, gar (Lepisosteus sp.), catfish (Ictalurus sp.),
bass (Micropterus salmoides), mullet (Mugil sp.), and slider
turtle (Pseudemys spp.). Four primarily terrestrial
vertebrates are also prominent in the faunas and these
include gopher tortoise (Gopheruspolyphemus), ducks
(Anatidae), rabbit (Sylvilagus sp.), and deer (Odocoileus
virginianus). Twenty-nine other vertebrate species were
identified but are considered relatively less important

Although the remains of the two dominant molluscs were
not quantified in these samples, they were extremely
abundant. They, in fact, appeared to form the matrix of
the cultural deposit and as mentioned before they are
associated with the two cultural horizons at the site, the
mussel with the historic levels and the snail with the
prehistoric levels. Little light is shed on the reasons for
this dissimilarity of the two horizons based on the ecology
of these two dominant species. Both Viviparus and Elliptio
are found in sandy bottoms of shallow parts of rivers and
lakes and are less frequently seen in the muddier

environments (Pennak 1978:723, 744-748). The Viviparus
snail can be collected by hand in areas where colonies exist.
Colonies may be somewhat separated from one another
and over exploitation might cause at least temporary
extirpation of a colony. Due to the prolific nature of the
snail, however, it is unlikely that populations would
undergo any permanent decline as a result of human
predation (Cumbaa 1976). The mussel was probably also
an easy prey though they have to be dug up from the sand.

The vertebrates identified from the Hontoon samples
reflect a fauna similar to that which exists around the
island today, perhaps with the exception of the gopher
tortoise (Gopheruspolyphemus) and the wild turkey
(Meleagris gallopavo) which prefer generally drier terrain.
The turkey is known to roost on branches overhanging the
water which affords the animal some protection against
nocturnal assailants (Schorger 1966) but the tortoise
remains almost exclusively in the sandy areas in which it
makes its burrows (Carr 1952).

Evidences of Seasonality

Among the species represented by the bone material from
Hontoon, only one would be available during only one
season. A fragment of the right tibiotarsus of a Canada
goose (Branta canadensis) was identified from zone 4 of
the 1980 pit. This species presently occurs as a migrant or
winter resident in the middle St. Johns region.

Fishes were probably caught all year around but are known
to be easier to catch in drier winters when the populations
are more concentrated in the deeper pools and lakes.
Mullet become increasingly abundant in freshwater
drainages in late summer, fall, and early winter, particularly
on sandy bottoms in well-rooted main channel vegetation
(McLane 1955). Both the striped mullet (Mugil cephalus)
and the white mullet (Mugil curema) were probably present
in some numbers during all seasons. Most of the other
fishes represented are year around residents of the St.
Johns River.

Reptiles are more easily available during some seasons.
Gopher tortoises retire to deep burrow during the cold
winter weather. Alligators, likewise, are not frequently
encountered during the winter.

Environment and Exploitation by Humans

One of the key species represented in the historic horizon
material is the gopher tortoise. Gopher tortoises are not
found on Hontoon Island today, nor is there evidence of
their presence on the Island in the past. The absence of
tortoises on Hontoon Island may be related to the Island's
soils which are unsuitable for the extensive burrows made
by the tortoises. About a mile away and to the west of the
Island are sand ridges which are a prime area for colonies
of the tortoises. They can be collected when found outside

of their burrows. During extreme temperatures, either
extremely hot or cold, they retire to the more equitable
temperatures of their burrows from which they are difficult
to extract (Taylor 1982).

The pond turtles of the genus Pseudemys are found in
various aquatic habitats particularly lakes, sloughs, and
rivers. They may congregate to sun themselves on logs
sticking out of the water. When they are disturbed during
their sunning they will often slide into the water. People
who catch turtles take advantage of this behavior by
hanging traps in the water under a log used by the turtle
for sunning so that they slide off into the trap. In the
absence of archaeological finds of traps, we have no way of
knowing whether such traps were used by the inhabitants
of Hontoon. The evidence from the faunal assemblage
indicates that turtles were frequently caught. Such
consistent success in catching pond turtles suggests that an
efficient technique, such as a trap, had been developed
based on a knowledge of the behavior of the prey species.

This same knowledge of animal behavior would of course
have been applied to the consistent capture of any animal.
The most abundantly represented fishes are catfish, bass,
and mullet. Today catfish and bass are most frequently
caught with hook and line. Catfishes are bottom feeders
and therefore are most successfully caught with a weighted
hook near the bottom. Bass feed closer to the surface of
the water and are most frequently caught in the upper
layers of the water. Mullet eat plants and are usually
caught in nets. Netting was found at the water logged
deposits at Key Marco (Gilliland 1975) and a small piece of
very similar two-ply cordage was found at Hontoon. A
shell fish hook has also been found at Hontoon. It is likely
that both of these basic fishing techniques were used at

Comparison of the Faunal Assemblage Represented in the
Three Samples

Based on these data some differences can be seen between
the predominant groups of vertebrates represented in the
two horizons of the 1980 pit and between the two historic
horizon samples. The differences and similarities that are
evident are as follows:

1) A comparison between the prehistoric and the historic
horizons of the 1980 pit shows a substantially greater
relative abundance of terrestrial vertebrates in the historic
horizon. Terrestrial turtles, primarily gopher tortoises, are
responsible for this change (Tables 1, 2, and 3).

2) A comparison between the two historic samples shows a
difference in respect to the relative abundance of
terrestrial vertebrates in the samples and that the 1985
historic sample appears to be more similar to the
prehistoric sample than to the historic sample from 1980.

3) When this is analysed in more detail and the relative
abundance of terrestrial turtle, primarily gopher tortoise, is

compared with the relative abundance of aquatic turtle,
primarily pond turtle, the two historic samples are
consistent in having significantly more terrestrial turtle
remains than is found in the prehistoric sample. On the
other hand when the relative abundance of all of the turtle
is compared with all of the fish, the two samples from the
1980 pit are similar to each other and differ significantly
from the 1985 sample which has much more fish than
turtle (Table 4 and 5).

4) Catfish is the most abundant fish in all of the samples.
They are followed in relative abundance by mullet and bass
or gar and bowfin. Many more remains of sunfishes are
recovered from the finely screened column samples and it
may be assumed that they were of significance in these
samples as well. When the relative abundances of the
three large important fishes are compared in the three
samples it is evident that the prehistoric 1980 sample and
the 1985 sample are more similar in this respect than either
is to the historic 1980 sample (Table 5).

5) The data from the column sample taken in 1982 and to
be published elsewhere (Wing in press) can only be
compared with the information reported here with great
caution because of the differences in the recovery
techniques used in collecting the samples. Each of the pit
samples analysed here is 40 times larger than the single
column sample. It is unlikely that everything encountered
in the 1982 square will be represented in the adjoining
column. We also know that most of the small remains,
those smaller than 0.6 cm were not recovered by the
procedure used to excavate the pits. This gap in
information provided by the two excavation procedures
can only be filled by analysis of father samples. However,
in many respects the trends in change seen between
prehistoric and historic samples are similar but differ in

6) The overall breakage of bone is similar in the three
samples ranging from an average fragment size of 3 gm to
5.3 gm. However, differential breakage between key species
is evident. For example, the gopher tortoise fragments are
three times larger in the 1985 sample than in the 1980
samples while catfish fragments are half as large in the
1985 sample as the other. The average bone size in the
1982 column sample is of course much smaller (0.4 gm)
because fine gauge screen was used to recover the
biological remains.


Despite the numbers of detailed studies of biological
remains from Hontoon our knowledge of the uses of plants
and animals by the site's inhabitants is still preliminary
(Purdy and Newsom 1985; Newsom 1986; Wing in press).
Some characteristics of the faunal and floral assemblages
seem to be consistent in the samples from different
locations within the site studied thus far. They may well be
characteristic of the site as a whole but further research
must be conducted on biological remains to verify this.

Table 1. Fragment count of major vertebrate groups identified from the three samples.

Vertebrate 1980 1980 1986
Group Prehistoric Historic Historic
Number Percent Number Percent Number Percent

Mammal 43 3.3 107 5.1 119 3.5

Bird 54 4.2 23 1.1 180 5.3

Terr. Turtle 221 17.0 1373 65.9 665 19.7

Total Terr. 318 24.5 1503 72.1 964 28.5

Aquatic Turtle 356 27.5 274 13.1 139 4.1

Fish 623 48.0 308 14.8 2279 67.4

Total Aquatic 979 75.5 582 27.9 2418 71.5

Total Fauna 1297 2085 3382

Table 2. Relative abundance of aquatic and terrestrial turtles and all turtles and fishes based on the
count of identified remains.

Vertebrate 1980 1980 1985
Groups Prehistoric Historic Historic
Number Percent Number Percent Number Percent

Terr.Turtle 221 38.3 1373 83.4 665 82.7

Aquatic Turtle 356 61.7 274 16.6 139 17.3

Total Turtle 577 1647 804

Turtle 577 48.1 1647 84.3 804 26.1

Fish 623 51.9 308 15.8 2279 739

Total 1200 1955 3083

Table 3. Weight (in gms) of identified fragments of the major vertebrate groups represented in the
three samples.

Vertebrate 1980
Group Prehistoric
Weight Percent

Mammal 210.9 5.5

Bird 107.2 2.8

Terr. Turtle 1065.1 27.6

Total Terr. 1383.2 35.9

Aquatic Turtle 2054.4 53.2

Fish 420.8 10.9

Total Aquatic 2475.2 64.2

Total Fauna 3858.4

Weight Percent

980.3 8.6

59.9 0.5

7896.6 69.5

8936.8 78.7

2166.4 19.1

253.7 2.2

2420.1 21.3


Weight Percent

717.4 7.0

149 1.5

4658.6 45.3

5525 53.7

1095.5 10.7

3664.8 35.6

4760.3 46.3


Table 4. Relative abundance of aquatic and terrestrial turtles and all turtles and fishes based on the
weight of identified remains.

Vertebrate 1980 1980 1985
Groups Prehistoric Historic Historic
Weight Percent Weight Percent Weight Percent

Terr. Turtle 1065.1 34.1 7896.6 78.5 4658.6 81.0

Aquatic Turtle 2054.4 65.9 2166.4 21.5 1095.5 19.0

Total 3119.5 10063 5754.1




3119.5 88.1

420.8 11.9


10063 97.5

235.7 2.3


5754.1 61.1

3664.8 38.9


Table 5. Relative abundances of key species: A weight of identified remains of gopher tortoise, pond
turtle, and catfish; B weight and number of identified remains of catfish, bass, and mullet.

Animal 1980 1980 1985
Prehistoric Historic Historic
Weight Percent Weight Percent Weight Percent

Gopher Tortoise 1006 40.3 7655 79.1 4536 71.8

Pond Turtle 1384 55.5 1925 19.9 1037 16.4

Catfish 106 4.3 93 1.0 741 11.7

Total 2496 9673 6314

Animal 1980 1980 1985
Prehistoric Historic Historic
Weight Percent Weight Percent Weight Percent

Catfish 106.4 65.6 92.7 61.3 741.3 70.7

Bass 39.9 24.6 10 6.6 188.3 18.0

Mullet 15.8 9.7 48.6 32.1 118.9 11.3

Total 162.1 151.3 1048.5

Animal 1980 1980 1985
Prehistoric Historic Historic
Number Percent Number Percent Number Percent

Catfish 141 68.8 97 57.4 1390 66.4

Bass 30 14.6 9 5.3 355 17.0

Mullet 34 16.6 63 37.3 348 16.6

Total 205 169 2093

Table 6 Faunal list for 1985 sample


indet. bird
indet. snake
indet. turtle
indet. fish





Table 8 Faunal list for 1980 Historic Horizon


indet. bird
indet. turtle
indet. fish



% MNI % WT




2224.0 155.0




Table 7 Faunal list for 1980 Prehistoric horizon


2.0 0.1 2.0 0.8 0.3 0.0
43.0 1.1 5.0 2.0 133.7 1.1
60.0 1.6 3.0 1.2 670.8 5.7
14.0 0.4 0.0 0.0 22.6 0.2
180.0 4.7 16.0 6.5 153.2 1.3
2.0 0.1 1.0 0.4 4.5 0.0
1.0 0.0 1.0 0.4 1.3 0.0
19.0 0.5 4.0 1.6 122.3 1.0
646.0 16.8 19.0 7.7 4536.3 38.5
1.0 0.0 1.0 0.4 0.7 0.0
11.0 0.3 2.0 0.8 8.2 0.1
113.0 2.9 5.0 2.0 1037.2 8.8
14.0 0.4 2.0 0.8 49.4 0.4
560.0 14.6 0.0 0.0 1372.2 11.7
1.0 0.0 1.0 0.4 0.2 0.0
92.0 2.4 4.0 1.6 71.8 0.6
59.0 1.5 6.0 2.4 41.0 0.3
1390.0 36.1 100.0 40.7 741.3 6.3
35.0 0.9 13.0 5.3 18.2 0.2
255.0 6.6 17.0 6.9 188.3 1.6
348.0 9.0 44.0 17.9 118.9 1.0
0.0 0.0 0.0 2485.3 21.1


indet. mammal
indet. bird
indet. snake
indet. turtle
indet. fish







% MNI % WT

0.1 1.0 0.9 1.7 0.0
0.1 1.0 0.9 0.4 0.0
0.6 3.0 2.7 8.1 0.1
0.2 1.0 0.9 17.3 0.2
0.3 1.0 0.9 123.0 1.6
1.4 0.0 0.0 60.4 0.8
0.1 1.0 0.9 0.6 0.0
0.1 1.0 0.9 3.2 0.0
0.3 2.0 1.8 3.6 0.0
0.1 1.0 0.9 0.1 0.0
0.2 2.0 1.8 15.5 0.2
0.1 1.0 0.9 1.3 0.0
0.2 1.0 0.9 4.4 0.1
0.2 2.0 1.8 9.5 0.1
0.4 2.0 1.8 24.0 0.3
0.1 1.0 0.9 1.3 0.0
0.3 2.0 1.8 16.3 0.2
1.4 0.0 27.4 0.4
5.6 7.0 6.2 3718.4 47.6
0.2 2.0 1.8 0.5 0.0
0.1 1.0 0.9 0.3 0.0
1.4 4.0 3.5 59.4 0.8
11.8 7.0 6.2 1005.7 12.9
3.2 3.0 2.7 379.6 4.9
1.5 2.0 1.8 20.4 0.3
0.1 1.0 0.9 2.2 0.0
0.4 2.0 1.8 6.3 0.1
13.3 12.0 10.6 1383.6 17.7
0.6 2.0 1.8 86.4 1.1
2.1 5.0 4.4 175.9 2.3
16.8 0.0 0.0 232.1 3.0
0.1 2.0 1.8 0.6 0.0
0.1 2.0 1.8 0.6 0.0
3.1 5.0 4.4 78.9 1.0
2.6 5.0 4.4 31.6 0.4
8.4 11.0 9.7 106.4 1.4
0.7 5.0 4.4 5.0 0.1
0.2 1.0 0.9 1.7 0.0
0.1 2.0 1.8 3.3 0.0
1.8 5.0 4.4 39.9 0.5
2.0 4.0 3.5 15.8 0.2
18.2 0.0 0.0 138.2 1.8


Further research is required to understand fully the nature
of the changes in the uses of plants and animals by the
Indian inhabitants of the site brought about by the advent
of the Spanish entrance upon the Florida scene. More
samples must also be studied to document variability
throughout the site and to gain an understanding of the
causes of this variability.

Certain characteristics of the biological remains seem to be
well established by the ethnobiological studies thus far.
The predominant species in the faunal assemblages from
the different parts of the site remain the same. The change
from a predominance of snails in the prehistoric horizon to
mussels in the historic horizon appears to be constant
throughout the site. Three groups of vertebrates
represented in great abundance in all samples (consituting
32, 54, and 82 percent of these samples) are the terrestrial
turtles primarily the gopher tortoise, the aquatic turtles
primarily the pond turtles, and fishes which includes
primarily catfishes. If comparison between the relative
abundances (based on weight of identified remains) of the
three primary species is made a trend towards increased
abundance of the gopher tortoise in the historic horizon is

A more general trend towards an increase in terrestrial
vertebrates in the historic horizon is less obvious because
of differences in the faunal assemblages of the different
samples. For example, the weight of alligator remains
identified in the prehistoric 1980 samples constitutes 47
percent of the assemblage and the weight of unidentified
fish remains is 21 percent in the 1985 sample whereas these
are negligible in the other samples.

Slender evidence points to year around occupation of the
site. Gopher tortoises and alligators are more easily
caught during the warmer months. The Canada goose and
some ducks are available during their winter migrations to
the south. Evidence from the ethnobotanical research
augments these data.

The faunal evidence indicates that throughout the history
of occupation at the site the inhabitants engaged in a
variety of subsistence activities. These included initially
gathering snails and later digging mussels, catching pond
turtles possibly with traps or by hand, gathering gopher
tortoises, and fishing with both nets and hook and line.
These activities were of course integrated with plant
gathering and cultivation (Newsom 1986).

More comparable samples are needed in order to
understand the differential distribution of species and
differences in the fragment size throughout the site.
Column samples should be increased and examined in such
a way so that one can be confident that they reflect the
faunal assemblage of the adjacent excavation unit. One
way to do this would be to identify and analyze the animal

remains from the unit and compare this with the faunal
remains recovered by identical screen size from the
column. If the faunal assemblages are similar then one
could be secure in the belief that the column represents
the fauna of the larger unit. Furthermore, the faunal
component that fell through the larger gauge screen and
could be identified from the column represents the fauna
lost in the larger excavation unit. By this device, the gap
that exists between the data from the excavation unit and
the column sample could be bridged. These data may then
permit more accurate interpretation of past economies and
site activity areas.


We are grateful to Dr. Barbara Purdy for the
opportunity to study these most interesting samples of
animal remains from the Hontoon site. Thanks are also
due to Tamara Markowitz who spent the major part of the
summer of 1986 identifying the remains from subsample
number 135 of the 1985 sample. Lee Newsom who has
worked intensively on the botanical remains from Hontoon
has helped us in many ways throughout our research on
the animal remains for which we are profoundly grateful.
References Cited

Calr. Archie F.
1940 A Contribution to the Ilerpetology ot Florida. Univeisill
of Florida Puhlication. Biological Science Series No. I p I-
Cumbiaa, Stephen L.
1976 A Reconsideration of Freshwater Shellfish Exploitation in
the Florida Archaic. The Florida Anthropologist 29(2):49-
Gilliland, Marian S.
1975 The Material Culture of Key Marco Florida. University
Presses of Florida. Gainesville 266 pp.
McLane, William M.
1955 The Fishes of the St. Johns River System. Ph.D.
Dissertation University of Florida.

Newsom, Lee A.
1986 Plants, Human Subsistence, and Environment: A Case
Study from Hontoon Island (8-VO-202), Florida. Masters
Thesis University of Florida 152 pp.

Pennak, Robert W.
1978 Freshwater Invertebrates of the United States (second
edition) Wiley-Interscience New York.

