@ 200o Florida Anthropological Society Inc. The Florida Anthropological Society Inc. holds all rights to the source text of the Florida Anthropologist and shall be considered the copyright holder for the text and images of these publications.
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.
THE FLORIDA i ..
VOLUME 40 NUMBER 1 MARCH 1987
SOCIETY PUBLICATIONS No. 13
Investigations at Hontoon Island (8Vo202),
An Archaeological Wetsite in Volusia
PUBLISHED BY THE
FLORIDA ANTHROPOLOGICAL SOCIETY, INC.
'' 2 ,
THE FLORID& 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 (Pa~ron) and $200 (life). Foreign subscriptions are an additional $5 US to cover added postage and handling coats 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, notifications 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.
OFFICERS OF THE SOCIETY
PRESIDENT: FIRST VICE PRESIDENT: SECOND VICE PRESIDENT:
Karen Malesky Harold Cardwell Marlene Levy
5312 Bay State Road 1343 Woodbine Street Dept. of Anthropology
Palmetto, FL 33561 Daytona Beach, FL 32014 Florida Atlantic Univ.
Boca Raton, FL 33431
SECRETARY: MEMBERSHIP SECRETARY: TREASURER AND REGISTERED
Elizabeth Horvath John F. Scarry AGENT: Wallace Spears
7404 12th Street N P.O. Box 1013 109 Blais Trail
Tampa, FL 33604 Tallahassee, FL 32302 Daytona Beach, FL 32018
(Three Years): (Two Years): (ome Year):
Ralph Gozlin Jeffrey Mitchem William Goza
7347 Hennessey Road Florida State Museum Florida State Museum
Jacksonville, FL University of Florida University of Florida
32210 Gainesville, FL 32611 Gainesville, FL 33611
EDITOR: EDITORIAL ASSISTANT: EDITORIAL ASSISTANT: PRINTERS:
Louis D. Tesar Joan Deming George Luer Gandy Printers, Inc.
P.O. Box 1013 308 6th St. NE 3222 Old Oak Drive 1800 S Monroe St.
Tallahassee, FL Largo, FL 33540 Sarasota, FL 33579 Tallahassee, FL
James J. Miller William H. Marquardt Morgan H. Crook
Div. of Historical Resources Florida State Museum Dept. of Sociology
Department of State University of Florida and Anthropology
The Capitol Gainesville, FL 32611 West Georgia College
Tallahassee, FL 32399-0250 Carrollton, GA 30118
J k n W. Griffin Glen Doran
Route 5 Box 19 Dept. of Anthropology
St. Augustine, FL 32084 G-24 Bellamy
Florida State University
Tallahassee, FL 32306
NMT: In addition to the above Editorial Board members, the review comments of others knowledgable in a manuscripts subject matter are solicited as part of our peer review process.
THIRTY-NINETH ANNUAL MEETING
FLORIDA ANTHROPOLOGICAL SOCIETY
o 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
Hof 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
I: on Saturday, there will be a Western cookout on the Hilton poolside patio, followed by a special guest speaker. Three concurrent Sunday morning workshops have also been scheduled, from 9
AM till noon.
o Details concerning accomodations, 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 586H2347). Special convention rates of $55/night waterfront, $50
streetside are being honored at the Hilton. To make reservations, 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. Registration 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.
z Hope you can join us!
REGISTRATION AND BANQUET FORM 39TH ANNUAL FAS MEETING (This form may be copied)
Name _ _ _ _ _ _ _ _
______Registration @ $10 each ___0
_____Banquet @ $15 each___TOTAL ____Please make checks payable to the Florida Anthropological
Society. Mail completed form and check to Joan Deming, 308
6th St. NE, Largo, Florida 33540.