Purdy, Barbara A. and Lee A. Newsom
1985 Significance of Archaeological Wet Sites: A Florida
Example. National Geographic Research 1(4):564-569.
Schorger, A. W.
1966 The Wild Turkey its History and Domestication.
University of Oklahoma Press, Norman OK 625 pp.

Taylor. Robert W. Jr.
1982 Human Predation on the Gopher Tortoise (Gopherus
polyphenus) in North-Central Florida. Bulletin of the
Florida State Museum 28(1-4):79-102.

Wing, Elizabeth S.
in press Integration of Floral and Faunal Data from Hontoon
Island. Florida. Archaeozoologia.

Elizabeth S Wing
Florida State Museum
University of Florida
Gainesville, FL 32611

Laurie McKean
and Zoology Department
North Carolina State University
Raleigh. NC 27695


Lee A. Newsom


This paper summarizes the analysis of plant remains from a
submerged shell midden at Hontoon Island (8Vo202), Florida.
The deposit has yielded an exceptional array of well pre-
served biological remains, the result of their having been
entombed in an oxygen-free matrix of peat and shell. This
report focuses on the plant remains obtained from two column

Ninety-seven vascular and nonvascular plants were identified.
Cultivated plants were among those identified, including
bottle gourd, at least three varieties of Cucurbita pepo
(pumpkin/squash/gourd), and corn. The relative distributions
of various wild and cultivated plants changed through time,
with cultivated and disturbed-habitat species becoming more
frequent in later levels. Though cultivated species were
present, the author believes that the inhabitants remained
primarily non-horticultural, pursuing a hunting-gathering-
fishing economy. Their way of life may have been more like
that of Indian groups to the south, than other groups of
Timucuan Indians to the north and west of the island.


Hontoon Island (8-Vo-202) is
located in western Volusia County,
Florida. It is situated on the
St. Johns River, nearly midway
along the river's 315 mile (504
km) course. Several archaeo-
logical deposits are on the
island, including two very large
shell middens, two middens of much
smaller extent, and at least one
shell field or scatter. The mid-
den on the northern end of the
island has been the focus of Uni-
versity of Florida excavations for
five years (1980-1985) under the
direction of Dr. Barbara A. Purdy.
The northeastern and eastern edges
of this midden slope off into a
large, low energy lagoon and
continue for some distance under
the surface of the water, there-
fore affording unusually good
conditions for organic preser-
vation. It is in this water-

saturated deposit that the
majority of the excavation units
were placed. The archaeobotanical
component of this wet deposit is
extensive and diverse. The
species identified represent a
lush river basin and hammock
environment characterized by great
species richness, and upon this
base the island's inhabitants made
their living. Through ceramic
chronology and absolute dating,
the deposit has been demonstrated
to range from the early St. Johns
I Period, or earlier (Purdy, this
volume; Milanich and Fairbanks
1980), through the Spanish Contact
and Mission periods of the six-
teenth and seventeenth centuries
(Ehrmann 1940; Deagan 1978)--
roughly A.D. 0 through A.D. 1750.

Two distinct cultural expressions
have been identified in the work

Volume 40 Number 1


March, 1987


at Hontoon Island. The earliest
is associated with an aboriginal
occupation of long duration,
having existed approximately one
thousand years. The evidence
suggests an extensive habitation
area occupied by a people who had
established a complex economy
based upon resources procured from
the river basin and flanking
uplands. In addition, their diet
may have been supplemented by
gardening or the encouragement of
at least two cultivated plants.
Early in the historic period
change occurred, and there is
evidence of more extensive use of
cultivated plants along with
certain other adjustments to the
subsistence base. The observed
changes in the archaeological
record at Hontoon Island may
reflect voluntary and involun-
tary disruptions in the aboriginal
culture arising with the initial
presence of Europeans in the
region. Further stimulus for
change may have come about dir-
ectly through influences exerted
by the Spanish mission system in
east-central Florida.


Hontoon Island is located in an
area of environmental diversity.
Vegetational complexes in this
region are governed less by soil
fertility than by texture and
moisture holding capacity or
permeability (Snedaker 1971:25).
The climate is relatively mild due
to close proximity to the Atlantic

Along the St. Johns River's north-
erly course the river channel, its
floodplain, and adjacent topogra-
phy, undergo several changes in
character. Because of this, the
river has been described as having
three distinct segments: the Upper
Basin, beginning with the river's

headwaters in southwest Brevard
County, and then north to approxi-
mately Lake Harney; the Middle
Basin, beginning with Lake Harney
and going north to the vicinity of
present-day Palatka; and finally,
the Lower Basin, from Palatka to
the mouth of the river near Jack-
sonville (Figure 1). Hontoon
Island is in the center of the
Middle Basin. The distance from
Hontoon Island to the coast is
approximately 40 km.

The evidence suggests that the
Hontoon Islanders conducted most
of their plant collecting/harvest-
ing activities within close proxi-
mity to home base. The greatest
intensity of collections appears
to have occurred within 5-10 km of
the site, with the majority of the
plants identified being represen-
tative of the river basin and
flanking uplands. Coastal
resources were also exploited,
though perhaps on an occasional
basis, being represented in much
lower frequencies than remains
from nearby habitats.

Within the intensive use area,
being that within 10 km of the
site, are five principal ecolo-
gical communities: freshwater
marsh, hardwood river swamp, hard-
wood hammock, pine flatwoods, and
sand pine scrub (Laessle 1942;
Puckett 1982; U.S.D.A. Soil Con-
servation Service 1985). On the
whole, these reflect a gradient
from wetter conditions and lower
elevations along the river's edge,
to progressively better drained
soils and higher elevations on the
nearby sand ridges. The diversity
in topography and soil/edaphic
conditions is reflected in the
associated plant assemblages, and
the species of plants identified
in the archaeological record
closely mirror the present day


St Augustine

Matanzas Inlet

7 7 Daytona Beach

S\\ New Smyrna

0 40
0 40

Figure 1. Northeast-central Florida including the St. Johns River Basin.


4 6

5 7

6 8

7 9

8 10







Figure 2. Two Column Samples from Hontoon Island. The Left
Represents the 30x30 cm Column; the Right Represents
the 30x15 cm column. Each is Divided into 10 cm


The recent excavations at Hontoon
Island include a 3-m square, exca-
vated in December 1980; a 2-m
square plus column sample, exca-
vated in February 1982; a 26x2-m
trench with column samples, exca-
vated during an archaeological
field school in the spring of
1984; and finally, a pair of 2-m
squares with two column samples,
in May 1985.

This report focuses on informa-
tion obtained from two of the
column samples: one from the pro-
file of the 1982 square, and the
other from the profile of 1985
square 59. The coarse-screened
material from the larger excava-
tion units will receive only
summary treatment as a supplement
to information from the column
samples since the large screen
size used (1.27 cm [0.5"] and
0.635 cm [0.25"]) introduces too
great a bias against smaller sized
plant remains. Both of these
column samples were excavated from
fully water-saturated deposits and
preservation can be considered
roughly equal. The two columns
were excavated and processed in
identical manner, although it must
be noted that they are not equal
in size. The 1985 column was half
as wide as the first column (the
1982 column measured 30x30 cm,
while the 1985 column was 30x15
cm) (Figure 2). The principal
investigator believed a smaller
size was necessary to facilitate
and shorten the time spent in
processing the samples. One
further difference is that the
1985 column was not excavated as
deeply as the 1982 column due to
problems with water control. These
differences are considered in the
discussion on the plant remains.

The column samples were excavated
in 10 cm segments. Each such
segment was bagged in its entirety
and sent directly to the labora-
tory for processing. The indivi-
dual segments then underwent a
series of fine screening and chem-
ical flotation (nested screen
sizes are: 4mm, 2mm, 1mm, and 125
kAm; see Newsom 1986 for a complete

After completion of the initial
flotation and separation process,
the plant remains were further
sorted and identified with the aid
of dissecting and transmitted-
light microscopes. All quantifi-
cations were made using volume as
a unit of relative mass since
measurements based on weight were
deemed unreliable due to the
waterlogged condition of the
material. The faunal material
from the heavy fractions of the
flotations underwent further pro-
cessing at the Zooarchaeology
Laboratory of the Florida State
Museum. All of the above was
carried out under the direction of
Dr. Elizabeth S. Wing, Florida
State Museum.

Seed identifications were made
with the aid of comparative col-
lections belonging to the Florida
State Museum Herbarium and to the
author, and with the use of
several reference works (e.g.,
Martin and Barkley 1961; Hitchcock
1971; Landers and Johnson 1976).
Wood and charcoal identifications
were made using anatomical keys
(Record and Hess 1943-1947; Record
and Hess 1943; Panshin and de
Zeeuw 1980) and by reference to
comparative specimens in the Her-
barium's wood collection, along
with samples collected at Hontoon
Island, Florida. In the interest
of brevity, an analysis of the
wood and charcoal samples will be
reserved for a separate report.

Certain features of the shell mid-
den must be made clear before con-
tinuing on to the description and
analysis of the plant remains.
Earlier it was mentioned that two
distinct components/deposits were
uncovered at the site. This is
evidenced most clearly by changes
in shellfish composition within
the midden which has resulted in
the identification of two separate
"macro-strata" or perhaps, more
accurately, two separate middens.
The first is a very extensive
midden containing evidence of a
wholly aboriginal material cul-
ture. Freshwater snail (Viviparus
georgianus) forms the primary
building material of this deposit,
which dates from ca. A.D. 0 (the
early St. Johns I period) through
approximately A.D. 1500. This
"snail shell" midden was
encountered in all excavations at
the site except the 1985 units.
The second deposit is much smaller
in extent and is situated strati-
graphically superior to the
original midden. It is distinc-
tive in being composed primarily
of freshwater mussel shell
(Elliptio buckleyi). These
superficial or "mussel shell"
levels yielded a number of
European artifacts, and radio-
carbon dates obtained on these
layers range from ca. A.D. 1540 to
the 1750s (Purdy, this volume).

The differences between the two
deposits or middens are being
emphasized as necessary to an
understanding of the following
analysis since marked changes in
the botanical and faunal compo-
nents occurred with the change in
shell matrix noted above. The
mussel shell layer comprised the
uppermost segments (Levels 5-8,
ca. 40 cm thick) of the 1982
column sample and excavation unit.
It likewise occurred in the 1985
units and adjacent column samples,
where it comprised the whole of

each (excluding sterile over-
burden) (i.e., approximately 45-50
cm thick; the underlying snail
shell midden was not encountered
before the water level precluded
further excavation). Thus, in the
1982 column both deposits are
represented, while in the 1985
column only the historic period
midden was encountered. Also, in
association with the shell midden
proper, there is a discontinuous
shell-free midden overlying the
mussel shell levels. Since, other
than the absence or near-absence
of shell, it is identical in plant
and animal remains to the mussel
shell midden, for these purposes
it is considered one and the same.

No attempt will be made here to
rank taxa according to their rela-
tive importance. This is being
delayed until more data are avail-
able, even though the present sam-
ple is thought to be representa-


A total of 3754 seeds and plant
parts were identified from the
1982 column sample (not including
the fragment counts of nut remains
and gourd/squash rind). In addi-
tion 1224 wood fragments were
examined and identified. The
identity of 95 seeds and fragments
remained uncertain. From the 1985
column, 2880 seeds were identified
(again, exclusive of nut remains,
cucurbit rind, and also corn cob
fragments), along with 320 wood
samples. Unidentified seeds and
fragments from the 1985 column
totaled 232.

Tables 1 and 2 show the division
of the plant remains from the two
column samples into gross cate-
gories. "Plant residue" repre-
sents unidentifiable, largely
fragmentary plant remains (e.g.,
leaf litter, bark, roots).

Table 1. Gross Volumes of 1982 Column Sample Floral Material*

LEVELS 5 6 7 8 9 10 11 12 13 14

WOOD 480 1135 1605 2750 2485 490 595 845 225 210

CHARCOAL 116 150 28 75 32 80 135 150 75 167

,IUT SHELL 7 15 36 51 13 7 22 110 23 57

SEEDS 21 41 57 21 16 16 15 11 10 15

CUCURBIT RIND 0.2 2.5 9 4 1.2 0.8 2.3 1 1.7


OTHER** 0.3 -

PLANT RESIDUE 525 1295 1070 2550 1170 905 920 1510 350 451

*Volumes are in cubic centimeters.

**"Other" includes pine cones and scales, cane stem, and fungi.

Table 2. Gross Volumes of 1985 Column Sample Floral Material*

LEVELS 7 8 9 10

WOOD 114 (228) 335 (670) 172 (344) 695 (1390)

CHARCOAL 48 (97) 105 (210) 161 (322) 75 (150)

NUT SHELL 5 (10) 72 (144) 126 (252) 49 (98)

SEEDS 8 (16) 27 (154) 33 (66) 27 (54)

CUCURBIT RIND 0.2 (.4) 1 (2)

ZEA MAYS (COB) -0.3 (.6) 8 (16)

OTHER** 0.1 (.2) 1 (2) 0.3 (.6) 0.2 (.4)

PLANT RESIDUE 659 (1318) 764 (1528) 634 (1268) 495 (990)

*Volumes are in cubic centimeters. Numbers in parentheses are the volumes doubled, to account
for this column being half the size of the 1982 column sample.

**"0ther" includes pine cones and scales, cane stem, and fungi.

The combined identifications from
the two column samples and the
larger excavation units yielded a
total of 97 vascular and
nonvascular plants, representative
of 82 different species of seeds
and miscellaneous plant parts
(e.g., cane stem and pine cone
scales), and 30 different woods.
Table 3 lists these taxa. The
identifications from the snail
shell and mussel shell deposits
closely paralleled each other
where native (wild) plants are
concerned, that is, in terms of
what species were actually repre-
sented. Differences were observed
though, in their frequency and
distribution through time. Marked
differences between the two depo-
sits were found in the presence
and relative abundance of culti-
vated plants. Corn was found only
in the historic period mussel
shell levels and, in addition,
noteworthy changes in the pumpkin/
squash remains were discovered.

Wild Plant Resources

At least 90 wild or indigenous
plants have been identified; among
these are bracket fungi, aquatic
plants, grasses, sedges, herbs,
vines, shrubs, and trees. With
the aid of ethnohistoric and
ethnobotanical literature we can
begin to identify which plants
might have been economically
valuable as food, oils, and
condiments to the human inhabi-
tants of Hontoon Island (medicinal
and other uses, such as dyes, will
not be considered here).

The early accounts of exploration
and attempts at colonization in
the New World are replete with
references to plants, both in an
effort to describe natural vegeta-
tion and to relate how and which
among the native plants were being
used by Native Americans. These

latter observations were parti-
cularly important to the first
European explorers since it often
meant the difference between life
and death as supplies from their
homelands were not reliable or
were not forthcoming at all. Reitz
and Scarry (1985) and others (cf.
Sturtevant 1962:68) have found
that sixteenth century Spanish
colonists came to rely almost
exclusively on New World plants
and animals primarily for these
reasons. Thomas Hariot, an
Englishman and "planter," came to
"Virginia" (now coastal North
Carolina) around 1585 (Hariot
1951). As a farmer, he paid par-
ticular attention to the soils,
climate, what was being sown and
gathered by the Indians, their
processing methods, and more.
Because many of the same plants
and associated human traditions
are common to the Southeastern
Coastal Plain and range into cen-
tral Florida, his work is most
enlightening. Table 4 lists the
plants mentioned in a few selected
sixteenth century descriptions of
Florida and the southeastern
United States.

Laudonniere's (1975) and Ribaut's
(1964) accounts have more direct
bearing on this study since most
of the Indians they had contact
with were members of the Timucuan
tribe--the parent group with which
the inhabitants of Hontoon Island
are believed to have been affili-
ated (Newsom 1986:12-22). A num-
ber of the plants mentioned by the
explorers and listed in Table 4
have been found in the Hontoon
Island samples, including: hic-
kory, bay, oak, grape, persimmon,
elderberry, sorrel, and more.
Table 5 lists the plants identi-
fied from the archaeological
deposits at Hontoon Island which
might have been utilized for their
potential economic value.

Table 3. Floral Species List: Hontoon Island, Florida (8-Vo-202).

Part One: Seeds, Nuts, Plant Parts


Cucurbita Repo L., type one
C. pepo, type two
C. pepo, type three
C. pepo, (?) type four
Lagenaria siceraria L.

Zea mays L.
? Pyrus malus L.

"ornamental gourd"
bottle gourd

domestic apple


Amaranthus sp.
Ambrosia sp.
Ampelopsis arborea (L.) Koehne
Asteraceae [inflorescence]
Berchemia scandens (Hill) K. Koch
Brasenia schreberi Gmel.
Capsella bursa-pastoris (L.) Medic.
? Carex sp.
Carva aquatic (Michx. f.) Nutt.
C. glabra (Mill.) Sweet
Celtis laevigata Willd. [leaves and seed]
Cephalanthus occidentalis L.
Cornus florida L.
C. foemina Mill.
Cyperaceae, type one
Cyperaceae, type two: tublar shape
Cyperus sp.
Dioda virginiana L.
Diospyros virginiana L.
Fomes sclerodermeus (Lev.) Cooke thalluss]
Gaylussacia sp.
Hypericaceae [inflorescence]

Ilex opaca Ait.
? Ipomoea sp.
Lyonia sp.

L. ferruginea (Walt.) Nutt.
Magnolia grandiflora L.
M. virginiana L.
Myrica cerifera L.
Naias guadelupensis (Spreng.) Magnus
N. marina L.
Nuphar lutea Sibth. & Small

amaranth; pigweed
amaranth family
sunflower family
rattan vine
water shield
shepherd's purse
water hickory
pignut hickory
goosefoot family
flowering dogwood
swamp dogwood
sedge, unidentified
sedge, unidentified
bracket fungus
St. John's-wort
American holly
rusty Lyonia
southern magnolia
sweet bay
wax myrtle
pondweed; naiad
spiny naiad

Table 3 continued.

Nymphaea sp., prob. N. odorata Ait.
Nyssa sylvatica Marsh.
Passiflora incarnata L.
Persea sp.
P. ? palustris (Raf.) Sarg.
Physalis sp.
Phytolacca americana L.
Pinus clausa (Chapm.) Vasey [cones]
P. elliottii Engelm. [cones]
P. palustris Mill. [cones]
Polygonum sp.
P. densiflorum Meisn.
P. hydropiperoides Michx.
P. punctatum Ell.
Polyporus hydnoides Swartz ex Fries thalluss]
P. sulfureous (Bull.) Fries thalluss]
Portulacca oleracea 1.
Prunus caroliniana Ait.

Quercus lyrata Walt.
Q. michauxii Nutt.
Q. nigra L./Q. laurifolia Michx.

Q virginiana Mill.
Rhynchospora sp.
Rubus sp., type one
Rubus sp., type two
Rumex sp.
Sabal palmetto (Walt.) Lodd. ex Schultes
Sambucus canadensis L.
Scirpus sp.
S. validus Vahl
Serenoa repens (Bartr.) Small
Setaria sp.
Sida sp. [seeds and fruit]

Smilax sp. [leaves]

? Sparganium sp.
Tillandsia usneoides (L.) L. [stems]
Ulmus americana L. [buds]
Vaccinium sp., prob. V. corymbosum L.

Vitis sp. (? cultivar; Watlington 1985)
V. aestivalis Michx.
V. cordifolia Lam.
V. munsoniana Simpson
unidentified charred rhizomes
unidentified thorns, prob. Rubus spp.

black gum
may pop
swamp bay
sand pine
slash pine
longleaf pine
dotted smartweed
bracket fungus
sulphur shelf
common purslane
Carolina laurel
overcup oak
swamp chestnut oak
water oak or
laurel oak
live oak
cabbage palm
soft-stem bulrush
saw palmetto
Indian hemp
Spanish moss
American elm
highbush blue-
summer grape
frost grape
muscadine grape
? waterlily

Table 3 continued.

Part II: Woods


Arundinaria gigantea (Walt.) Chapm.
Sabal palmetto (Walt.) Lodd. ex Schultes

switch cane
cabbage palm


Juniperus sp.
Pinus sp., section Diploxylon
Taxodium sp.

red cedar
hard pines


Acer sp., prob. A. saccaharinum L.
Carya sp.
C. aquatica (Michx. f.) Nutt.
Cephalanthus occidentalis L.
Diospyros virginiana L.
Fraxinus sp.
F. caroliniana Mill.
Gordonia lasianthus (L.) Ellis
Liquidambar styraciflua L.
Lvonia sp.

Nyssa svlvatica Marsh.
Persea sp.
Prunus serotina Ehrh.
Quercus sp., red oak group
Q. virginiana Mill.
? Rhus sp.
Salix caroliniana Michx.

Sapindus marginatus Willd.
Ulmus americana L.
Ulmaceae, ? Celtis sp.
Vitis sp.
Unidentified hardwood no. 1, 1984

Unidentified hardwood no. 2, 1985
Unidentified hardwood no. 3, 1985

silver maple
water hickory
water ash/pop ash
loblolly bay
magnolia family
black gum
black cherry
water or laurel oak
live oak
Coastal Plain
American elm
grape vine
? Bursera sp.
(gumbo limbo)
tropical wood
tropical wood

Table 4. Plants Recorded in Early Historic Period Documents.*

Plants described as foods:

1. Plants managed by the Indians.

maize (corn) . . . .

squash . . . .
pumpkins . . . .
gourds . . . .
kidney beans . . .
beans . . . .
sunflower . . . .
"melden" (possibly Polygonum spp.) .
"wickonzowr"/"peaze" (planted peas). .
"plantafolis" (planted with corn) .
"uppowoc" (tobacco) . .

2. Plants gathered from the wild.

roots, water plant . .
blackberries . .. .
walnuts . .
prickly pears . . .
mulberries . . .
cocoplums . . . .
palmetto berries, "roots" . .
acorns, five types . . .
arrowroot/coontie (Zamia pumilia) ..
root (prob. Apios sp., groundnut). .
plums . . . *
"medlar"/"ameixas" (persimmon) ...
grapes . . . .
pecan/hickory . . .
chestnut . . . .
onions . . . .
peas, native . . . .
roots, general . . .
"opeepenauk," a dry ground root/tuber.
"kaishucqenauk," a root/tuber .
"tfinaw," catbrier (Smilax sp.) .
"coscushaw," a root . . .
"habascon," a root . . .
"metaquesunnauk" (? cactus) . .
"sacquenummener," berries from a
water plant (? waterlily) .
strawberries . . . .
reed with seeds (prob. switch cane)
blueberries . . . .
"graynes and herbes" . . .
cassina drink . ....
"maste" (probably nut foods) ...
simpless" (? plums, cherries) .

CV, SM, F, R, TH, D, L,
CV, KE, R, TH, H, D, JR, L
TH, D, JR, L
CV, KE, TH, H, JR, L
TH, ? L

CV, KE, R, H, L
SM, D, L
SM, KE, R, TH, L
F, L
F, ? TH
KE, H, L
KE, TH, H, L
KE, TH, H, JR, L

H, L
H, D, L

Table 4 continued.

"mil" or "mill," "graines of
palm berries, roots (cabbage
"peasen" (?) . .
"sorrel-like plant in Indian
(? goosefoot/amaranth) .
"pinocqs," waterlily seeds .
"chinaroot bush" . .

Plants mentioned in descriptions of "the country:"

walnut and hickory trees . .
"laurel" (probably bay tree) .
"liquid-amber" (sweetgum) .
cedars . . . .
savings (?) . . .
evergreen oak (live oak) . .
red oaks . . . .
oak, unspecified .. . ..
pines . . . .
"palmitos" (palmetto) . .
sassafras . . .
mulberry trees .. .
fir trees . . . .
"rakiock," (for canoe manufacture) .
cypress trees . . .
maple trees . # .
witch-hazel . .
holly . . . .
willow . . . .
reeds . . .
beech .. . . .
ash . . . .
elm . .. . ..
"ascopo" (probably Zanthoxylem sp.)
black cherry . . .
"lentiskes" (?) . . .
chestnut trees .. . .
bays . . . .
palm trees . . ..
magnolias . . .
grape vines . . .
yucca . . .
Spanish moss . . .

CV, TH, H, L
CV, TH, H, L
TH, H, L
CV, TH, H, L
TH, H, L
TH, H, L
TH, H, L
H, JR, L
H, L
H, L
H, L

"chinaroot bush" (?) .. . L

*Key: SM=Solis de Meras (1964); CV=Cabeza de Vaca (1871); F=Fontaneda
(1854); KE=Knight of Elvas (1922); R=Ranjel (1922); TH=Thomas
Hariot (1951); H=Hakluyt (1909); D=Dickinson (1945); JR=Jean Ribaut
(1964); L=Laudonniere (1975).

Mill" .
palm) .

. .
* .
* .

. H
. D, L
. JR

Table 5. Wild Plants Having Potential Subsistence Value.

Plant Species

Part Used

Acer sp. (maple)
Amaranthus sp. (amaranth)
Amaranthaceae (amaranth family)
Arundinaria gigantea (switch cane)
Brasenia schreberi (water shield)
Capsella sp. (shepherd's purse)
Carya spp. (hickory)
Celtis laevigata sugarberryy)
Cenopodiaceae goosefoott family)
Cyperus sp. (nut sedge)
Diospyros virginiana (persimmon)
Gaylussacia sp. (huckleberry)
Nuphar lutea spatterdockk)
Nymphaea sp. (waterlily)
Nyssa sylvatica (black gum)
Passiflora incarnata (may pop)
Persea borbonia (red bay)
Physalis sp. (groundcherry)
Phytolacca americana (pokeweed)
Pinus spp. (pine)

Polygonum spp. (smartweed/knotweed)
Polyporous sulfureous (sulphur shelf)
Portulacca oleracea (common purslane)
Prunus serotina (black cherry)
Quercus spp. (oak)
Rhus sp. (sumac)
Rubus spp. (blackberry)
Rumex sp. (dock)
Sabal palmetto (cabbage palm)
Sambucus canadensis (elderberry)
Sapindus marginatus (soapberry)
Scirpus spp. (bulrush)
Serenoa reopens (saw palmetto)
Setaria sp. bristlegrasss)
Smilax sp. (greenbrier)
Vaccinium sp. (blueberry)
Vitaceae (grape family)

seeds, greens
seeds, greens
seeds, shoots
roots, tubers, leaves
greens, seedpods
seeds, greens
tubers, seeds
tubers, seeds
mature fruit
young shoots and greens
young shoots, young male
cones, inner bark
seeds, whole plant
tender edges of thallus
stems, leaves, seeds
fruit, young shoots, leaves
greens, stems
fruit, buds
seeds, shoots, tubers
fruit, terminal bud
young shoots, leaves, tubers
fruit, young leaves

Source: Angier 1980; Coon 1974; Hudson 1976; Murray and Sheehan 1984;
Munson 1984; Peterson 1977; Peterson and Munson 1984; Swanton
1946; Wilder 1975.

Since many of the plants listed in
Table 5 occur naturally in the
marsh/river swamp ecotone where
the midden is located, care must
be exercised in interpreting them
as food remains as their pres-
ence in the archaeological record
may be more incidental than from
deliberate use, given the quality
of preservation and natural seed
rain. The fact that most of the
seeds show no evidence of human
manipulation or alteration, such
as being charred, broken (or
ground), or clustered in caches,
tends to support a conclusion of
their natural intrusion in the
site, although it does not pre-
clude the possible use of certain
of these plants (e.g., amaranth
and spatterdock) for greens, tub-
ers, or other plant parts (cf.
Peterson 1977). These parts would
have been consumed in their
entirety, leaving the seeds as
unused residue. In the absence of
strong contextual data, much of
the preliminary interpretation has
been based on analogy with ethno-
historic and ethnobotanical infor-
mation, keeping in mind certain
problems with doing so (cf. Gould
1978; Gould and Watson 1982;
Cordell 1977).

Wild Plants Utilized for Food.
The wild plant resources for which
direct evidence of usage (subsis-
tence) exists are: hickory, oak,
may pop, grape, saw palmetto, and
cabbage palm. In addition, at
least thirteen more plants are
believed to have contributed to
subsistence, though the evidence
is less clear. Each of these is
described below.

Persimmon seeds and portions of
pericarp were recovered in the
1985 unit excavations and column
sample (one seed). Though not
found in abundance (7 seeds
total), persimmon was undoubtedly
valued as a food resource at

Hontoon Island as it was among
other Southeastern Indians
(Swanton 1946:265, 286). Persimmon
was mentioned by several early
explorers (see Table 4), being
referred to as medlarss" or
"ameixas," and perhaps also "sim-
ples" by the French. A medlar is
a European fruit in the rose fam-
ily--Mespilus germanica (Tutin et
al. 1968), which is very similar
to the native persimmon. Hariot's
description leaves little doubt as
to which fruit he is referring:

Medlars a kind of verie
good fruit, so called by
us chieflie for these
respects: first in that
they are not good until
they be rotten: then in
that they open at the head
as our medlars, and are
about the same bigness:
otherwise in taste and
colour they are farre
different for they are as
red as cheries and very
sweet: but whereas the
cherie is sharpe sweet,
they are lushious sweet.
(Hariot 1951:D)

In addition to ethnohistoric
documentation, evidence for the
utilization of those plants men-
tioned above comes in the form of:
(1) the condition of the remains
themselves (i.e. charring, break-
age); and (2) in their relative
quantities, keeping in mind that
certain plants are more prolific
than others at a given time. Since
no firehearths or features were
defined and excavated as separate
units, precise contextual evidence
as to utilization is presently

Tables 6 and 7 show frequencies
and/or volumes of the most con-
spicuous plant remains identified
in the column samples (excluding
cultivated plants which will be

Conspicuous Plant Remains (Seeds, Nuts, llisc.) From the 1982 Column Sample.

6 7 8 9 10 11 12 13 14

1.6 3.4 4.0 1.2 3.0
5.0 11.5 32.0 50.0 10.0

4 207 71 27 15
0 0 3 2 9
0 0 0 0 0
0 0 0 0 1
0 0 0 0 1
8 2 3 1 1
1 0 0 0 2
10 3 0 3 0
53 38 8 52 18
0 54 26 28 26
2 36 130 37 2
0 0 0 0 0

4.6 14.4 38.0 7.0 7.1
2.0 8.0 72.0 16.0 50.0


Cabbage palm
Saw palmetto
'lay pop
Black gum

hut sedge
Water shield

*Nut remains are

represented by volume only (in cubic centimeters); other plants by numerical

Table 7. Most Cons
From the


picuous Plant Remail
1985 Column Sample.

ns (Seeds, Nuts, Misc.)

1.4 61.7
3.1 10.0

1 219
0 1
1 14
1 6
0 5
0 0
7 4
0 0
0 0
2 7

Nut sedge
Water shield


*Nut remains are represented by volume only (in cubic centimeters);
other plants by numerical count.


0 0 3
0 12 15
0 0 0
0 0 4
1 11 13
0 0 0
0 0 0
1 13 27

0 9 28
0 11 16
0 3 0
0 9 35
34 40 94


Cabbage palm
Saw palmetto
May pop
Black gum



Table 6. Most

described separately). Nut
remains, specifically oak and
hickory, are represented by volume
since their high degree of frag-
mentation renders a numerical
value meaningless. The rest of
the plant remains are represented
by absolute number.

Nut remains were among the most
abundant wild plant resources in
all tests from the site. Acorns
appear to have been more inten-
sively used in later time periods
corresponding with the mussel
shell midden. Acorn had a volume
of about half that of hickory in
the lower snail shell strata
(Levels 10-14, 1982 column) but
increased to more than twice as
much as hickory in the mussel
shell layers (in particular the
1985 column and larger excavation
units) (Tables 6 and 7). Hickory
does not seem to follow as clear a
trend as acorn and was probably a
less important food. At least a
third of the hickory remains are
from the bitter-tasting water
hickory (Carya aquatica, and
these show no sign of use, indi-
cating deposition under strictly
natural conditions. The apparent
increase in acorn remains from
Hontoon Island runs contrary to a
general pattern outlined by
Yarnell and Black (1985:97) for
other roughly contemporaneous
cultural areas in the southeastern
United States, although Alexander
(1984a:22-24) documented a similar
increase in acorn usage at three
Fort Walton Period sites in Leon
County, Florida.

Two oak species occurred exclu-
sively in the mussel shell strata:
Quercus michauxii (swamp chestnut
oak), and Q. lyrata (overcup oak).
These species bear exceptionally
large acorns and are members of
the white oak group (Sargent 1965:
240), which would make them pref-
erable to the common (in this

region) red oaks. This is because
acorns from white oaks can be
eaten directly, whereas those from
red oaks, which have a high tannin
content, must first be subjected
to a long leaching process (Larson
1980:188). These species cur-
rently range north of Hontoon
Island, making their presence in
the samples especially inter-
esting. Two (or three) additional
oak species were identified:
Quercus virginiana (live oak),
which is also in the white oak
group, and either, or perhaps
both, Q. laurifolia and Q. nigra
(laurel and water oaks)--their
acorns being very difficult to
distinguish. These latter two
species are members of the red oak
group. Hull fragments from live
oak and laurel/water oak acorns
were recovered throughout all
temporal levels of the site.

Acorns were often mentioned in
ethnohistoric accounts (cf. Solis
de Meras 1964:175, 226; Hariot
1951:DI-D2; Ranjel 1922:107). They
were used by the Indians in a
variety of ways. Acorns were
boiled to extract their oil--
Hariot (1951:D1-D2) recorded three
species which he says were pre-
ferred for this purpose. The nut
meats were dried on "hurdles of
reeds" over fires so that they
could be stored for later consump-
tion, at which time they were
soaked in water and then either
eaten directly, or pulverized and
made into "loaves or lumpes of
bread" (Hariot 1951:D1-D2).
Hickory nuts and walnuts were used
in similar fashion: that is, for
direct consumption, in a bread, or
for oil content (Cabeza de Vaca
1871; Hariot 1951:11; Knight of
Elvas 1922:74, 221; Swanton 1946:
365 -366).

Most of the acorn recovered was in
the form of uncharred, fragmented
hull, with the exception of one

intact charred nut meat of the
laurel/water oak type, and a few
charred hull fragments from the
1985 column sample. These, we
might infer, were accidentally
charred, perhaps during the drying
process described above. Very
little hickory shell was charred.

After acorn and hickory, cabbage
palm was the next most abundant
wild plant resource recovered.
Like acorn, cabbage palm under-
went a substantial increase in the
superficial levels. It averaged
47 fruits per column sample level
in the lower five levels (snail
shell midden), whereas in the
mussel shell strata (evidenced
most clearly in the 1985 column),
palm averaged 208 fruits per
level, with the exclusion of
uppermost Level 7, where it under-
went a marked reduction in number
(Tables 6 and 7). Correcting for
the smaller size of the 1985
column as compared to the 1982
column, this increase in abundance
is even more dramatic--averaging
416 fruits per level in the mussel
shell layers. This increase
could, in reality, be more a
factor of the particular location
or placement of the excavation
units since cabbage palm is an
abundant producer of fruit and
perhaps a tree(s) was once near
the location of the 1985 units. An
cultural explanation does not
appear to be indicated though,
since at least eleven additional
plants, including oak, underwent
an identical pattern of increase
during the historic period, fol-
lowed by a sudden marked reduc-
tion in the early-middle eight-
eenth century. The fruits of
cabbage palm are edible, and can
be eaten fresh (Larson 1980:204-
205; Swanton 1946:283). It may be
that the "palm berries" consumed
by Dickinson and his party (1945:
61, 70) during their unfortunate
trek up the Florida coast in 1696

and 1697, were cabbage
or the saw palmetto).
shoots of the palm are
(Swanton 1946:284).

palm (and/
The new
also edible

Plants Associated Primarily With
the Mussel Shell Levels. At
least ten other plants underwent
the same temporal pattern of
increase and decline as palm and
acorn. These include: amaranth
(Amaranthus sp.), and an addi-
tional member of the amaranth
family (Amaranthaceae), bristle-
grass (Setaria sp.), nut sedge
(Cvperus sp.), huckleberry (Gay-
lussacia sp.), blueberry (Vacci-
nium sp.), pokeweed (Phytolacca
americana), and at least three
more--Chenopodiaceae goosefoott
family), groundcherry (Physalis
sp.), and elderberry (Sambucus
sp.)--though these latter three
were far less abundant than the
others. All of these are poten-
tial food plants (see Table 5).
The amaranths and Chenopodiaceae
provide both edible seeds and
greens, nut sedge has edible
tubers, pokeweed has edible greens
when they are young, and fresh
fruit could be had from the
huckleberry, blueberry, ground-
cherry, and elderberry (Peterson
1977). Blueberry is mentioned dir-
ectly in ethnohistoric documents
(e.g., Hakluyt 1909:451), and the
descriptions could apply to both
blueberry and huckleberry since
they are closely related plants
and are very similar in appear-
ance. There are vague references
to weedy plants growing in agri-
cultural fields (Hariot 1951:C2-
C3; Laudonniere 1975:122) which
could apply to the amaranths and
goosefoot, but also to a few other
plants which frequent tilled soil.
"Roots" or tubers (from nut sedge,
for example) were often described
as foods by early explorers or
settlers (Hakluyt 1909:451; Ribaut
1964:73; Laudonniere 1975:122).
Hariot described at least five

such "Rootes," and in addition
related their Indian names, habi-
tat, and how they were used
(Hariot 1951:C4-C5).

The identification of bristle-
grass (Setaria sp.) was made by
Dr. David Hall, Florida State
Museum Herbarium, who is an expert
in the grass family. This plant
is not presently known to have
been used as a food resource by
southeastern Indians, but similar
grasses played very important
roles among these people--
especially Phalaris spp. (may
grass), which has recently been
placed in a quazi-cultivar cate-
gory (cf. Asch and Asch 1985;
Cowen 1978; Yarnell and Black
1985:99). Setaria sp. has been
found in abundance in human copro-
lites recovered from archaeolo-
gical sites in the Tehuacan Valley
in Mexico (Bender 1975:188). There
its use as a food was apparently
long in duration. It too has been
proposed as a possible candidate
for having been cultivated (in
Mexico) (Callen 1965).