TA SEE JAK L 71558 SARASOTA BirJ
TARPONASINOORT GULF OF MEXICO erar
SOCIETY PUBLICATIONS No. 13
THE FLORIOA ANTHROPOLOGIST
VOLUME 40 NUMBER I
Investigations at Hontoon Island (8Vo202),
An Archaeological Wetsite in Volusia
Ed itor'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 |ontoon 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
BOOK REVIEWS, CURRENT RESEARCH, COMMENTS AND EVENTS .............. 90
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 Archaeological 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
2 EDITOR'S PAGE
This is the first issue of my fourth year issues. (I will make this suggestion as a
as the Editor of The Florida Anthropologist, formal motion to our Board of Directors at the journal of the Florida Anthropological our next Annual Meeting on May 8-10, 1987. Society. With this issue we enter our 39th If you have any opinions on this matter,
year of publication. While in past issues please let be know them at or before that
several changes have been introduced to im- meeting). prove the journal's format and broaden its
content, this issue introduces yet another A change which can be made effective this ischange to simplify citation of our journal sue is the elimination of the part one/part
and regularize its quarterly publication, two subdesignation when combined FASP/FA isThis change is to correct the complicated sues are published. Furthermore, when such
numbering system which has occurred over the combined issues are published, the unexpended years. The problem rests with two aspects regular issue funding will be used for the
of our publication -- both concern funding. remaining regular issues, which may thereby be enlarged in content. Thus, all issues,
The first occurred as a result of editorial regardless of size or funding source, will be budget constraints which limited the number issued quarterly and designated as single of pages which could be published to monies numbered issues of The Florida Anthropologist. available from annual membership dues. Thus, This should not cause any problems since it if the Editor received material which he or does not affect membership fees and pricing she believed should be presented in a single for back issues will be based on the relative publication, rather than divided between two cost of producing each issue. or more quarterly issues, to operate within
budget the material was presented as combin- It is with pleasure that I publish Florida ed issues, i.e., FA 22(1-4), 36(1-2) and 39 Anthropological Society Publications Number (1-2). When issues have been combined in 13/The Florida Anthropologist Volume 40 Numthis manner, it has been my experience (and ber 1. This special issue contains reports presumably that of my predecessors) that on the results of work conducted at the Honsome of our institutional subscribers auto- toon Island (8Vo202) site in Volusia County, matically send the Editor notice of non- Florida. The preservation of normally perreceipt for the second quarter when the ex- ishable organic remains at this wetsite gives pected quarterly issue does not arrive, us a unique view of Native American culture.
The second occurs when our occassional ser- Funding for this issue is derived from memberies, Florida Anthropological Society Publi- ship dues, back issues sales, a $400 donation cations (FASP), has been published. Fund- from the Apalachee Anthropological Society of
ing for this series is deposited in our mon- Tallahassee, Florida, and a $1000 donation reograph account and is obtained through do- ceived from the Friends of Anthropology Account, nations, grants and back issues sales. While University of Florida Foundation as a result of the first five issues of this series occur- efforts by Dr. Barbara Purdy. Dr. Purdy is also red as an independent monograph series, be- to be thanked for arranging the preparation of ginning with FASP No. 6/FA 25(2 Pt. 2) in the galley texts for five of the Hontoon Island
1972 the monograph series has had a dual articles; Dr. Wing and Lee Newsom prepared their
designation to indicate that it is a suppli- final galley texts. Dr, Purdy's efforts are ment to our regular quarterly series, The greatfully acknowledged. Phil Pollock and Kevin
Florida Anthropologist. In such combined McCartney are thanked for preparing the final
issues, while their special funding status galleys of their articles. I prepared the rehas been acknowledged through their FASP mailing galleys and the final paste-up.
designation, the pagination has continued
that of the regular FA series. I hope that you enjoy this issue.
A possible solution to the above problem
would be to drop the FASP joint designation Louis D. Tesar
AND to indicate the "Special Issue" status March 5, 1987
of FA issues in which monograph funds are
used to distinguish them from regular FA
Volume 40 Number 1 THE FLORIDA ANTHROPOLOGIST March, 1987
FLORIDA ANTHROPOLOGICAL SOCIETY PUBLICATIONS NUMBER 13
The investigations reported in this volume were funded by the National Endowment for the Humanities, the National Geographic Society, Wenner-Oren 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 Y. LeCompte achieved a high level of excellence in her role as editorial assistant and we commend Louis D. Tesar, Editor of ThFlir1 Anthroplgist, 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 ANTHOPOLOGIST March, 1987
FLORIDA ANTHROPOLOGICAL SOCIETY PUBLICATIONS NUMBER 13
INVISTICATIONS AT HONTOON ISLAND (8-VO-202), AN ARCHAEOLOGICAL
WETSITE IN VOLUSIA COUNTY, FLORIDA: AN OVERVIEW AND CHRONOLOGY
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.
Overview estimated to contain one thousand human burials with
accompanying grave goods, has been excavated by
Lacking the opportunity to observe or participate Clausen et al. (1979). The recently discovered deep peat personally in the cultures they excavate, deposit at Titusville contains a full range of
archaeologists would agree that the next best situation archaeological materials, including wooden artifacts, is to discover sites where all industries are preserved and 7000-8000-year-old human skeletons with brains so that the interrelationships among material items still remaining in the crania (Doran, personal and the representativeness of each to the total way of communication 1984). There are many other sites life can be studied. Such sites occur but, only in areas where wooden artifacts and human skeletal material of continuous dryness or wetness and few have been have been found (MacDonald and Purdy 1982). investigated thoroughly. Analyses of the contents of
wetsites have revealed that less than 10% of material Because of the growing awareness that archaeological culture consists of nonperishable items, i.e., the wetsite components offer a more comprehensive view typical pottery and stone excavated from terrestrial of past cultural activities and because a number of sites. Without the organic fraction, previous views of impressive wooden artifacts have been recovered from the past may be distorted or erroneous and they are Florida's wetlands, usually as a result of development missing information about the environment, diet, and projects, it seems timely to report the results of technologies in wood, fiber, and bone. Even the elusive excavations and analyses from an archaeological functions of stone and shell implements can often be wetsite in Florida that have been ongoing since 1980. determined by studying manufacturing techniques on
wood and bone preserved in wetland deposits. State-owned Hontoon Island is located in Volusia County, Florida midway along the 350-mile (560-km)
Wetsites include the famous Swiss Lake Dwellings (see course of the St. Johns River (Figure I). It is managed by Ruoff 1981), the Maglemosian site at Star Carr (Clark et the Florida Department of Natural Resources, Division al. 1954), the Somerset Levels in southern England of Recreation and Parks. In December 1980, based on where many kilometers of wooden trackways and other information gathered during a statewide survey of wooden remains have been excavated (Coles 1982), the wet- sites, an archaeological research permit was Ozette site on the Olympic Peninsula where a mudslide obtained from the Florida Department of State, Division entombed a village preserving its contents in pristine of Historical Resources and a 3-meter square was condition (Daugherty 1980), the 2500-year-old fishing excavated below the water table in a shallow lagoon camp on the Hoko River (Croes and Blenman 1976), the area adjacent to a large shell midden on the island. A 12,000-year-old Monte Verde site in Chile (Dillehay combination of hydraulic excavation and evacuation 1984:1) and others such as the Bronze and Iron Age bog techniques was used as a primary recovery method for sites in Denmark and England and ancient sites wholly submerged deposits. The unit was excavated to containing wooden artifacts in Australia. a depth greater than 1.5 in below ground surface, I m of which was cultural material and the remaining was
In Florida, the Court of the Pile Dwellers at Key Marco culturally sterile overburden. The screening of all the with its wooden masks, figurines, and utilitarian samples with water and sluicing was an effective way objects was excavated in the 1890s (Cushing 1897; to recover even the most fragile materials. The Gilliland 1975). The Fort Center site on Fisheating organic fraction had survived in excellent condition. Creek near Lake Okeechobee with its large wooden Numerous bone pins and awls, adzed wood, and items sculptures recovered from a charnel pond was suggestive of trade were found. Remains of 48 species investigated throughout the late 1960s and early 1970s of animals and 28 species of plants were identified. (Sears 1982). The slough at the Little Salt Spring site, From the single 3-meter square, 5,777 pieces of bone,
Volume 40 Number 1 THE FLORIDA ANTHROPOLOGIST March, 1987
FLORIDA ANTHROPOLOGICAL SOCIETY PUBLICATIONS NUMBER 13
Figure 1. Hontoon Island, Florida (8-VO-202) (prepared by L.A. Newsom).