Of the eleven plants just men-
tioned (cabbage palm, amaranth,
Amaranthaceae, bristlegrass, nut
sedge, huckleberry, blueberry,
pokeweed, Chenopodiaceae, ground-
cherry, and elderberry), the seeds
of only three--cabbage palm, poke-
weed, and Chenopodiaceae--were
found charred and then only in
very small percentages. But since
the subsistence value of these
plants rests primarily in their
fresh fruits and greens, direct
contact with fire seldom should
have occurred.

Stronger support for their infer-
red function as subsistence items
(versus a strictly environmental
inclusion in the samples) can be
had with a consideration of their
temporal distributions and habi-
tat. All of these plants under-

went an increase during the period
in which the mussel shell midden
was accumulating, and for some
species, this increase was very
pronounced. Further, following
the expansion, each plant under-
went a simultaneous and perhaps
rather sudden, decline or almost
complete disappearance by Level 7
(the uppermost cultural stratum of
the mussel shell midden, 1985
column). The fact that each one
of these plants followed the same
pattern of abundance and expan-
sion, followed by a marked reduc-
tion in number, is a good indica-
tion that their presence in the
samples might have been due to
human agency and not simple coin-

In addition, two other events
occurred in conjunction with those
outlined above: (1) bone, char-
coal, wood, and ceramics also
underwent a pronounced decrease in
Level 7; and, (2) a second group
of plants underwent a reverse
trend, which is a notable increase
in Level 7 (see below). To illus-
trate the former, faunal remains
had a volume of 1103 cc in Level 8
of the mussel shell midden (1985
column sample), but only 17 cc in
Level 7. Likewise, nut shell
dropped by a similar percentage,
and wood and charcoal dropped by
at least a half (see Table 2).

Plants Indicating a Return to
Natural Conditions. In contrast
to those plants and other biolo-
gical remains which decreased in
Level 7 of the 1985 column, at
least three plant species
increased in abundance. These are
buttonbush (Cephalanthus occiden-
talis), knotweed (Polygonum spp.),
and bulrush (Scirpus spp.). The
habitat occupied by buttonbush is
quite limited. It is a component
of the river swamp and occupies
the outer fringe of the swamp--the
zone in closest proximity to the

freshwater marsh and river. As
observed on the island today,
buttonbush, along with Coastal
Plain willow (Salix caroliniana),
forms dense thickets between the
lagoon and more open vegetation of
the inner, or landward, swamp
zone. Neither plant is known to
have any subsistence value, and it
is conceivable that they might
have been partially cleared away
during the occupation of the site
to facilitate access to the
lagoon. Adzed chips and/or cop-
piced stems of both woods have
been identified in historic and
prehistoric levels of the site.
Seeds were not recovered at all,
except for the abundance of but-
tonbush seeds in Level 7 of the
1985 column sample (Table 7) and
contemporaneous levels of the
adjacent units. Subsequent to the
abandonment of the area by humans,
buttonbush (and willow) would have
quickly regained its position bor-
dering the lagoon, a likely reason
for the abundance of buttonbush
seeds in Level 7. Willow seeds are
very small in size, plumed and
light-weight for wind dispersal.
Consequently, the chances of their
being present are minimal.

The situation is slightly differ-
ent for the other two plants,
knotweed and bulrush, which
increased as buttonbush did. Like
buttonbush, these plants inhabit
hydric environments--the river
swamp and marsh respectively--but
unlike buttonbush, both have
edible parts. The seeds and shoots
of bulrushes are edible, and their
possible use by aboriginal inhabi-
tants of the island has been
considered, but the plant, like
buttonbush, appears to have under-
gone a recolonization or radiation
after, or as, humans left the
area. Knotweed was probably used
as a food at Hontoon Island, both
in prehistoric and historic times,
as it was throughout the South-

east. The whole plant is edible
(cf. Murray and Sheehan 1984:282-
298). In limited support of this,
is that some of the seeds from
Hontoon Island are charred, but
this plant too, is interpreted
(also) as an indicator of a return
to natural swamp vegetation since
it became far more common by Level
7. Additionally, a greater variety
of knotweed species appeared
(going from one species in the
lower levels to at least three in
Level 7).

To reiterate, an inverse relation-
ship was documented between twelve
native plants (oak, cabbage palm,
amaranth, Amaranthaceae, bristle-
grass, nut sedge, huckleberry,
blueberry, pokeweed, Chenopodia-
ceae, groundcherry, and elder-
berry) which were common in the
accreting stages of the mussel
shell midden and which were then
drastically reduced in number by
the final stages, and those plants
(buttonbush, bulrush, knotweed)
which were relatively inconspic-
uous or absent in the lower
levels, but then underwent an
increase when the others declined.
Like the first group of plants,
other categories of remains were
also abundant prior to Level 7 and
then experienced the same decline
--all of which is taken as evi-
dence that these plants were being
concentrated at the site during
the time prior to Level 7, after
which the evidence for them is
meager as the site saw a drastic
reduction in usage, or was aban-
doned altogether. The fact that
twelve plant candidates for human
subsistence underwent a substan-
tial reduction in number along
with other subsistence remains
(fauna) and items of material cul-
ture, is an indirect indication
that these particular plants were
being used by the Indian occupants
of the site.

There is a possible alternative
explanation which could account
for the more abundant presence of
at least seven of these plants
during site occupation, and
likewise, their subsequent decline
as human presence lessened. Unlike
oak, cabbage palm, blueberry,
huckleberry, and elderberry, which
normally occur in the hammock/
river swamp environment surround-
ing the site, the other seven
plants--amaranth, Amaranthaceae,
bristlegrass, nut sedge, pokeweed,
Chenopodiaceae, and groundcherry--
characteristically inhabit
disturbed sites and two even have
species which tend to be found on
shell middens (i.e., Setaria
macrosperma and Cyperus plani-
folius [Wunderlin 1982:91, 104]).
What this means is that though
they might have occurred in the
vicinity of the site anyway, with
the presence of humans their habi-
tat would have been greatly
enhanced and expanded, and they
might, therefore, occur in greater

This is a phenomenon known to
botanists as the "door-yard" or
"dump-heap garden" effect (Harlan
and de Wet 1963:19; Rindos 1984:
128-134). Human foot traffic,
deliberate and/or unintentional
soil disturbances, and a human
induced concentration of nitrogen
and phosphates in the soil favor
growth of certain plants (weeds)
which would otherwise be less able
to compete. Humans uncon-
sciously (in a passive role) widen
and strengthen their niche by
eliminating potential compet-
itors and by churning and enhanc-
ing the soil in a way for which
these plants are suited (or have
become adapted). Thus, the
"anthropogenic environment"
(Rindos 1984:134) created with the
formation of the midden and the
concomitant disturbances to the

soil and natural vegetation may be
the primary factor accounting for
the progressively increased abun-
dance of these seven plants as
site occupation continued. Then,
logically, as humans quit the
area, and natural conditions began
to prevail, their niche would
begin to fade and a reduction in
their numbers would be expected,
as was observed for Level 7 and
the overlying, culturally sterile

This second hypothesis, which
emphasizes life cycles and biology
of individual plants, is just as
likely an explanation for the
changing frequencies of the seven
plants in question, as the first
hypothesis of direct human usage.
Even if it is eventually shown
that the second proposed hypo-
thesis better explains the plants'
presence and temporal distribu-
tion, it must be noted that this
does not preclude their having
been used as foods or otherwise by
the Indian occupants. Their close
proximity and ready availability
would have made them attractive
candidates for use. Rindos (1984:
138-143) and others (e.g.,
Flannery 1968) have postulated
that this is a likely avenue of
early plant domestication--taking
the form of a human-plant co-
evolution. With this in mind,
others have come to the conclusion
that it is probably no simple
coincidence that many known and
potential domestics among the
southeastern Indians are plants of
this type (see for example,
Yarnell and Black 1985). Analysis
of human coprolites would contri-
bute a great deal toward resolving
questions surrounding the plants
described above.

Other Potentially Utilized
Plants. Nine additional plants
also might have been utilized,
though they occurred in much lower

frequencies than the plants men-
tioned above. Three occurred
almost exclusively in the mussel
shell levels, while three others
were more prominent in the snail
shell midden. Finally, the last
three were quite scattered
throughout all levels. The three
which are associated with the
superficial layers are Sida sp.
(Indian hemp or broomweed), black-
berry (Rubus spp.), and purslane
(Portulaca oleracea) (Tables 6 and
7). Those associated with the
underlying, snail shell midden are
grape (Vitis spp.), black gum
(Nvssa sylvatica), and spatterdock
(Nuphar lutea), and lastly, those
which are more evenly distributed
are saw palmetto (Serenoa reopens ,
may pop (Passiflora incarnata),
and water shield (Brasenia
schreberi) (Tables 6 and 7).

Both purslane and blackberry are
food plants (Table 5), and occur
in disturbed habitats, including
human living and activity areas.
Purslane was found exclusively in
the historic period Levels 8-10 of
the 1985 column. Blackberry
occurred in both historic and
prehistoric contexts (more in the
upper levels), though never in
abundance. Broomweed was found
only in Levels 8 and 9 of the 1985
column sample. Its potential
value, if any, is not known,
though its alternative common
name--Indian hemp--would suggest
that it could possibly have pro-
vided fibers for twine or the
like. Since it too characteris-
tically inhabits disturbed sites
and hammocks, its presence alone
is not a viable indicator of
importance to the Indians.

Saw palmetto (Serenoa reopens may
pop (Passiflora incarnata), and
water shield (Brasenia schreberi)
were identified in both pre-
historic and historic period con-
texts in roughly the same fre-

quencies (Tables 6 and 7). All
three plants have edible parts:
the roots, tubers, and leaves of
water shield; mature fruit and
terminal bud of saw palmetto; and,
mature fruit of may pop (Peterson
1977). Water shield is a fairly
common plant in the freshwater
marsh and its presence in the
samples is just as likely inci-
dental as it is the result of
human utilization. The seeds of
saw palmetto though, were always
found charred, which is a good
indication that it was used. The
"ripe berries on the palm shrubs"
[emphasis added] eaten by Jonathan
Dickinson and his fellow shipwreck
victims during their travels along
the east Florida coastline
(Dickinson 1945:70) might have
come from the saw palmetto. The
members of the first Menendez
expedition ate palmetto berries
(Solis de Meras 1964:127), and
Laudonniere (1975:123) mentioned
having eaten the "roots" (probably
young shoots) of palmettos while
in Florida. Charred saw palmetto
seeds have been identified from a
number of archaeological sites on
the Southeastern Coastal Plain,
including the Jungerman Site on
the Indian River just south of
Cape Canaveral (approximately 112
km southeast of Hontoon Island)
(Jordan et al. 1963:17), and at
St. Augustine, 112 km to the
northeast (Reitz and Scarry

May pop is well known as a food
item among southeastern Indians,
particularly during the Mis-
sissippian and historic periods.
Yarnell and Black (1985:99) have
suggested it be placed in a quazi-
cultigen category. It is used as
a fresh fruit, having a citrus-
like pulp, but can be eaten and
processed in a variety of ways
(Peterson 1977:94-95; Spencer
1985). Many of the seeds from
Hontoon Island were charred,

suggesting perhaps that mature
seeds were being expectorated into
the fire during consumption of the
pulp. In the immature state the
seeds are edible.

The last three wild plants--grape
(Vitis spp.), black gum (Nyssa
sylvatica), and spatterdock
(Nuphar lutea), seem to have been
more important during the pre-
historic period, being more
abundant in the snail shell
midden. Spatterdock went from an
average of 54 seeds per column
sample level in the prehistoric
midden, to approximately 5 per
level in the superficial mussel
shell strata. This is a member of
the water lily family (Nympha-
ceae), and both its tubers and
seeds are edible (Peterson 1977:

The use of the "roots" of water
plants by southeastern Indians was
mentioned by Thomas Hariot, and by
Cabeza de Vaca of the Narvaez
expedition. Cabeza de Vaca (1871:
167, 103) recounted the use of
"roots," some walnut-sized, and
some smaller, that were "taken
from under the water" and eaten.
This could apply to spatterdock or
to several other water plants,
including the nut sedge (Cypreus
sp.) mentioned earlier. Hariot
(1951:C4-C5) was more specific,
though again, it is difficult to
determine to which water plant he
was referring. He described six
"rootes," at least two of which
were from wetland or marsh plants:

Openauk are a kind of
roots of round forme,
some of the bignes of
walnuts, some far greater,
which are found in moist
and marish grounds growing
many together one by another
in ropes, or as though they
were fastnened with a string.
Being boiled or sodden they

are very good meate .
Coscushaw it growth
in very muddle pooles and
moist grounds. Being
dressed according to the
country manner, it maketh
a good bread, and also a
good sponemeate, and is
used very much by the inhab-
itants: The juice of this
root is poison, and there-
fore heede must be taken
before any thing be made
there-withall: Either the
rootes must bee first sliced
and dried in the Sunne, or
by the fire, and then being
pounded into floure will
make a good bread .
(Hariot 1951:C4)

The seeds of spatterdock are what
has actually been positively
identified in the archaeological
samples. Interestingly, keeping
in mind Hariot's foregoing
description of the Indians'
methods for preparing rootstock
for consumption, charred rhizomes
have been recovered although they
have not been securely identified
as to which plant they belonged.
If inclusion of spatterdock seeds
in the deposits was a result of
human activity (versus natural
deposition from the marsh), then
we might account for their pres-
ence if they were gathered unin-
tentionally during procurement of
rootstock or, more likely, both
seeds and rootstock were used.

Black gum occurs in wet environ-
ments including bayheads and river
swamps, such as that in which the
midden is located. This tree is a
prolific producer of fruit and
tree rain could account for its
seeds in the samples. There is
nothing to suggest that its fruits
were utilized at Hontoon, but
ethnographic accounts suggest such
evidence would be difficult to
recognize. According to Swanton

(1946:360), the fleshy pericarp
was eaten fresh and the seeds then
ejected, so that only by chance,
would the seeds be subjected to
fire, and in any case, carbon-
ization would not in itself be a
reliable indication that the fruit
was actually used. Since the
fruits are edible and occur in
abundance, it is likely that both
cultural and cultural factors
attributed to their inclusion in
the samples.

Grape was undoubtedly utilized by
prehistoric and historic period
inhabitants of the island, though
more seeds were recovered from
prehistoric levels. Four species
were identified (Table 3)
(Watlington 1984, 1985). At least
four early chroniclers mentioned
grapes (Knight of Elvas 1922:82,
221; Hakluyt 1909:451; Ribaut
1964:72, 95; Harlot 1951:B2,D1).
They were eaten fresh or dried
over a fire for future consumption
(Speck 1909:45). Accordingly, a
number of grape seeds from Hontoon
Island were charred.

Brief mention of a small group of
miscellaneous plant remains will
finish this discussion of the wild
plant species. Many of the sam-
ples from the columns and larger
units contained charred and
uncharred stem fragments of the
switch cane (Arundinaria gigan-
tica). This plant had many uses.
According to Harlot (1951), its
seeds were ground into meal, and
the cane was used for building
material, arrow shafts, fish
weirs, drying racks, basketry, and
as a pipe (Ehrmann 1940:13).

Pine cones and loose scales, char-
red and uncharred, were another
common component of the archaeo-
logical samples. This may indi-
cate the use of pine nuts for
food, or perhaps the cones them-
selves were desired for tinder or

some other purpose. Cones from at
least three species have been
identified (Table 3).

Another unusual component in the
deposits were bracket fungi, three
species of which have been iden-
tified (Table 3). The use of
these fungi at the site, if at
all, is not understood, but they
are known from ethnographies to
have been used as paints, tinder,
hallucinates, and more (Findlay
1982; Ison 1985). One species,
the "sulphur shelf," is edible
(Peterson 1977:238). Shelf fungi
such as these have been identified
from at least two other archaeo-
logical sites in eastern North
America. They occurred at a rock
shelter in southwestern Missouri
(Gilmore 1931:92) and in Archaic
Period deposits from the Windover
Site near Titusville, Brevard
County, Florida (Kimborough 1986).

Also among the miscellaneous mate-
rials were Spanish moss (Tilland-
sia usneoides), and bundles of
fiber and pediole fragments from
the cabbage palm. A small amount
of two-ply cordage was recovered
in the 1985 excavations which
appears to be made from twined
leaf fibers of cabbage palm and/or
a palmetto.

Cultivated Plants

At least three genera of culti-
vated plants have been identified
in the archaeological samples from
Hontoon Island (Table 3). The
seeds and rind fragments from
members of the family Cucurbi-
taceae (pumpkins, squashes, and
gourds) are among the most volumi-
nous plant remains from the site.
More than 1300 seeds and an abun-
dance of rind fragments of Cucur-
bita pepo have been recovered from
the column samples (Table 8). Many
more have come from the larger

excavation units. The other posi-
tively identified cultivated
plants were bottle gourd (Lage-
naria siceraria) and corn (Zea

Several interesting temporal
trends with regard to domesticated
species have been discovered, and
in particular, within the family
Cucurbitaceae (which includes both
C. pepo and Lagenaria siceraria).
From prehistoric levels at the
site, bottle gourd and a small-
seeded variety of C. pepo were
identified. While in the historic
period mussel shell layers, not
only were these earliest two cul-
tivated plants found, but also two
to three more varieties of C.
pepo, corn, and possibly one addi-
tional non-native domesticated
species (apple).

Beginning with the earliest incep-
tion of cultural material (ca.
A.D. 0) in the 1982 column sample

Figure 3.

and contemporary levels of the
larger units, came evidence of the
first two cultivated plants--
bottle gourd and a small orna-
mental gourd-like form of Cucur-
bita pepo (Figure 3). The bottle-
gourd seeds are quite small com-
pared to modern analogues, and
closely resemble seeds from the
Phillips Spring Site in Missouri
which date from around 4300 B.P.
(Kay et al. 1980), and seeds from
Key Marco in Collier County, Flor-
ida (Cushing 1896; Cutler, in
Gilliland 1975:255-256) of roughly
the same time frame as Hontoon
Island. The small pepo, labeled
C. pepo Type One for convenience,
also possesses relatively small
seeds and very thin, smooth-sur-
faced rind. Similar ornamental
gourd-like pepo and bottle gourd
remains have been identified from
other sites in Florida and the
Southeastern United States. The
earliest date to the Late Archaic
and Early Woodland periods (cf.

Bottle Gourd (Lagenaria Siceraria) (top) and Cucurbita
epo Type One (bottom). Small Squares on Grid Equal 2 mm.

Newsom and Decker 1986; Yarnell
and Black 1985; Watson 1985:101-
103; King 1985:76-97).