3,624 pieces of wood, and 3,265 pottery sherds were Examination of the materials from the test squares, recovered from a matrix of peat and shell. The actual the first column sample, and observations of the trench dynamics of site formation at Hontoon Island are still deposits reveal that the major food items utilized at not understood completely but judging from the vast Hontoon Island, as expected, were aquatic species. Of accumulation of wood chips and a broken canoe paddle, great interest, especially because America will soon be it is virtually certain that canoe making and mooring celebrating the quincentenary of Columbus' voyage, is were taking place; the recovery of numerous cut, the discovery of abrupt changes in diet, woodworking, shaved, and charred posts suggests that simple pile and artifacts around A.D. 1500 including (1) complete structures were built in or near the water edge. Three abandonment of shellfish as the food staple, (2) large wooden carvings were found near Hontoon Island indication that wood may have been cut with metal in the past, an owl (Bullen 1955), pelican, and otter instead of aboriginal marine shell or stone, (3) decline (Figures 2-4). The owl is now located at Fort Caroline in numbers of marine shell tools but increase in other and the latter two figures are on display at the Museum artifact classes, (4) quantum increase in squash of Florida History, Tallahassee. remains and the addition of new varieties, (5)
In Fbrury 982 a scon arhaelogial eserchincreased use of turtle and channel catfishes and Ineruary ob982,e acnd archsqaegia reea decreased diversity of faunal species utilized, (6) permt waslotidand, apoiatl 2-in qurome aprenedoat change in design on some bone pins, (7) appearance of Hoon Islnd aprpreohxim vat w0m aro the previous Spanish objects including a copper coin and Majolica woe Them purpose loch nation warfies ta triee dating to the early 16th century, and (8) European could be radiocarbon dated to determine if the peat ifuneo h hpso oepteyvses matrix was contaminating the wood samples thereby Stratiaraphv and Chronology yielding inaccurate dates. After culturally and
biologically sterile sand was reached at a depth of 1.5 The primary reasons for excavating a 3-in square in m, the profile was straightened and a column sample 1980 were to expose an area large enough to view the divided into 30x30xl0-cm segments was cut from the stratigraphy and to retrieve a representative sample of face extending from the ground surface to 140 cm the fragile artifacts, flora, and fauna for analysis. below surface. Volumetric samples furnish an exact Zone I consisted of approximately 50 cm of culturally representation of each species and, thus, a unique sterile organic overburden followed by 10 cm (Zone 2) opportunity to gain a fuller understanding of the of midden containing copious amounts of pottery, bone, prehistoric economy and the natural environment. With wood, etc., but no shell. Zones 3 and 4 contained all such information it is possible to assess the manner of artifacts and ecofacts in a matrix of shells importance of the use of plants relative to the use of of freshwater species. Zone 5, a sand deposit, animals by prehistoric people. apparently predated human occupation in the area
because artifacts were found only in the poorly defined
Since the processing of the column sample produced transition area between Zones 4 and 5 (Figures 6 and 7). some startling results, a third archaeological research This stratigraphic sequence has been observed in all permit was obtained and a 2-in x 26-in trench with a subsequent excavations at Hontoon Island except that 6-in lateral extension adjacent to the trench (Figure 5) Zone I is thicker farther out in the lagoon and, on the was opened during a University of Florida terrestrial portion of the site, Zones I and 2 are missing,
archaeological field school, Spring 1984 to (1) verify Zone 3 is truncated, and the shell midden proper (Zone finds of previous tests including evidence of extreme 4) is much thicker. modifications that occurred following European contact
in the 16th century, (2) expose the shell mound
stratigraphy to culturally sterile sand to determine the Dr. J. J. Stipp, Director of Alpha Analytic, Inc. and Beta original configuration of the island, (3) document the Analytic, Inc., Coral Gables, Florida and his students at loss of information as one moves from the permanently the University of Miami, Craig Barker and Patricia saturated zone to the dry part of the inidden, (4) obtain Johnson, generously provided free of charge all of the large quantities of wood to study woodworking chronometric dates received for Hontoon Island techniques and preservation requirements, (5) recover beginning with the 2-in square excavated in 1982. wooden carvings similar to those found previously near Table I lists the radiocarbon dates received for Hontoon, (6) obtain and identify floral remains to build Hontoon Island in 1980,1982, and 1984. No dates are a comparative collection such as has been completed available yet for the 1985 excavations. for faunal remains at the Florida State Museum, and (7) Thermoluminescence dates also are shown for 1984. procure volumetric samples from the profile of the From the results of radiocarbon analysis of wood trench and continuous core samples adjacent to the samples from the 1980 excavation, we had a general trench to confirm that the previous column sample, idea of the time period of the site. In 1982 we dated taken in 1982, represents a pattern and not just a samples taken at the contacts between Zones I and 2 fortuitous occurrence from an arbitrarily chosen (60 cm below surface), Zones 2 and 3 (74 cm below location. In May 1985, when the water table was much surface), and Zones 4 and 5 (122-140 cm below lower than during the excavation of the trench in 1984, surface). When the 1984 trench was dug at Hontoon
itwsposbeto excavate partially two additional Island, we made arrangements with Dr. Stipp to date sitos posible trnhadtecxeso.samples by both radiocarbon and thermoluminescence
tz-AP 4 vo-
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).
53 54 55
24 25 261 29 30 31 32 33 34 35 36
0 2 4
I M I e I
Figure 5. Plan View of 1984 Trench at Hontoon Island, Florida (8-VO-202) (prepared by L.A. Newsom).
50 cm b.s.
Zone 2 60 cm b.s.
Zone 3 74 cm b.s.
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 (EIliptio sp.), and (5) culturally sterile sand.
Lab Sample Depth Age Calendar Age
(cm bs) (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
Dicarb 1993 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 74 350+55 400+70 1540
2637 wood 90-100 480+90 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
35 45 pottery 440100 AD 1510 35 45 pottery 420 90 AD 1530 35 45 pottery 400190 AD 1550 34 105 pottery 1310200 AD 64 Viviparus 317050 1220 BC 3003 34 105 pottery 1260190 AD 690 34 105 pottery 1200180 AD 750 34 145 pottery 2040200 90 BC Viviparus 309070 1140 BC 3005 34 145 pottery 1990190 40 BC wood 60040 AD 1350 3009
34 145 pottery 1920 70 AD 30 33 201 pottery 2690240 740 BC charcoal 2120+100 170 BC 3007 33 201 pottery 2600230 650 BC wood 127050 AD 680 3008
33 201 pottery 2500210 550 BC Viviperus 378070 1830 BC 3006 26 130 pottery 1850160 AD 100 Marine shell 202060 70 BC 3001 26 130 pottery 1750150 AD 200 Charcoal 208060 130 BC 3000 26 130 pottery 1700140 AD 250 Viviparus 2950+50 1000 BC 3002
lModified from Barker (1982:22).
Table I. Hontoon Island Chronology
analyses. We were interested (t) in receiving imported species would De expected it mere had teen
additional verification of the date for the contact more recent European influence in the area. Other
between Zone 2 (shell-less cultural zone) and Zone 3 changes that occurred in the contact zone are described
-(shell midden), (2) in determining the earliest in other papers about Hontoon Island (this issue).
occupation of the midden within the area of the trench, Evidence of an earlier Preceramic Archaic occupation
and the differences in the dates depending upon the at Hontoon Island was found in a 2-m square excavated
horizontal position in the trench (i.e., terrestrial or 40 m weot of the trench. mnd ;n im;fA- t; ,
lagoonward), (3) in comparing the results of large shell midden on the south end of the island.
radiocarbon with thermoluminescence, and (4) in documenting the age of the water-saturated zone, i.e., where wood preservation occurs both vertically and References Cited
horizontally. The data are revealing although additional studies are needed because all of the Barker, Craig
objectives were not met and some of the results were 1982 Radiocarbon and Thermoluminescent Analyses of Specimens
from I Hontoon Island. Work conducted under the direction of Dr. inconclusive. J.J. Slipp, Professor of Geology, University of Miami. Ms. on file,
Purdy's office, Department of Anthropology, University of Florida. Thermoluminescence samples taken in square 35 at 45
cm below surface yielded an average date of A.D. 1530. Bullen, Ripley P.