The small, hard-shelled gourds
have been interpreted as (most
likely) weedy forms, probably more
camp followers or semi-encouraged
plants than fully obligate culti-
gens (Yarnell and Black 1985:99,
104). Their use is believed to
have been more for a container
than for the edible seeds and
bitter flesh (Speck 1941; Heiser
1985:66). This might explain the
occurrence of the great numbers of
uncharred seeds from Hontoon
Island, if indeed, upon maturity
the seeds and pulp were being
separated from the rind and dis-
carded. All of the rind fragments
from Hontoon Island were from
mature fruits, and in addition,
all but one seed were fully
mature. Also, only two seeds (out
of more than 2000) were carbon-
ized. The bottle gourd was pro-
bably also used as a container or
utensil, although its seeds, like
those of the ornamental gourds,
are edible and high in oil content
(King 1985:77). Gourds as con-
tainers were described by several
early explorers including at least
one who visited Florida (Ribaut

The Cucurbita spp. seeds from the
site clustered into four distinct
groups based on size and overall
morphology. Seed measurements and
analyses of the size groups are
reported by Newsom (1986), and
Newsom and Decker (1986).

Seeds from a second form of C.
pepo, referred to as C. pepo Type
Two (Figure 4), were recovered
only from levels dating to the
Contact Period and perhaps some-
what earlier. These seeds are
similar to those of the Type One
cucurbit, but are consistently
larger and have a different width

to length ratio. They have been
classified on the basis of numeri-
cal analysis with cucurbits such
as the scallop and acorn squashes
(see Decker and Newsom 1987;
Newsom and Decker 1986). Seeds
possessing a similar morphology
and dimensions were recovered from
Key Marco (Newsom and Decker
1986). They also compare with a
sample of larger (than the
ovifera-type) seeds from Salts
Cave, Kentucky (refer to King
1985:80 for discussion on similar
finds from the Southeast).

At least one, probably two, addi-
tional C. pepo varieties appeared
at Hontoon Island coincident with,
or slightly later than C. pepo
Type Two and likewise, are found
in the historic period levels
(Figure 4). One of these, desig-
nated C. pepo Type Three, is dis-
tinctive in having a relatively
long, narrow seed. Ninety seven
such seeds have been recovered.
C. pepo Type Four is also large in
size, but much wider than Type
Three. The possibility of their
belonging to the same variety has
not been ruled out since Type Four
could simply represent the extreme
for width. Eleven Type Four seeds
have been recovered. The seeds of
types Three and Four could have
come from a pumpkin-like fruit(s)
since they are similar to the
seeds of modern pumpkins and since
they fit within Decker and
Wilson's (1986) "pumpkin" genetic
class based on numerical analysis
(Newsom and Decker 1986). Large
peduncles (stems), which also fall
within the pumpkin size range,
have been recovered from the same
contexts. Dickinson (1945:69)
observed pumpkins growing in the
vicinity of Cape Canaveral, Flor-
ida, only 64 km southeast of Hon-
toon, during his coastal trek in
1696-1697. Large seeds much like
Type Three from Hontoon Island
were recovered from two late his-



Figure 4. (a

I i f i i i f I I i 1 t I 1 y' "


"* ~^^^ ^H ^^ ^^^^ ~ ~'~^ ^^^ L]



Cucurbita pepo Type One (top) and C. pepo Type Two
ttom); (b) C. pepo Type Three (top) and C. pepo Type
r (bottom). Small Squares on Grid Equal 2 mm.

Table 8. Distribution of Cultivated Plants from the 1982 and 1985 Column Samples.

COLUMN LEVELS 1982 1985 -
5 6 7 8 9 10 11 12 13 14 7 8 9 10

Cucurbita pepo Type
One 141 273 400 98 60 149 101 19 8 11 10 20 6 20

C. pepo Type Two 0 17 0 1 -

C. pepo Type Three 14 -

C. pepo Type Four* -

Lagenaria siceraria 1 1 2 4 1

Zea mays: cob fragment 3
cupule 1 23
kernel 1

? Pyrus malus 1

*C. pepo Type Four was only identified in the larger excavation units.

toric period sites occupied by
Missouri and Osage Indians dating
from the mid-1600s to late 1700s.
These are believed to have
resulted from, or been influenced
by, a Mexican variety of pumpkin/
squash (Blake 1986:6).

No thick "pumpkin-like" rind has
been discovered in association
with these latter two pepo seed
types. This could be a matter of
preservation, but it is unlikely
because bottle gourd rind and the
thin rind (believed associated
with Type One) were present. More
likely, this is a reflection of
the manner in which the Indians
prepared and utilized the
(assumed) fleshier fruit(s).
Pumpkins, "pompions," or squash
were mentioned in virtually every
ethnohistoric account describing
the southeastern Indians (Table
4). The Knight of Elvas, in his
description of the "fruits encoun-
tered in Florida," made reference
to a fruit which might be inter-
preted as similar to modern
"yellow squash": "everywhere in
the country a fruit, the produce
of a plant like liqoacam, that is
propogated by the Indians, having
the appearance of the royal pear,
with an agreeable smell and taste
. ." (1922:221). According to
ethnographic information, pepos
were normally prepared by the
southeastern Indians in two ways.
"Summer" or immature squash (e.g.,
yellow crookneck) was boiled and
eaten in its entirety much as is
done today, while "winter" or
mature varieties (e.g., acorn
squash) were generally cut into
strips, the rind removed, and the
flesh then dried for consumption
at a later date (Swanton 1946:269,
275, 360). The Knight of Elvas
(1922:134) reported on the "roast-
ing" of pumpkins as a preparative
technique. Based on the foregoing
discussion, we would not neces-
sarily expect to recover rind, or

seeds in large quantity, from the
fleshier, squash/pumpkin varie-
ties (as is the case at Hontoon
Island). In contrast to the large
quantities of seed and rind frag-
ments from the small, ovifera-like
Type One cucurbit, which may have
been used as a container rather
than for edible parts, the other
two (three) cucurbits (types Two,
Three, and ? Four) are represented
by much less seed and apparently
no rind or mesocarp.

Another important southeastern
cultivated plant, Zea mays (corn),
was identified in Hontoon Island
samples. Like the larger pumpkin/
squash seeds, it occurred exclu-
sively in the historic period
horizon. Eighteen cobs or cob
fragments were recovered, along
with two kernels and a number of
loose cupules and cupule fragments
(this is for the site as a whole,
only 26.10 g total). All were
thoroughly carbonized, with the
exception of two partially charred
cobs. Less than half of the cobs
were judged adequate on which to
base measurements so that taxo-
nomic placement is tenuous, but a
beginning since little is known
about corn from this region. Half
of the cobs have a row number of
eight, and the other half are ten-
row. The ten-row cobs averaged
13.6 mm in diameter, and the
eight-row cobs averaged 11.8 mm
(all measurements were taken from
cob midsections). Based on cupule
size, kernel morphology and size,
and row number, the Hontoon Island
corn falls within the range for
the variety known as "Eastern
Flint" (Reitz and Scarry 1985:59-
60). A few of the cobs and one
kernel were examined by Margaret
Scarry, who has spent a great deal
of time studying southeastern and
non-local varieties of corn. In
her opinion, the Hontoon Island
samples were typical of Eastern
Flint. Both kernels exhibit the

cresent shape and smooth surface
characterizing this variety.

Corn was very important to south-
eastern Indian subsistence. Like
the cucurbits, it was mentioned in
virtually every ethnohistoric
account (Table 4). Processing
methods included grinding into
meal and flour, and then making
bread or corn cakes (Ranjel 1922:
87; Hariot 1951:C1; Knight of
Elvas 1922:38); the kernels were
sometimes parched and eaten
(Ranjel 1922:96), or processed by
"seething them whole until they be
broken" (Hariot 1951:C); the
"floure" was boiled in water to
make a "pappe" (Hariot 1951:C1);
and it was also dried and stored
for later use (Ranjel 1922:86;
Knight of Elvas 1922:46).

The presence of another possible
cultivated plant in the historic
period samples is interesting. One
thoroughly carbonized seed from a-
plant in the rose family (Rosa-
ceae) came from Level 9 of the
1985 column sample. It is in poor
condition but has been tentatively
identified as cultivated apple
(Pyrus malus), an Old World
species. Since apples were a
likely cargo on ships bound for
Spanish Florida in the sixteenth
through eighteenth centuries, the
presence of a seed at Hontoon
Island is tenable, especially
given the nearness of the coast
and St. Augustine, and later, of
the missions.

In summation, at least five, pos-
sibly as many as seven, domes-
ticated plants were identified.
These are corn, bottle gourd, at
least three forms (possibly four)
of Cucurbita pepo (pumpkin/squash/
gourd), and finally, a tentatively
identified domestic apple. Corn,
bottle gourd, and "pumpkin/squash"
are well known as having been
important components of aboriginal

subsistence complexes throughout
the southeastern United States,
especially during the Missis-
sippian Period (ca. post 900 A.D.)
(Yarnell and Black 1985). Together
with beans (Phaseolus spp.), these
domesticates comprised what has
been designated the "eastern
Mexican crop complex" (Watson
1982). Bottle gourd and a "weedy"
form of Cucurbita pepo seem to
have made the earliest entries
into the east, at least as early
as the late Archaic. Corn, beans,
and possibly other varieties of C.
pepo--"the fleshy cultigen squash"
known ethnohistorically (Yarnell
and Black 1985:99)--appear in the
Southeast at much later dates.

At first glance, the Hontoon
Island data appears to mirror this
sequence of domesticated plant
entry and inferred usage (though
beans are missing) in the South-
east, but certain inconsistencies
have been recognized. As in the
Southeast proper, the small, weedy
cucurbit and bottle gourd were the
first to occur, and they may or
may not have appeared at roughly
the same time as elsewhere.
Following these, corn began to
appear in other areas of the
Southeast, during the Late Middle
Woodland, and became a very impor-
tant crop by Mississippian times.
It continued to dominate the sub-
sistence base during the proto-
historic and historic periods.
Where Hontoon Island differs is
that there is no evidence for corn
until the middle to late historic,
the Mission Period (being found
only in the middle to upper levels
of the mussel shell midden and
overlying shell-free midden). The
same situation was found for corn
at Mount Royal, another long term
village site on the middle St.
Johns River (Jones, personal com-
munication, 1986). In contrast,
corn has been recovered at least
as early as Weeden Island times

from other areas in Florida to the
west (e.g., A.D. 955 + 85, Syca-
more Site, A.D. 1000-1300, Patton
Leslie Site, Curlie Site [Milanich
1974:15; Cutler and Blake 1976]).
This conforms more with the rest
of the Southeast. Beans are known
to have been part of the south-
eastern Indian diet by the Missis-
sippian Period (Yarnell and Black
1985:103), and at least as early
as the late Fort Walton Period
(ca. A.D. 1300-1500) in north-
western Florida (Alexander 1984;
Cutler and Blake 1976). This
again conforms well with evidence
from the core southeast. There-
fore, at Hontoon Island, and by
inference east-central Florida in
general, the absence of corn prior
to the middle or late historic
period (even though it was being
grown in other parts of the state
much earlier), along with the
absence of beans, is perplexing.
Beans though, are difficult to
trace because processing and con-
sumption often entailed their
being made into gruels and eaten

Some possible answers are
suggested. (1) One or both of
these members of the corn-beans-
squash trinity may have been pre-
sent at earlier dates, but were
missed during testing due to the
vagaries of sampling, preserva-
tion, and aboriginal processing
methods; or (2) they were never
present at all; that is, they may
not have been a component of the
subsistence practices employed by
people inhabiting the Middle St.
Johns River Basin. East-central
Florida is both peripheral and
distinctive in terms of environ-
ment and culture from the South-
east as a whole, and even from
other areas of Florida. Cultures
in east-central Florida should not
necessarily be expected to conform
or pursue the same spectra of
resources, especially if the

naturally occurring mix of
exploited plants and animals was
adequate and dependable. This is
certainly true for areas further
south of Hontoon, where the over-
all environment becomes increas-
ingly more dissimilar (Sigler-
Lavelle and Russo 1983:17). Given
the lushness and high productivity
of the natural environment within
the river basin, along with the
nearness of coastal resources,
corn cultivation may not have been
desirable or advantageous. These
people had long since developed a
fine-grained adaptation to the
aquatic and terrestrial resources
of the basin.

The Middle River Basin is bisected
by Larson's (1980:213) inferred
southern limit for agriculture
(just south of Lake George), and
Hontoon Island is approximately 24
km south of this imaginary line.
The soils in the area are gener-
ally acidic and poorly suited to
agriculture, being mostly either
excessively drained, nearly
sterile sands or excessively wet
mucks with high organic content
(U.S.D.A. Soil Conservation Ser-
vice 1980). Indians are not known
to have fertilized their fields,
which would probably have been
necessary to grow corn in this
area on the scale that Indians to
the north and west of the Middle
River Basin did. Even if small
garden plots were maintained, the
chances of producing enough corn
for the maintenance of a large
human population are minimal so
that the investment in time and
energy may not have been war-
rented. It is thus probable that
full scale plant cultivation was
not a particularly feasible econo-
mic choice for prehistoric people
residing in the Middle St. Johns
River Basin. The question still
remains open. Based on ethno-
historic documents (e.g., Fontan-
eda 1854; Dickinson 1945) and the

archaeological record, limited as
it is, it does not appear that
corn was under cultivation.

The cucurbits present a different
picture. These plants, especially
the early varieties, are much
easier to grow, requiring little
or no cultivation. They have
often been described as weeds or
"camp followers" and will volun-
teer or sprout independently in
mulch piles and abandoned gardens
(King 1985:78-79). Unlike corn,
they will tolerate wet, organic
soils and have been observed grow-
ing untended in swamp hummocks
(Stephens, personal communication,
1985). Appending cucurbits to the
aboriginal inventory of utilized
wild plants would have neces-
sitated little effort or change in
subsistence or other activities,
and indeed Cucurbitaceae were
present at least as early as
A.D. 0 at Hontoon Island.

Corn does occur, along with the
larger pepos, at Hontoon in strata
dating to the middle to late
1600s. Certain extraneous factors
may have played a role in this.
One, by this time a great deal of
Indian movement, especially from
the north, had begun. As popu-
lations were decimated by disease,
and balances of power and alli-
ances reorganized, people were
fleeing impact areas, migrating,
and reassociating into new groups
(Milanich and Proctor 1978). If
some of these displaced people
moved up the St. Johns River, they
would have carried ideas about
subsistence, including techniques,
equipment, and even seed stock
from cultivated plants. This
could account for the introduction
of corn and new cucurbit types at

Another possible influence may
have been through the Spanish
missions. By the middle to late

sixteenth century at least two
missions were located close to
Hontoon Island (Newsom 1986). The
missionaries in eastern Florida
were urging the Indians to culti-
vate corn and other produce to
benefit the missions and to cause
the Indians to settle in one place
(making the Indians more acces-
sible and therefore, more easily
Christianized). The missionaries
also encouraged the importation of
foodstuffs from other areas of
Florida, especially the Apalachee
Province (Sturtevant 1962:64). An
established trade of corn to these
missions presents the possibility
that Hontoon Islanders did not
cultivate corn at all, especially
considering the small quantities
recovered archaeologically at
Hontoon and nearby Mount Royal
(Jones, personal communication,
1986). Further, in 1602 the
Freshwater Timucua were described
by the Governor of Florida as
primitives who only subsisted on
fish (and by inference, did not
cultivate plants) (Arnade 1951:

In addition to importing food-
stuffs from within Florida, food
was also being delivered to the
Florida missions and garrisons
from Cuba and Mexico. In evidence
of this, Scarry has identified
exotic New World cultivars from
sixteenth century St. Augustine
(Reitz and Scarry 1985:64). This
might serve as an explanation for
the unusually long, narrow squash
seeds (Type Three) from Hontoon
Island if Blake (1986) is correct
in identifying seeds of such size
and shape with Mexican varieties.

The presence of cultivated plants
at Hontoon Island has raised more
questions than provided answers.
At present we cannot assess how
they might have impacted tradi-
tional foodways--whether they
provided a minor supplement to a

subsistence system based on wild
plants and animals, or if they
might have altered or have been
increasingly incorporated into the
existing subsistence system.


The plant remains occurring in two
column samples from Hontoon Island
have been identified and analyzed.
The resulting information has been
supplemented with data from the
larger, coarse-screened exca-
vation units when appropriate
(e.g., corn remains). Plants were
important to the river-oriented
inhabitants of Hontoon Island and
they undoubtedly supplied a major
portion of the subsistence base.

Marked differences in relative
frequencies and overall species
content exist between the two
column samples. The 1982 column,
which primarily represents the
prehistoric period occupation, is
dominated by wild plants, parti-
cularly marsh/river swamp natives.
Oak and hickory, cabbage palm,
grape, peppervine, black gum,
Polygonum spp., bulrush, nut
sedge, water shield, spatterdock,
Amaranthaceae, and pokeweed are
abundant in this column. Two
domesticated plants also occur in
the prehistoric levels: a small,
thin-shelled gourd (Cucurbita pepo
cf. var. ovifera), and bottle
gourd (Lagenaria siceraria).

Some of the same plants continued
to be prevalent in historic period
deposits, but in general, a dif-
ferent mix of species dominates.
In the 1985 column, successional
or disturbed-habitat species are
more common than the marsh/river
swamp dominants of the first
column. Three additional culti-
vated species appeared in these
levels--corn, and at least two
varieties of pumpkin/squash.

A minimum of eleven wild plants
are believed to have furnished
food during the occupation of the
site. One of the most significant
questions generated from the
investigations at Hontoon Island,
is when and why did corn and the
obligate cultivated pumpkin/
squashes enter the middle St.
Johns region. These plants are
only found in the upper levels at
the site--the middle to late
historic occupation. Their
appearance is accompanied by the
first occurrence of some addi-
tional wild plants (overcup and
swamp chestnut oaks, broomweed,
and purslane), along with other
changes in the plant assemblage.
Overall, both wild and cultivated
plants appear to have been used
with greater intensity in the
historic period and a shift from
emphasis on aquatic/wetland
species to terrestrial plants is
evident. A comparable shift was
observed in the faunal material--
going from 94%-99% aquatics in the
lower levels to only 75% aquatic
vertebrates (by weight of their
remains) in the upper levels (Wing
1986). A change in the size of
fishes caught was also docu-
mented, probably indicating the
adoption or addition of new tech-
nologies or methods of capture.
The shift to terrestrial animals
could at least partially be cor-
related with the expansion of
gardening/horticultural pursuits.
Gardens are known to draw various
animals, resulting in an oppor-
tunistic sort of hunting (see
Linares 1976 for a complete
discussion). The point is that
changes in the subsistence base
and technology were occurring
simultaneously, and on several

The question remains why the
changes occurred at all, and what
provided the impetus for these
changes. Human beings are gen-

erally most resistant to change
where subsistence is concerned,
often relying on a number of
alternate choices or feedback
loops which work to counteract
change (Cowen 1985:226-228;
Flannery 1968). Therefore, change
often results only under very
stressful conditions such as the
failure of a primary resource or
human population expansion
stretching the limits and
potentiality of the environment,
and thus necessitating the
adoption of new techniques to
boost production (Butzer 1982:281-
294, 301-318; Cowen 1985:228-242;
Flannery 1968, 1973; Hassan 1978:

That the environment had ample
food resources appears to be in
evidence at Hontoon Island, based
on the richness of the plant and
animal species in the archae-
ological record. The change in
shellfish could be an indication
of a failure of an important
resource, but enough alternative
providers existed that this should
not have stressed the exploitive
system to the point that a whole
range of alterations were insti-
tuted. A normally reproducing
human population would have under-
gone an increase over the fifteen-
hundred-plus years of occupation
at the site, but this must have
been checked during the historic
period due to the rapid spread of
European disease and other social
factors (cf. Crosby 1972). So,
likewise, population expansion and
its concomitant stresses, does not
appear to be a viable hypothesis
to explain the cultural disrup-
tions or changes which occurred at
Hontoon beginning in the historic

An hypothesis positing an influx
or immigration of people and ideas
is under consideration. It is
reasonable to assume that the

St. Johns Culture-descendant
Freshwater Timucua were still
living at the site, whether
reduced in number or not,
primarily because ceramics
remained unchanged. There were
changes, though, in other cate-
gories of the artifact assemblage,
most notably in the appearance of
new art styles on bone carvings
(Purdy, this volume)--perhaps an
indication that disruptions occur-
red not only on the infrastruc-
tural level, but on other levels
as well (i.e., ideological

If members of other native peoples
were being incorporated into the
group of inhabitants at Hontoon
Island, they would have influenced
or have had the potential to
encourage a redirection of sub-
sistence technologies and the
resource base. Also, people
themselves do not necessarily have
to move to influence change; the
movement of ideas and commodities
through down-the-line exchange and
other forms of trade have also
been cited as primary directives
of culture change (cf. Binford
1972:195-197, 252-258). Another
potential motivator of change was
the Spanish Mission system itself.