The dates obtained from this depth are significant 1955 Carved Owl Totem, Deland, Florida. The Florida Anthropologist
because they relate to the historic contact period. The 8: 61-73.
organic fraction from the same location was Clark, vatioD s at Star carr. University of Cambridge Press.
insufficient for analysis. Thermoluminescence analysis of pottery from square 34,105 cm below 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 surface, provided an average date of A.D. 700 for the 203:609-613.
location where wood and other plant materials first Coles, John M.
appeared both vertically and horizontally in the profile. 1982 The Somerset Levels: A Waterlogged Landscape. Proceeding f
Unfortunately, radiocarbon analysis of Viviparus sp. ICOM Waterlogged Woodworking Conference, Ottawa, pp 129Ufrortunatleel, radicarbo aal of thefre w atersl Published by International Council of Museums, Ottawa.
from this level, as with all of the freshwater shell Croes, D.R. and Eric Blenman, Editors
analyzed, produced a date far older than expected. 1980 Hoko River: A 2,500-year-old Fishing Camp on the Northwest
Pottery and charcoal collected in square 33 at the Coast of North America. Laboratory of Anthropology, Reports of
contact with the sand horizon, 201 cm below surface, Investigations, No. 58, Washington State Univrsity, Pullman.
produced an average thermoluminescence date of 650 Cushing, Frank Hamilton
B.C. and a radiocarbon date of 170 B.C. The dates 1897 A Preliminary Report on the Exploration of the Ancient
Key Dweller Remains on the Gulf Coast of Florida. Proceedings of received on wood and freshwater snail are not the American Philosophical Society 35(153).
considered accurate (see Table I). From square 26 of Daugherty, Richard D.
the terrestrial portion of the midden, 130 cm below 1980 Wetlands Reseasrch at the Ozette Site. In Florida Mar itim
surface, thermoluminescence dates averaged around Heritage. Proceedings of a conference held in ampa, March 1980.
A.D. 175, radiocarbon analysis of marine shell was 70 Edited by Barbara A. Purdy.
B.C. and charcoal was 130 B.C. Again, snail shell Dillehay, Tom D.
1984 A Late Ice-age Settlement in Southern Chile. Scientific (Viviparus sp.) was considered inaccurate. American 251 (4):106-117.
Doran, Glen H.
The thermoluminescence dates provide a nice sequence 1984 Personal communication. Dr. Doran is the direction of the
and dovetail well with European contact, stone points, Windover Site project and is a faculty member in the Department
and Sof Anthropology, Florida State University, Tallahassee. and St. Johns ceramics (Milanich and FairbanksGilanMroSpu
Gilliland, Marion Spjut
1980:148). The radiocarbon dates on wood from the 1975 The Material Culture of Key Marco, Florida. The
1984 trench are disappointing and suggest that a source University Presses of Florida, Gainesville.
of contamination exists. The freshwater snail shell MacDonald, George F. and Barbara A. Purdy
provides dates consistently older than expected, by at 1982 Florida's Wet Sites: Where the Fragile Past Survives.
least a thousand years. The best correspondence in Early Man 4(4):6-12.
dates between thermoluminescence and radiocarbon are Ruoff, Ulrich
those from charcoal and marine shell. 1981 Die Entwicklung der Unterwasserarchaologie im Kanton
Zurich. Helvetia Archaeologica 45/48-62-70.
Until the chronology at Hontoon Island is better Sears, William H.
defined, we conclude that the portion of the shell 1982 Fort Center: An ArchaeologicalSite in the Lake
Okeechobee Basin. University Presses of Florida, Gainesville. 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 Barbara A. Purdy, Professor
the presence of some Spanish artifacts and all manner Department of Anthropology
of changes that occurred about A.D. 1500. The site was University of Florida
probably abandoned around 1763 when the British took Gainesville, Florida 32611
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
SAMPLING AND EXCAVATION STRATEGIES AT HONTOON ISLAND (8-VO-202) 13
Ray M. McGee and Bruce K. Nodine
The techniques utilized at Hontoon Island (8-VO-202) material below the present watertable without to ensure that the fraction of material recovered and resorting to diving gear, while not original at Hontoon, analyzed is representative of the site included survey, involves techniques not usually utilized by sampling, excavation, microstratigraphy, screening, archaeologists. After logistical problems were solved, processing, recording, preservation, and analyses. Some excavation and recovery proceeded more efficiently. of these procedures are discussed elsewhere. (See Appendix for a description of pumps and problems associated with wetsite excavations.)