In conclusion, we as yet do not
know if Hontoon Island is typical
or atypical of the rest of the
Middle River Basin, or even of the
St. Johns Culture and descendant
Timucua as a whole. That the
people of Hontoon Island seem to
stray from the norm for Timucuan
groups, which relied heavily on
corn-beans-squash agriculture
(Deagan 1978; Milanich 1978), is
already apparent. The Hontoon
Islanders were, perhaps, more like
non-horticultural Indian groups
(e.g., Ais) to the south, who
practiced fully hunting-gathering-
fishing economies. The analysis
of the plant remains from Hontoon

Island, limited as it is, is a
step toward the reconstruction of
human-plant interactions in east-
central Florida and the under-
standing of selective human
behavior in such a resource-rich


I would like to thank David W.
Hall, Brenda J. Sigler-Eisenberg,
Barbara A. Purdy, and Elizabeth S.
Wing, each of whom read pre-
liminary drafts of this paper and
offered suggestions and comments.
Several people assisted in some of
the more difficult identifica-
tions, including David W. Hall of
the Florida State Museum Herba-
rium, Regis B. Miller of the Wood
Products Laboratory, Madison, Wis-
consin, Margaret Scarry of the
State of Florida Bureau of Archae-
ological Research, Gayle Fritz,
Smithsonian Institution, Hugh
Cutler, formerly of the Missouri
Botanical Garden and Washington
University, and Deena S. Decker,
Marie Selby Botanical Gardens,
Sarasota, Florida. To these
people and others not mentioned,
and to the University of Florida,
Department of Anthropology, I am


Alexander, M. M.
1984 Analysis of Plant Materials from
Three Fort Walton Sites: High
Ridge, Velda, and Lake Jackson.
Paper presented at the Forty-First
Annual Meeting of the Southeastern
Archaeologicl Conference, Pensa-
cola, Florida.

Angler, B.
1980 A Field Guide to Edible Wild
Plants. Stackpole Books, Harris-
burg, Pennsylvania.

Arnade, C. W.
1959 Florida on Trial: 1593-1602. An
Eyewitness Account of Life in Flor-
ida Datelined at St. Augustine in
1602. Univ. of Miami Hispanic
American Studies No. 16. Univ. of
Miami Press, Coral Gables.

Asch, D. L. and N. B. Asch
1985 Prehistoric Plant Cultivation in
West-Central Illinois. In Pre-

America, R. I. Ford, ed., pp. 149
204. Anthropological Papers,
Museum of Anthropology, Univ.
of Michigan No. 75. Ann Arbor.

Bender, B.
1975 Farming in Prehistory: From
Hunter-Gatherer to Food Producer.
St. Martin's Press, New York.

Binford, L. R.
1972 An Archaeological Perspective.
Seminar Press, New York.

Blake, L. W.
1986 Cultivated Plant Remains from
Historic Missouri and Osage
Indian Sites. Paper presented at
the Fifty-First Annual Meeting of
the Society for American Archaeo-
logy, New Orleans, Louisiana.

Brinton, D. C.
1859 The Florida Peninsula, its Lit-
erary History, Indian Tribes,
and Antiquities. J. Sabin Co.,

Butzer, K. W.
1982 Archaeology as Human Ecology:
Method and Theory for a Con-
textual Approach. Cambridge Uni-
versity Press, New York.

Cabeza de Vaca, A. N.
1871 Relation of Alvar Nunez Cabeza
de Vaca. Trans. by Buckingham
Smith. J. Munsell, New York.

Callen, E. O.
1965 Food Habits of Some Pre-Columbian
Mexican Indians. Economic
Botany 14:335-343.

Coon, N.
1974 The Dictionary of Useful Plants.
Rodale Press, Rodale, Penn.

Cordell, L. S.
1977 Late Anasazi Farming and Hunting
Strategies: One Example of a Prob-
lem in Congruence. American Anti-
quity 42(3):449-461.

Cowen, C. W.
1978 The Prehistoric Use and Distri-
bution of Maygrass in Eastern
North America: Cultural and Phyto-
geographical Implications. In The
Nature and Status of Ethnobotany,
R. I. Ford, ed., pp. 263-288.
Anthropological Papers, Museum of
Anthropology, Univ. of Michigan
No. 67. Ann Arbor.

1985 Understanding the Evolution of
Plant Husbandry in Eastern North

America: Lessons from Botany,
Ethnography, and Archaeology. In
Prehistoric Food Production in
North America, R. I. Ford, ed.,
pp. 205-244. Anthropological
Papers, Museum of Anthropology,
Univ. of Michigan No. 75.

Crosby, A. W.
1972 The Columbian Exchange: Biolo-
gical and Cultural Consequences
of 1492. Greenwood Press, West-

Cushing, F. H.
1896 Exploration of Ancient Key Dwel-
ler's Remains on the Gulf Coast
of Florida. Proceedings of the
American Philosophical Society
of Philadelphia Volume 35, No. 153.

Cutler, H. C.
1975 Two Kinds of Gourds from Key Marco.
In The Material Culture of Key
Marco, Florida by M. S. Gilliland,
Appendix D, pp. 255-256. University
Presses of Florida, Gainesville.

Cutler, H. C. and L. W. Blake
1976 Plants from Archaeological Sites
East of the Rockies. American
Archaeology Reports No. 1. Uni-
versity microfiche, L. C. #49422.

Deagan, K. A.
1978 Cultures in Transition: Fusion and
Assimilation among the Eastern Tim-
ucua. In Tacachale, Essays on the
Indians of Florida and Southeastern
Georgia During the Historic Period,
J. T. Milanich and S. Proctor, eds.
Florida State Museum Ripley P.
Bullen Monographs in Anthropology
and History No. 1, University
Presses of Florida, Gainesville.

Decker, D. S. and L. A. Newsom
1987 Numerical Analysis of Archaeo-
logical Cucurbit Seeds from Hon-
toon Island, Florida. Paper pre-
sented at the 10th annual confer-
ence of the Society for Echno-
biology. Gainesville, Florida.

Decker, D. S. and H. D. Wilson
1986 Numerical Analysis of Seed Mor-
phology in Cucurbita pepo.
Systematic Botany 11:595-607.

Dickinson, J.
1945 Jonathan Dickinson's Journal or,
[1699] God's Protecting Providence,
E. W. Andrews and C. M. Andrews,
eds., Yale Univ. Press, New Haven.

Ehrmann, W. W.
1940 The Timucua Indians of Sixteenth
Century Florida. Florida Hist.
Quarterly 18:168-174.

Findlay, W. P. K.
1982 Fungi: Folklore, Fiction, and
Fact, Richmond Publishing Co.,
Ltd., Surrey.

Flannery, K. V.
1968 Archaeological Systems Theory and
Early Mesoamerica. In Anthropo-
logical Archaeology in the Ameri-
cas, B. J. Meggers, ed., pp. 67-
87. Anthropological Society of
Washington, Washington, D. C.

1973 The Origins of Agriculture.
Annual Review of Anthropology

Fontaneda, Do d'Escalente
1854 Memoir of Do d'Escalente Fon-
taneda Respecting Florida. Writ-
ten in Spain about the Year 1575.
Trans. by Buckingham Smith. Wash-
ington, D. C.

Gilmore, M. R.
1931 The Vegetal Remains of the Ozark
Bluff-Dweller Culture. Papers of
the Michigan Academy of Science,
Arts, and Letters Vol. 24. Univ.
of Michigan Press, Ann Arbor.

Gould, R. A.
1978 Beyond Analogy in Ethnoarchaeology.
In Explorations in Ethnoarchaeo-
logy, R. I. Gould, ed., pp. 249-
293. Univ. of New Mexico Press,

Gould, R. A. and P. J. Watson
1982 A Dialogue on the Meaning and Use
of Analogy in Ethnoarchaeological
Reasoning. Journal of Anthropo-
logical Archaeology 1:355-381.

Hakluyt, R.
1909 The Principal Navigations, Voy-
ages, Traffiques, and Discoveries
of the English Nation. Made by
Sea or Over-land to the Remote
and Farthest Distant Quarters of
the Earth volumes 8 and 9. Trans.
by Richard Hakluyt of the (1587)
Narrative Notable Historie Con-
taining Foure Voyages Made by
Certain French Captaynes into
Florida. James MacLehose and
Sons, Glasgow, Scotland.

Harlot, T.
1951 A Brief and True Report of the
Newfound Land of Virginia. Fac-
simile edition of the 1588 Quarto
with an introduction by Randolph
G. Adams. The Clements Library
Association, Ann Arbor.

Harlan, J. R. and J. M. J. deWet
1965 Some Thoughts about Weeds.
Economic Botany 18(1):16-24.

Hassan, F. A.
1978 Demographic Archaeology. In
Advances in Archaeological Method
and Theory. Selections from Vol.
1-4, M. B. Shiffer, ed., pp. 225-
281. Academic Press, New York.

Heiser, C. B.
1985 Some Botanical Considerations of
the Early Domesticated Plants

North of Mexico. In Prehistoric
Food Production in North America,
R. I. Ford, ed., pp. 57-72.
Anthropological Papers, Museum of
Anthropology, University of Mich-
igan No. 75. Ann Arbor.

Hitchcock, A. S.
1971 Manual of the Grasses of the
United States. Two volumes.
Dover Pub. Inc., New York.

Hudson, C. F.
1976 The Southeastern Indians. Univ.
of Tennessee Press, Knoxville.

Ison, C. R.
1985 The Role of Fungi in Prehistoric
Subsistence Systems. Paper pre-
sented at the Forty-Second Annual
Meeting of the Southeastern
Archaeological Conference, Bir-
mingham, Alabama.

Jones, C.
1986 Personal Communication, Archaeo-
logist, State of Florida Bureau
of Archaeological Research, Div.
of Archives, History, and Rec-
ords Management, Tallahassee.

Jordan, D. F., E. S. Wing, & A. K. Bullen
1963 Papers on the Jungerman and
Goodman Sites, Florida. Contri-
butions of the Florida State
Museum, Social Sciences No. 10.
Univ. of Florida, Gainesville.

Kay, M., F. B. King, and C. K. Robinson
1980 Cucurbits from Phillips Sping:
New Evidence and Interpreta-
tions. American Antiquity

Kimbrough, J. W.
1986 Personal communication, Botany
Department, Univ. of Florida.
Letter to author dated 10 October

King, F. B.
1985 Early Cultivated Cucurbits in
Eastern North America. In Pre-
historic Food Production in North
America, R. I. Ford, ed., pp. 73-
98. Anthropological Papers,
Museum of Anthropology, Univ. of
Michigan No. 75. Ann Arbor.

Knight of Elvas
1922 Narrative of the Hernando de Soto
Mission of 1539. In Narratives
of the Career of Hernando de
Soto in the Conquest of Florida,
as told by a Knight of Elvas and
in a Relation by Luys Hernandez
de Biedma, Factor of the Expedi-
tion. Together with an Account
of de Soto's Expedition Based on
the Diary of Rodrigo Ranjel, his
Private Secretary. Tran. by
Buckingham Smith (1905) from
Oviedo's Historia General y
Natural de las Indias. American
Explorer Series, New York.

Laessle, A. M.
1942 The Plant Communities of the
Welaka Area, With Special Ref-
erence to Correlations Between
Soils and Vegetational Succes-
Sion, The University Presses of
Florida, Gaine viLi.

Landers, J. L. and A. S. Johnson
1976 Bobwhite and Quail Food Habits
in the Southeastern United States
with a Seed Key to Important
Foods. Miscellaneous Pubs.,
Tall Timbers Research Station,
Tallahassee, Florida.

Larson, L. H.
1980 Aboriginal Subsistence Tech-
nology on the Southeastern
Coastal Plain During the Late
Prehistoric Period. Univ.
Presses of Florida, Gainesville.

Laudonniere, R.
1975 Three Voyages. Originally
published (1586) as L'Histoire
Notable de la Florida, Situes
es Indes Occidentales, Charles
E. Bennett, translator. Univ.
Presses of Florida, Gainesville.

Linares, O. F.
1976 Garden Hunting in the American
Tropics. Human Ecology 4(4):

Martin, A. C. and W. D. Barkley
1961 Seed Identification Manual.
University of California Press,

Milanich, J. T.
1974 Life in a Nineth Century Indian
Household. A Weeden Island Fall-
Winter Site on the Upper Appala-
chicola River, Florida. State of
Florida, Bureau of Historic
Sites and Properties Bulletin No.
4. Tallahassee.

1978 The Western Timucua: Patterns of
Acculturation and Change. In
Tacachale: Essays on the Indians
of Florida and Southeastern Geor-
gia During the Historic Period,
J. T. Milanich and S. Proctor,
eds., pp. 59-88. Florida State
Museum Ripley P. Bullen Mono-
graphs in Anthropology and His-
tory No. 1, Univ. Presses of
Florida, Gainesville.

Milanich, J. T. and C. H. Fairbanks
1980 Florida Archaeology. Academic
Press, New York.

Munson, P. J.
1984 Comments on Some Additional
Species and Summary of Season-
ality. In Experiments and
Observations on Aboriginal Wild
Plant Food Utilization in East-
ern North America, P. J. Munson,
ed., pp. 459-473. Prehistoric
Research Series 6(2). Indiana
Historical Society.

Murray, P. M. and M. C. Sheehan
1984 Prehistoric Polygonum Use in
the Midwestern United States.
In Experiments and Observations
on Aboriginal Wild Plant Food
Utilization in Eastern North
America, P. J. Munson, ed., pp.
282-298. Prehistoric Research
Series 6(2). Indiana Histori-
cal Society.

Newsom, L. A.
1986 Plants, Human Subsistence, and
Environment: A Case Study From
Hontoon Island (8-Vo-202), Flor-
ida. Masters thesis, Dept. of
Anthropology, Univ. of Florida,

Newsom, L. A. and D. S. Decker
1986 Archaeological Cucurbitaceae
from Peninsular Florida. Paper
presented at the 1986 Southeast-
ern-Archaeological Conference,
Nashville, Tennessee.

Panshin, A. J. and C. deZeeuw
1980 Textbook of Wood Technology,
Fourth Edition. McGraw Hill,
New York.

Peterson, F. A. and P. J. Munson
1984 Amaranth as a Food Resource in
the Prehistoric Midwestern
United States. In Experiments
and Observations on Aboriginal
Wild Plant Food Utilization in
Eastern North America, P. J.
Munson, ed., pp. 317-337. Pre-
historic Research Series 6(2).
Indiana Historical Society.

Peterson, L. A.
1977 A Field Guide to Edible Wild
Plants. Houghton Mifflin,

Puckett, C.
1982 Natural Resources, Facilities,
and User Policy and Procedures
at the Welaka Research and Edu-
cation Center. Univ. of Florida,
Institute of Food and Agricul-
tural Sciences, Gainesville.

Ranjel, R.
1922 Narrative of the Hernando de Soto
Mission of 1539. In Narratives
of the Career of Hernando de
Soto in the Conquest of Florida,
as told by a Knight of Elvas and
in a Relation by Luys Hernandez
de Biedma, Factor of the Expedi-
tion. Together with an Account
of de Soto's Expedition Based on
the Diary of Rodrigo Ranjel, his
Private Secretary. Trans. by
Buckingham Smith (1905) from
Oviedo's Historia General y
Natural de las Indias. American
Explorer Series, New York.

Record, S. J. and R. W. Hess
1943 Timbers of the New World. Yale
Univ. Press, New Haven.

1943-1948 Keys to American Woods.
In Tropical Woods various volumes.

Reitz, E. J. and C. M. Scarry
1985 Reconstructing Historic Subsis-
tence with an Example from Six-
teenth Century Spanish Florida.
Society for Historical Archaeo-
logy, Special Publication Series
No. 3.

Ribaut, J.
1964 The Whole and True Discoverye
of Terra Florida. A facsimile
reprint of the London edition of
1563. Univ. Presses of Florida,

Rindos, D.
1984 The Origins of Agriculture.
An Evolutionary Perspective.
Academic Press, New York.

Sargent, C. S.
1965 Manual of the Trees of North
[1905) America. Dover Publications,
Inc., New York.

Sigler-Lavelle, B. J. and M. Russo
1983 Production Strategies within
Florida's Coastal Lowland: Evi-
dence from an Early Formative
Period Household. Paper pre-
sented at the Third Annual Meet-
ing of the Society for Economic
Anthropology, Iowa City, Iowa.

Snedaker, S. C.
1971 Soils-Vegetation Complex of the
Oklawaha Regional Ecosystem. In
Environmental Impact of the Cross-
Florida Barge Canal with Special
Emphasis on the Oklawaha Regional
Ecosystem. Report prepared by
Florida Defenders of the Environ-
ment, Inc., Gainesville.

Solis de Meras, G.
1964 Pedro Menendez de Aviles, Mem-
orial by Gonzalo Solis de Meras.
Translated by Jeannette Thurber
Connor. The University Presses
of Florida, Gainesville.

Speck, F. G.
1909 Ethnology of the Yuchi Indians.
Anthropological Publications,
University Museum, Univ. of Penn-
sylvania Vol. 1(1). Philadelphia.

1941 Gourds of the Southeastern
Indians. The New England Gourd
Society, Boston.

Spencer, R. C.
1985 The Economics of Passiflora Use
in the United States. Paper
presented at the Thirty-Sixth
Annual Meeting of the American
Institute of Biological Sciences,
Gainesville, Florida.

Stephens, J. M.
1985 Personal Communication. Univ.
of Florida, Institute of Food

and Agricultural Sciences, Vege-
table Crops Division.

Sturtevant, W. C.
1962 Spanish-Indian Relations in
Southeastern North America.
Ethnohistory 9(1):41-94.

Swanton, J. R.
1946 The Indians of the Southeastern
United States. Bureau of Amer-
ican Ethnology Bulletin No. 137.
Washington, D.C.

Tutin, T. G., V. H. Heywood, N. A. Burges,
D. M. Moore, D. H. Valentine, S. M. Waters,
D. A. Webb, eds.
1968 Flora Europa. Volume Two. Rosa-
ceae to Umbielliferae. Cambridge
University Press, Cambridge.

United States Department of Agriculture,
Soil Conservation Service
1980 Soil Survey of Volusia County,
Florida. In cooperation with
the University of Florida Insti-
tute of Food and Agricultural
Sciences, Agricultural Experiment
Stations, and Soil Science Dept.

1985 Twenty-Six Ecological Commun-
ities of Florida. Gainesville.

Watlington, F.
1984 Adaptive Viticulture in the
Lower Latitudes of the New
World: A Geographic Synthesis.
Manuscript on file with the Dept.
of Anthropology, Univ. of Florida,

1985 An Archaeological Grape Cultivar
from East-Central Florida. Manu-
script on file with the Dept. of
Anthropology, Univ. of Florida,

Watson, P. J.
1982 The Beginnings of Plant Culti-
vation in the Eastern Woodlands.
Paper presented at the Florida
State Museum, University of
Florida, Gainesville.

1985 The Impact of Early Horticulture
in the Upland Drainages of the
Midwest and Midsouth. In Prehis-
toric Food Production in North
America, R. I. Ford, ed., pp. 99-
148. Anthropological Papers,
Museum of Anthropology, Univ. of
Michigan No. 75. Ann Arbor.

Whitaker, T. A. and G. F. Carter
1946 Critical Notes on the Origin and
Domestication of the Cultivated
Species of Cucurbita. American
Journal of Botany 33:10-15.

Wilder, T. V.
1975 Plant Life in Sixteenth Century,
Florida. In Three Voyages by
Rene Laudonniere, trans. by
Charles D. Bennett, Appendix D.
Univ. Presses of Florida,

Wing, E. S.
1986 Integration of Floral and
Faunal Data of Hontoon Island.
Paper presented at the Ninth
Annual Meeting of the Society
of Ethnobiology, Albiquerque,
New Mexico.

Wunderlln, R. P.
1982 Guide to the Vascular Plants
of Central Florida. Univ.
Presses of Florida, Tampa.

Yarnell, R. A. and M. J. Black
1985 Temporal Trends Indicated by
a Survey of Archaic and Woodland
Plant Remains from Southeastern
North America. Southeastern
Archaeology 4(2):93-106.

Lee Ann Newsom
Department of Anthropology
University of Florida
Gainesville, Fla. 32611


Elise V. LeCompte

This article is a condensation of a paper presented by the author at the Southeastern
Archaeological Conference in Pensacola, Florida, 1984. It was part of a session devoted to the
results of archaeological research at the Hontoon Island site (8-VO-202) in Volusia County,
Florida. However, its results have applicability to the conservation of wood from other
archaeological wet sites.

Wood can survive for long periods of time only under
continuously dry or continuously wet, oxygen-free
conditions. The latter condition exists at many
prehistoric sites in Florida. Although waterlogged
wood looks sound when it is recovered, it is actually
highly degraded and, therefore, requires some form of
preservation to maintain its integrity.

It has been found that the inner structure of
waterlogged wood deteriorates and is replaced by the
surrounding water. Degradation is caused by chemical
processes and biological effects (de Jong 1979:33).
Wood is composed primarily of cellulose,
hemi-cellulose, and lignin. The cellulose within the
wood is decomposed by exposure to oxygen, heat, and
ultra-violet light (Jespersen 1979:69). Both fungi and
bacteria attack and destroy these components. In
addition, water dissolves and hydrolizes other
structural components such as pectines and pentosans
(de Jong 1979:33). Because of this extensive internal
decomposition, wood becomes spongy, soft, and very
fragile. The presence of water, however, maintains the
outer structure of the wood by bonding with the broken
cellulose chains. As waterlogged wood dries out, these
chains begin to bond to each other, thereby drawing the
piece together; this is how shrinkage occurs
(Rosenqvist 1975:12). In addition, the movement of
water molecules out of the drying wood creates
surface tension along the remaining cell walls. The
stress arising from this tension eventually ruptures
the cell walls, resulting in cracking and checking. The
amount of deformation can be so great as to make the
piece unrecognizable. Resoaking the wood will not
restore it to its original size and volume because of
the linkage of the broken cellulose chains and the
additional damage caused by the liquid surface tension
as the water leaves the sample.

Before fieldwork begins at a wetsite, the
archaeologist, in cooperation with a trained
conservator, should plan carefully to ensure the
survivability of the fragile and perishable material
that will be recovered. Organic material must never
be allowed to dry out or else irreparable damage will
occur. At Hontoon Island, we found the easiest way to
remove the wood without damaging it was to use gentle
water pressure from a hose to loosen and to dislodge
material which was then scooped up into a wire tray.
Larger pieces were removed by hand and placed
directly into plastic bags. Smaller pieces were
separated from other material at the screens and
placed in bags. All perishable materials, including

artifacts, were stored in bags which contained a small
amount of water obtained from the square or lagoon.
These bags were then kept in a large vat with several
inches of water in the bottom of it. This prevented
accidental drying if the bags leaked.

Excavated wood cannot be kept in water indefinitely.
Conservators working in England found that they could
store wood in water for only a year because it absorbed
additional water to the point where it began to
deteriorate due to the solvent action of the water
(Murray 1982:14). In addition, this procedure is
space-consuming and the artifacts cannot be put on
display in any satisfactory manner.

Many conservative techniques have been used on wooden
artifacts around Florida including alum, polyvinyl
acetate (PVA), polyethylene glycol (PEG), and
freeze-drying. The idea behind using alum,
PVA, or PEG is to replace the water with a substance
that will support the cell walls and fill cell voids
within the degraded wood, and not evaporate as water
does. Alum, however, is not much more successful than
air-drying. In this procedure, objects are soaked in a
bath of hot, saturated potassium aluminum sulphate
solution (Jespersen 1979:71). The alum replaces the
water within the wood and also penetrates the cell
walls. Upon drying, it crystallizes, a process at first
thought to bulk the remaining intact cell structure, but
later found to damage it further (van der Heide
1979:18). Jespersen (1979:71) actually suggests
retreating any artifacts that have been preserved with

A study carried out at the University of Florida
indicates that the use of PVA solutions is also an
undesirable technique (Dochniak 1982b). This
experiment involved soaking pieces of wood
approximately 3 cm in diameter and I cm thick (all cut
from one large branch piece) in 10%, 25%, and 50%
solutions of PEG 540 Blend and PVA over a month's
time. Several specimens remained in water as
controls. Upon drying, it was found that the pieces
soaked in PVA all lost about 85% of their original
weight and 65% of their original volume. However, the
loss in weight and volume of the pieces soaked in PEG
decreased with increasing concentrations of PEG. A
specimen preserved in a 10% PEG solution lost 85% of
its wet weight and 42% of its volume, whereas the
sample from a 50% solution decreased only 38% in
weight and 7% in volume. One of the control samples
which was removed from the water and allowed to air

Volume 40 Number 1


March, 1987


dry lost 91% of its original wet weight and 71% of its
volume. Figures I and 2 illustrate graphically the
increased efficiency of a 25% and 50% solution of PEG
540 Blend in maintaining the sample's original weight
over these same concentrations of PVA, and also the
improvement when a higher percentage of PEG is used.

As illustrated, the wood in 50% PEG loses much less
weight than the piece in 25% PEG. The probable reason
for the failure of the PVA is that the PVA molecules
are too large to penetrate the wood completely
(Huffman, personal communication 1980). In summary,
preservative solutions of PEG 540 Blend were more
effective than solutions of PVA or air drying, and
higher concentrations of PEG worked better than lower

Based on these preliminary results, a second
experiment was performed in which one of the control
specimens from the first experiment was submitted to
increasingly concentrated solutions from 10-80% PEG
540 Blend over a I-month period (Dochniak 1982a).
Upon drying, this piece retained its original wet weight
and volume, thus showing 100% efficiency for
preservation using a final solution of 80% PEG. The
original samples were reweighed and remeasured on
November 1, 1985, three years after the termination of
the PEG/PVA experiment. Table I clearly shows that
the specimens preserved in PEG have remained stable,
thus providing confidence in PEG as a long-term
preservation method.

Several conservators have found that superior results
can be obtained by using PEG 540 Blend as a bulking
agent instead of chemicals such as acetone rosin,
tertiary butanol, or other molecular weights of PEG
(Barbour 1983:1; Grattan 1982:132-34). We have found
the following procedure most successful in treating
woods from Hontoon Island: immerse the waterlogged
material in increasingly concentrated solutions of PEG
540 Blend from 10-80%, raising the concentration by
10% every month (duration may be increased if
specimens are large). This stepped procedure is
necessary since high initial concentrations draw the
water out faster than it can be replaced by the PEG,
causing excessive shrinkage and cracking. We do not
find it necessary to heat the solution or use biocides;
instead we simply stir it occasionally and skim off
organisms if they are present.

We are presently conducting a long-range study
monitoring the stability of various types and species of
wood conserved with PEG 540 Blend in 1980.
Preliminary results show that the majority of
softwood species, such as pine, cypress, and cedar are
preserved excellently using the method described
above, while almost all of the hardwoods such as oak,
hickory, and persimmon are not. The cellular structure
of hardwood is considerably more complex than that of
softwoods. Although the basic functions of tissues in
hard- and softwoods are the same, they are carried out
by a large number of specialized cell types in the
hardwoods (Wilcox 1973:110), but by only a few kinds of
cells in the softwoods (Lee Newsom 1984: personal
communication). Most of the cell wall material in

hardwoods is contained in tough fibers with small
diameters, narrow lumina, and thick walls (Wilcox
1973:110). The efficiency of a preservation process,
such as soaking in PEG, depends on the permeability of
the wood to liquids. The complex structure in the
hardwoods makes them partially impenetrable to PEG
and, therefore, when the specimen is removed from the
solution, the residual water begins to evaporate and
the wood begins to split, crack, and check.

The technique of freeze drying provides a method which
removes almost all of the water within the wood while
maintaining the original shape and size of the
specimen. Freeze drying is a process whereby liquid is
removed from a substance by first freezing the
specimen and then subliming the ice rapidly under
vacuum. If the pressure and temperature are kept low
enough, the liquid will go directly from a solid to a
vapor without passing through the liquid phase. (This
is what is meant by sublimation.) The vapor is then
carried away from the material and condensed outside
of the freeze drier. This process can be very efficient,
leading to 90-100% removal of the liquid (Flosdorf

Several schedules for freeze drying have been
developed by conservators at other institutions (for
example, R.J. Barbour at the University of Washington,
and David Grattan and Cliff McCawley at the Canadian
Conservation Institute). Following their method, we
begin by soaking the wood in a 20% solution of PEG
1000 for three months (Ambrose 1972; Barbour 1983:1,
1984:1; Smith and Barbour 1983:23). This concentration
of PEG 1000 is employed because it can be frozen more
effectively than can other concentrations or other
molecular weights of PEG. A concentration of 20% puts
enough PEG into the degraded wood to improve its
mechanical properties without making it waxy. There
are several reasons for soaking the wood in PEG prior
to freeze drying. Water expands as it freezes and this
damages the wood. The contraction of PEG during
freezing counteracts the expansion of the water
(Rosenqvist 1975:13). Frozen PEG is soft, not rigid like
ice crystals (Grattan and McCawley 1978:158). In
addition, upon sublimation of the water, the PEG is left
behind as a supportive medium (Hartley and Grosso

Preliminary results from this part of our experiment
show that some of the wood has a tendency to break
apart while soaking in the PEG 1000 and, when removed
from the solution, must be frozen immediately or else
severe cracking results. This suggests that further
experimentation with solution concentrations and
soaking times is needed. After soaking, the material is
placed in plastic bags or foil and frozen at -200C. This
process, as opposed to rapid freezing in dry ice, helps
to prevent cracking (Grattan 1982:127). The frozen
wood can then be stored for long periods without
decay, or freeze dried immediately (Murray 1982:18).
After freezing completely, the wood is placed in the
vacuum chamber of the freeze drier while the
temperature is raised incrementally until sublimation
is complete (Grattan 1982:127; Barbour 1984:1). The

Solution Sample. Drying


Samples Drying



L --------- --13\

4 26 28 30 1438

TIME (Days)

TIME (Days)

Treatment With 25% PEG

Sample A-2

Treatment With 25% PVA Treatment With 50% PEG
Sample B-2 Sample A-3

___ Treatment With 50% PVA
Sample 8-3

Figure 1. Changes in Sample Mass During and After Treatment in
a 25% Solution

Figure 2. Changes in Sample Mass During and After Treatment in
a 50% Solution





88 PEG Group

1981 1985 19

181 1985

A-I 10%
A-2 25%
A-3 50%
C-I 80%
C-3 control

PVA Group
B-I 10%
B-2 25%
B-3 50%
C-2 Control
C-4 Control






0.4 0.3 1.2-1.7 1.1-1.7


0.6 0.8
0.7 0.8
0.8 0.8
0.8 0.8

1.4 1.0
1.7 1.5
1.4 1.4
4.2 4.2

0.7 0.7 1.2 1.1

The original measurements of each sample were: weight 4.5 g, diameter 2.4-2.8 cm, thickness
0.8 cm volume 4.2 cm3- Volume based on diameter and thickness measurements:
V = nrrh = (H)(d/2) (thickness).

Table I. Remeasurement of Samples Used in PEG/PVA Experiment

temperature is then slowly brought up to 20C and held
there until drying is complete (Barbour 1984:1).
Experimentation with freeze drying at the University of
Florida is in progress.

The primary goal of any conservative technique with
waterlogged wood is to maintain the original
dimensions of the wood by preventing deformation due
to shrinking, cracking, and checking. Any method which
accomplishes this can be considered effective and
viable. However, the cost and safety factors of the
various techniques must also be taken into account
when choosing a specific procedure. The final
appearance and texture of the object after completion
of the treatment must also be considered. The most
important challenge is to determine whether or not an
object will be preserved permanently with the method
chosen. Only time will answer that question.

Organic artifacts recovered from archaeological wet
sites constitute a unique and non-renewable heritage.
Archaeologists and conservators, therefore, have a
responsibility to preserve them for future study.
Archaeologists must be prepared to implement the
proper excavation and conservative techniques
necessary when organic material is going to be
recovered. This requires establishing cooperative
relationships with conservators before fieldwork
begins and necessitates requesting enough money in a
proposal budget to cover the high costs of conservation.
In addition, serious
consideration should be given to finding adequate
storage facilities for the artifacts after they are

The prehistoric inhabitants of Florida could not build
monuments and statues of stone because suitable raw
material was not available. Their achievements in wood
should be preserved for future generations to enjoy in
the same way as we enjoy the accomplishments of the
Egyptians, Greeks, and Romans (Purdy 1982:2).

References Cited

Ambrose, W.R.
1972 The Treatment of Swamp Degraded Wood by Freeze-drying.
Report to the International Council of Museums Committee for
Conservation, Madrid.
Barbour, R.J.
1983 Personal communication to Barbara A. Purdy, January 13.
(Letter on file Purdy's office, Department of Anthropology,
University of Florida.)
1984 Personal communication to Barbara A. Purdy, February 9.
(Letter on file Purdy's office, Department of Anthropology,
University of Florida.)
de Jong, J.
1979 The Deterioration of Waterlogged Wood and Its Protection in the
Soil. Conservation of Waterlogged Wood: International Symposium
on the Conservation of Large Objects of WaterloggedI Wood.
pp. 31-40. Government Printing and Publishing Office, The Hague.
Dochniak, Craig C.
1982a The Maxium Efficiency of PEG in Waterlogged Wood
Preservation. Report presented to Dr.Barbara A. Purdy, March 30,
1982. (On file Purdy's office, Department of Anthropology,
University of Florida.)
1982b Polyethylene Glycol Compared to Polyvinyl Acetate in
Waterlogged Wood Preservation. Report presented to Dr. Barbara A.
Purdy, January 30, 1982. (On file Purdy's office, Department of
Anthropology, University of Florida.)
Flosdorf, Earl W.
1949 Freeze-drying: Drying by Sublimation. Reinhold Publishing
Corp., New York.

Weight Diameter
(g) (cm)
19811985 1981 1985


Grattan, D.W.
1982 A Practical Comparative Study of Several Treatments for
Waterlogged Wood. Studies in Conservation 27:124-136.
Grattan, D.W. and J.C. McCawley
1978 The Potential of the Canadian Winter Climate for the
Freeze-drying of Degraded Waterlogged Wood. Studies in
Conservation 23:157-167.
Hartley, Emily and Gerald H. Grosso
1979 An Evaluation of Various Methods of Wet Wood Conservation.
Paper presented at the Society for American Archaeology meeting,
Vancouver, B.C.
Huffman, Jacob
1980 Personal communication. School of Forestry, University of
Florida, Gainesville.
Jespersen, Kirsten
1979 Conservation of Waterlogged Wood by Use of Tertiary Butanol,
PEG, and Freeze-drying. Conservation of Waterlogged Wood:
International Symposium on the Conservation of Large Objects of
Waterlogged Wood, pp. 69-76. Government Printing and Publishing
Office, The Hague.
Murray, Howard
1982 The Conservation of Artifacts from the MARY ROSE. Proceedings of
the ICOM Waterlogged Wood Working Group Conference. David Grattan.
ed., pp. 12-18. The International Council of Museums, Ottawa.
Newsom, Lee Ann
1984 Personal communication. Ms. Newsom is in the doctoral program
in the Department of Anthropology, University of Florida. Her
expertise is paleoethnobotany.
Purdy, Barbara A.
1982 Preservation of Woods from Archaeological Wetsites in Florida.
(Ms. on file Purdy's office, Department of Anthropology, University of
Rosenqvist, A.M.
1975 Experiments on the Conservation of Waterlogged Wood and Leather
by Freeze-drying. Maritime Monographs and Reports No. 16, pp. 9-23.
National Maritime Museum, Greenwich.
Smith, W. Ramsay and R.J. Barbour
1983 A Water Based Dimensional Stabilization Treatment for Highly
Deteriorated Waterlogged Archeological Wood. National Museum Act
grant proposal. (On file Purdy's office, Department of Anthropology,
University of Florida.)
van der Heide, Gerrit
1979 A Piece of History in Conservation of Waterlogged Wood.
Conservation of Waterlogged Wood: International Symposium on the
Conservation of Large Objects of Waterlogged Wood, pp. 17-24.
Government Printing and Publishing Office. The Hague.
Wileex, Wayne W.
1973 Degradation in Relation to Wood Structure. In Wood Deterioration
and Its Prevention by Preservative Treatment, Vol. I, pp. 107-108.
Darrel D. Nicholas, ed. Syracuse University Press, New York.

Elise V. LeCompte
Department of Anthropology
University of Florida
Gainesville, Florida 32611


Florida Bureau of Archaeological Research,
Tallahasssee, 1986. vi + 225 pages,
figures, tables, bibliography. $10 (paper)

This document certainly deserves a better
title than Number 2. At least, Dr. John H.
Hann, whose work is here presented, should
be celebrated somewhere on the cover.

four sections some of Dr. Hann's work:

Translation of the Ecija Voyages of 1605
and 1609 and the Gonzalez Derrotero of 1609

Translation of Governor Rebolledo's 1657
Visitation of Three Florida Provinces and
Related Documents

Church Furnishings, Sacred Vessels and
Vestments Held by the Missions of Florida:
Translation of Two Inventories

Translation of Alonso de Leturiondo's
Memorial to the King of Spain

Each section is of significant value to
anyone who may be interested in any portion
of the Spanish contact period. Each section
is considerably enhanced by the introduc-
tion, explanation and notes of Dr. Hann.
The voyages of Ecija and Gonzalez are very
much easier to follow through the excellent
maps of Charles Poe. Likewise the study of
the visitations of Governor Rebolledo are
aided by Mr. Poe's illustration of the
geography involved, although the mission
locations will certainly be disputed.

The Ecija Voyages

Dr. Hann provides introductory comment to
give perspective before undertaking the
somewhat stilted language of Ecija's trip
logs. Ship's logs haven't changed much
over the years and remain terse, often in
a colloquial shorthand and require a
thorough knowledge of surrounding circum-
stances to arrive at a correect interpre-
tation. Dr. Hann carefully notes the

"surrounding circumstances" as he proceeds
with his translations. He also lets you
know when he encounters difficulties and
why. His most severe comment, however,

The elliptical features of this sentence
make its full thrust difficult for the
translator to resolve. (Note 95)

Although not as harsh as "elliptical," he
sometimes slips in a "muddy" when the ink
has splashed or the copyist has changed
gears. Despite these difficulties, he has,
within the limitations he describes, made
them very readable.

In this first section, Dr. Hann also has
supplemented the Ecija material with trans-
lations of several relevant documents that
bear on the subject of the navigation of
the Atlantic coast from the Bahama Channel
to the bar of Jacan (mouth of Chesapeake
Bay), and the activities of the French and
English in this area during the period.

One need not be an avid student to appre-
ciate these documents. A novelist could
get several good plots out of the Ecija
voyages: the Frenchman who was captured
and taken as a slave to the interior of
the Carolinas just before Ecija arrived in
the area on his 1605 voyage and was later
ransomed back from his Indian captors by
Ecija during the 1609 trip. Or the con-
frontation between the Spanish and an
English ship at the mouth of the Chesa-
peake Bay as they jockeyed back and forth
for position while Ecija and his committee
of officers tried to determine if a shoot-
ing match would be worth the effort to
accomplish their mission. Discretion led
Ecija and his men to return without having
forced the issue and without having actual-
ly seen the new English settlement, reason-
ing that all they really had to do was
locate it, they really didn't have to see
it. This latter is a prime example of how
Ecija survived his more than thirty years
of sailing up and down the southeast
Atlantic coast for the Spanish.

Volume 40 Number 1


March, 1987


Each reader will find something different
to whet the appetite. I was struck by the
Spanish concern that the French might know
something about the interior of what is now
the southeastern part of the United States
that the Spanish didn't. Something that
perhaps Pardo forgot to mention.

Finally, it should be noted that Dr. Hann
continues the struggle with the Spanish
word monte, which he notes "can mean woods
or thicket or hill." Other translators
have had the same experience. Where we run
into this in the interior we tend to sub-
stitute the word "hammock." That may not
necessarily be correct in all instances. I
should think that Dr. Hann's notes would
serve as a textbook on the problems experi-
enced in attempting translations of these
old documents.

Rebolledo's 1657 Visitation

Turning next to the section on the Rebolledo
visitation, practically all students have
covered some phase of this somewhere in some
article, but not like it is here presented.
As Dr. Hann, with his own pick and shovel
well in hand, noted in the April 1986
Florida Historical Quarterly:

Although that (Rebolledo) visitation
record has been mined extensively during
the last two decades for two disserta-
tions and for a number of journal arti-
cles, a valuable aspect of that docu-
ment's contents has not yet received
much attention.

Indeed. This presentation, supplemented as
it is by extensive comment and authority,
should be made mandatory for any student of
Spanish-American history.

With my own little "pick in hand," I note
that Dr. Hann and his map-maker are contin-
uing the controversy concerning the loca-
tion of Azile (San Miguel de Azile mission,
See Map page 82 and Note 3, page 138). Was
it east or west of the Aucilla (Azile or
Asile) River? In Ustega (Yustega) or
Apalachee? And, was the Aucilla River
really the eastern boundary of Apalachee?

Diego Peia in 1716 (Boyd translation) puts
Azile west of the Aucilla. So does B.
Calvin Jones in 1987 (Jones and Shapiro:
"Nine Mission Sites in Apalachee," Society
of Historical Archaeology Annual Meeting,
Savanna). But was Azile town in Apalachee?
As Dr. Hann notes, the literature generally
treats Azile as Timucuan and therefore part
of Ustega. Somewhere I read (Swanton, I
think) about the Creek use of "peacetowns"
along the Chattahoochee. It is just possi-
ble that Azile town was in normally Apala-
chee territory though populated with Uste-
gans, that it was a "peacetown" established
shortly after Prieto's 1608 visit, when
Ivitachuco was the first town west of the
Aucilla. It is possible that Azile was es-
tablished with Ustegans (as sort of hos-
tages) just within the Apalachee territory
to cement the peace just then established
between the formerly warring Ustega and
Apalachee. This would make the answer to
the third question: "Yes," and possibly
explain why, by the time of Rebolledo, the
nephew of an Apalachee chief could be chief
of Asile without it necessarily being an
"Apalachee town."

I have struggled with the place names in
the last paragraph because Dr. Hann empha-
sizes that the correct spelling should be
Azile, not Asile or Assile or Assily, etc.;
Ustega not Yustega and Abalache not Apala-
chee. But none of us could break the habit
completely (See Note mentioned above).

There is such a wealth of information con-
tained in the Rebolledo visitation records
that the "mining" is sure to continue for
some time to come. No doubt those who have
spent all that de Soto money looking for
the mission road knowing that someday the
auditors will discover that de Soto did not
travel up Route 41 and hang a left at U.S.
90, must have seized on Rebolledo's remark

... as the places of this said province
of Timucua are far apart from one another
along crosswise paths and not along the
royal road, ... (p. 104).

Thus, they will know that finding the mis-

sion road, "the royal road," will not help
much. The early missions (which no doubt
were located at the oldest villages, the
towns most likely subject to the earliest
European visits) were on "crosswise paths"
and "far apart" and "not along the royal
road." So it appears that locating the
camino real where some of these missions
were relocated after 1657 will be an exer-
cise in futility as far as the earliest
European (make that de Soto) contact is
concerned. Disease, rebellion, death and
official edict dissipated the clues.

The quoted remarks of Rebolledo both star-
tled and refreshed those in Utina who are
joined in the search; closing, perhaps,
one line of endeavor but opening up other
possibilities. To date Utina has suffered
from the provencialiasm of those employed
in this field. For the most part (happily,
there are exceptions) these "professionals"
have become oriented to their habitant
province, either Potano or Apalachee. It
is refreshing and most exceedingly helpful
to have Dr. Hann turn his attention to the
middle provinces. A few more diggers would
be of considerable help also. Hopefully
there will soon be more than just press
releases to show for these efforts.

Church Furnishings

Hann observes, as have others, that the
1657 visit of Rebolledo was probably set
up to secure testimony favorable to the
governor, his troops and his prior policies
and unfavorable to the priests with respect
to the treatment of the Indians. History
shows that this was an exercise in futility
on the part of Rebolledo, his fate having
already been sealed by other powers. Here
Dr. Hann, with his third contribution on
Church Furnishings, adds balance to the
equation. He points out that which has
been glossed over in most studies. That
is, it appears the priests did very well
economically. Far better than they were
in the practice of telling the king or
maybe even their church superiors for that

Either a substantial part of the testimony

of the Rebolledo visit concerning the
priest's taking the Indian products and
reselling them at considerable profit is
true, or God was truly working miracles
at the missions. Dr. Hann notes that
just the pieces of art averaged twenty
per mission. If you take into account
the plunder taken by the British and
their Indian allies from the various
missions, the amount stated to be buried
and/or burned by the Spanish themselves
as they abandoned the missions and add
it to the amount brought back to St.
Augustine and actually inventoried, you
must conclude that the priests did much
better than heretofore thought and the
raids on the missions could have been as
much for the plunder as for being an ex-
tension of English foreign policy.

Bushnell and others in their studies also
have been led to the conclusion that there
was a lot more wealth in La Florida than
can be accounted for officially. Take a
look at this third section, let your mind
wander around a bit and see.

Leturiondo's Memorial to the King

The last section in Number 2 is Dr. Hann's
translation of Leturiondo's memo to the
King. This one turned out to be not my
cup of tea, but three out of four is not
bad. For those not familiar with this
document it is a very long letter from a
Florida born, St. Augustine priest to his
European king addressing what the priest
considers to be the major problems with
respect to Florida and offering solutions,
or at least suggesting changes, that might
improve the situation as he sees it. Actu-
ally, the entire section including the
introduction, notes and references by Dr.
Hann runs only 59 pages, but it seems to
go on forever. Leturiondo's prose moved
Dr. Hann to his most violet comment in the
entire 225 pages: "turgid." It certainly
is. But so is quantum physics and this
piece contains a lot of information.

Most intriguing is the part about how the
local farmers and ranchers with the assis-
tance of the local government officials


paid a very small percentage of the taxes
or tithes properly due. Admittedly a mat-
ter of interpretation but of vital inter-
est to those who depend on the tithes.
Leturiondo mentions a number of ways that
the situado (subsidy) is misdirected, mis-
used or just mislocated by these same local
Florida officials. The predictions that he
made, if matters were not corrected for the
people of Florida, were remarkably accurate
and he states movingly that

... if a remedy does not come from out-
side, they will perish in the grip of
need and of the tyranny that the
Christians practice against their
brothers and neighbors ...

So it was. So it came to pass.


This volume, Number 2, as was the case with
Number 1, is being distributed from the main
State Library at the R. A. Gray Building in
Tallahassee, Florida to 24 other depository
libraries around rthe state. However, I
have checked the few libraries we have in
Utina and none are on hand. It little mat-
ters, for the excellent job that Jim Miller
has made in this presentation of Dr. Hann's
work will only increase the call for more.
I expect that he and his staff will receive
more than a few requests for Number 1 as
well as Number 2, but they can not laurel-
sit. There must be more.

Reviewed by:
W.S. Eubanks, Jr.
P.O. Box 880
Live Oak, Florida 32060


To order the above publication, please
write to Florida Archaeology, Bureau of
Archaeological Research, Division of
Historical Resources, Department of State,
The Capitol, Tallahassee, Florida 32399-
0250. The cost is $10.00 plus 5% (50C)
sales tax for Florida residents payable by
check or money order in advance. Add $1.00
postage for the first copy, and $0.30 for
each additional copy. A discount is
available to qualified vendors for resale.

Archaeological Geology edited by George
94 Rapp, Jr. and John A,Gifford. 435 pages,
hardbound. Published by Yale University
Press, New Haven and London. ISBN 0-300-
03142-4. Retail price: $35.
Geoarchaeology quarterly journal published
by John Wiley & Sons, 605 Third Avenue,
New York, NY 10158. Subscription price:

Archaeology and Geology have long been
held as distinct subjects in the public
and scientific mind. The geologist and
archaeologist are separated by different
backgrounds and by interests that do not
appear to overlap geology in general
does not apply directly to man. These
differences have long obscured the many
similarities between these fields. Both
involve the retrieval and examination of
objects obtained within the earth, in
order to interpret the past. In recent
years, many techniques and methodologies
commonly used in geology have been refined
to the point that they can be used in very
young strata.

"Archaeological geology" is defined as
"geology pursued with an archaeological
bias or application." "Geoarchaeology" is
archaeology practiced with the aid of
geological techniques and methodology.
The difference between these may be simply
a matter of emphasis, or the background of
the principle investigator. By whatever
name it might be called, the field has
very recently come into its own. At the
most recent national meeting of the Geo-
logical Society of America (November,
1986, San Antonio), there were about
twenty talks (in two sessions) and a number
of poster presentations relating to this

The book Archaeological Geology does a good
job in showing the variety of geological
subjects that have applications to archaeo-
logy. In general, geology can be used in
determining age, provenance, or the envir-
onment of archaeological sites or arti-
facts. Technologies commonly used in
geology, such as geophysical surveying
methods, can also be used in finding or
determining the extent of an archaeological
site. In each of these cases, there are a
number of geological specialties that can
provide information to help solve
archaeological problems.

Perhaps the most important similarity
between geology and archaeology is the
concern with time. In each science the
determination of relative and absolute
time is essential in understanding the
history of the past. Geologists use a
variety of methods to determine age, in-
cluding paleontology, isotopic analysis
and geomagnetism. Of these, only iso tpic
analysis has seen extensive use, as C
dating, in archaeology. Articles can be

found here on the use of palynology
(paleontology of spores and pollen) and
archaeomagnetism. Archaeomagnetism,
according to one article, may have a
potential accuracy of 25 years. In
addition, there are articles on tephro-
chronology (the dating of volcanic ejecta)
and a technique for the radiocarbon dating
of lime mortar.

Determination of the source of material
found at archaeological sites has become an
important part of recent archaeologic work.
Such information can be used to interpret
trading routes or cultural influence. A
variety of geochemical and geological
techniques that have archaeologic appli-
cations are discussed in this book,
including aerosol (soil) analysis, the
mineralogic and petrologic analysis of
ceramic and building materials. There is
also an interesting article in which
isotopic analysis is used to determine the
provenance, and thus the authenticity, of
an ancient marble bust.

There are also articles on geomorphology,
sedimentology, soil analysis, and
their use in interpreting many aspects of
ancient environment. Much more could have
been said about these applications, al-
though that would have required a more
lengthy and expensive text. But of
enormous value to the researcher is a 55-
page selective bibliography, which is in
addition to the references cited in the
articles themselves. Thus, Archeological
Geology makes very large amount of the past
work in this interdisciplinary field avail-
able to the reader. The book is a very
useful reference work, and is highly recom-
mended for research archeologists.

An additional source of information on
this interdisciplinary field is the new
journal Geoarchaeology. This journal
seeks to publish articles from "all
aspects of geology, paleontology, climato-
logy, oceanography, and geochronology
which either include archaeological data
or can be applied to archaeological
studies." The issue reviewed here (vol.
1, no. 3) has articles on geophysical
methods, sedimentation rates, sea level
change and their archaeological
implications. There are also very useful
book reviews and a calendar of upcoming
meetings. The issue has 5 articles and
about 100 pages, although a number of the
pages are blank. The subscription cost,
unfortunately, is more than many people
can justify, but hopefully it will soon be
available at many libraries.

Submitted by:
Kevin McCartney
Department of Geology
Florida State University
Tallahassee, Florida 32306-3026

Volume 40 Number 1


March, 1987


set reproducing "in facsimile over 370
of the most important articles on a num-
ber of topics in Indian studies." Each
Volume Editor provides an introduction
to the articles in the volume and a bibli-
ography for further reading. Many of the
reproduced articles are extremely rare.
The volumes in this series are printed on
acid-free, 250-year-life paper. David
Hurst Thomas is the General Editor for
this series, which is published by Garland
Publishing, Inc., 136 Madison Avenue, New
York, NY 10016. (202) 686-7492.

This is an important series which will be
an asset to any reference library, and will
be of interest to many of our readers. Re-
views of the individual volumes are being
presented in clusters over a series of four
issues of this journal. The first twelve
volumes were reviewed in the last two is-
sues. The review of this series continues
in this issue, and will be completed in the
next issue.

Volume 13. A Plains Archaeology Sourcebook:
Selected Papers of the Nebraska State His-
torical Society, edited with an introduction
by Waldo R. Wedel.

1. Hill, AT, and Paul Cooper. The Archaeological Campaign of 1937 NHM
2. Cooper, Paul L. The Archaeological Exploration of 1938. NHM (1940)
3. Hill, AT, and M.F Kivett. Woodland-Like Manifestations in Nebraska.
NHM (1941).
4. Hill, AT, and George Metcalf A Site of the Dismal River Aspect in
Chase County, Nebraska. NHM (1942).
5. Hill, AT., and Waldo R Wedel. Excavations at the Leary Indian Village
and Burial Site, Richardson County, Nebraska NHM (1936).
464 pages LC 83-47639 ISBN 0-8240-5887-9 S45

The Nebraska State Historical Society was
incorporated in 1878 as a private organi-
zation, and later designated in 1883 as a
state agency. Dr. Wedel provides a brief
overview of this society as a background
setting for the five articles which he
selected for reprinting in this volume.
The five articles are important, relative-
ly little known, works originally publish-
ed by the Nebraska State Historical Society.
I found the text and illustrations in these
reports to be interesting and informative.
They provide essential background informa-
tion to researchers in Nebraska. I recom-
mend the acquisition of this volume.


Volume 14. A Blackfoot Sourcebook: Papers
by Clark Wissler, edited with an introduc-
tion by David Hurst Thomas.

1. Material Culture of the Blackfoot Indians. AMNHAP (1910)
2 The Social Life of the Blackfoot Indians AMNHAP (1911).
3 Societies and Dance Associations of the Blackfoot Indians AMNHAP
4. The Sun Dance of the Blackfoot Indians. AIMN1I.AP (1918)
510 pages LC 83-47645 ISBN 0-8240-58887 $65

This volume reprints four of the six vol-
umes which Clark Wissler prepared for pu-
blication by the American Museum of Natural
History between 1909 and 1918. The two
volumes omitted are the lengthy work on
Blackfoot mythology (Wissler and Duval
1909) and the extensive descriptive work
dealing with Blackfoot ceremonial bundles
(Wissler 1912).

The Volume Editor, David Hurst Thomas,
reports that:

It is important to note the degree to
which Wissler was assisted in his re-
search by David C. Duvall (18779-1911).
Duvall's mother was a full-blood Piegan,
and his father was a French-Canadian
fur trader ...

Wissler engaged Duvall as an interpreter
in 1903. ... He took an increasingly in-
novative role in Wissler's Blackfoot re-
search, ultimately contributing several
hundred pages of manuscript to the effort.

Duvall's untimely death in 1911 abruptly
terminated Wissler's research on Black-
foot medicine rituals ...

As a result of his informant, Wissler's
work is dominated by Piegan Material, with
less treatment on other Blackfoot groups.
Nevertheless, I found these works to be
very interesting and informative, includ-
ing the graphics which provide a greater
depth of understanding than one might ob-
tain from the text alone. I recommend
the acquisition of this volume.

Volume 15. The Dunbar-Allis Letters on the
Pawnee, edited with an introduction by Waldo
R. Wedel.

Volume 40 Number 1


March, 1987


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