The 1980 and 1982 test excavations at Hontoon Island
were conducted below the watertable and a A system to water screen the excavated material was combination of hydraulic excavation and evacuation built by Dr. Paul Gleeson, an archaeologist from the techniques, using pumps, was the primary recovery state of Washington, who had worked for 15 years at method. The 20x26-m trench with 6-m lateral the Ozette Village wetsite on the Olympic Peninsula. extension dug during the 1984 field session, and the The excavated material was dumped from a two additional units excavated in 1985 (Figure I), wheelbarrow onto a hopper, flushed through a 2.5-cm extended from the wholly submerged lagoon edge to the grid which collected large objects, then onto a 1.25-cm top of the terrestrial portion of the shell mound. mesh screen where the soil matrix was rinsed away, and removed to a table where it was sorted into
The location of the trench was selected as a result of categories (Figure 2). This proved to be such an the 1980 and 1982 test excavations and a three-month efficient method that a second water screen was survey of Hontoon Island carried out in 1983 by Nodine. erected so that two squares could be excavated Although the entire 660-ha island was tested at the simultaneously. The spoil from the water screen was time of the survey, the primary intent was to identify dumped into the mined area of the midden west of the water-saturated areas where organic material was trench to prevent contamination of the site. preserved, to ascertain the depth and extent of the
preservation, and to delimit (and thus avoid) areas that The use of 1.25-cm screen can be defended for the had been disturbed by shell mining in the 1930s and the Hontoon Island site. From several levels of one 2-m construction of a picnic area in the 1970s. The results section of the trench, .625-cm mesh was used for of the survey revealed that there are a number of other one-half of the square and 1.25-cm mesh for the other archaeological sites on the island, including large shell half. Upon comparing the results of these two screen middens, but the deposits containing the superb sizes, we learned that we had lost very little preservation of organic materials occur only on the information by 1.25-cm screening because we island's northern perimeter. The survey then recovered, primarily, only smaller fragments of the concentrated on this area and 204 soil samples were same materials in the .625-cm screen. In addition, collected using a bucket auger. From this population, recovery was slowed to a standstill using the smaller 20 samples representing two transects from the lagoon mesh. The use of 1.25-cm screen can be justified also up the shell midden, perpendicular to the contour lines by noting that any size mesh will result in the loss of of the midden, were chosen for analysis. Analysis some information and that is why we opted to take revealed that organic materials were preserved along column samples every 2 meters along the trench to the entire lagoon and extended under the dry part of the obtain a controlled 100% sample for flotation. Figure 3 midden at an elevation of 2.75 m below site datum. shows the location of the column samples and Figure 4 shows the depth of excavation in each square of the
In January 1984, using standard procedure, the trench trench. In addition, we took continuous core samples was laid out perpendicular to the contours of the from several locations adjacent to the trench and into midden as far north as possible without encountering the lagoon, on line with but beyond the trench. The the area of recent disturbance. During excavation of continuous cores permit sampling far deeper than is the trench, all elevations and distances were possible with excavation. Furthermore they can be calibrated to a datum that had been established at the preserved and can furnish a stratigraphic record for time of the 1983 survey. The datum was located at the future reference. Portions of the cores were used to foot of a fire tower. Each 2-m section of the trench obtain samples for radiocarbon dating and for pollen was taken down in arbitrary 25-cm units. Water analysis. excavation was used to recover material from the
submerged deposits, and shovels and trowels were used Microstratigraphy of each 2-m section was drawn by to excavate the portions of the trench that lay above Robin Teas, a crew member, for the south and west the permanently saturated zones. The recovery of -valls of the entire 26-meter trench and extension.
Volume 40 Number 1 THE FLORIDA ANTHROPOLOGIST March, 1987
FLORIDA ANTHROPOLOGICAL SOCIETY PUBLICATIONS NUMBER 13
Hontoon Island 1984 Trench and Test Pit
1 985 Excavation Units 24 N
34 54 35 55
39 59 "0.
Figure I. Hontoon Island Site, Florida (8-VO-202) Showing the Figure 2. Method of Water Screening (Top) at the Hontoon Island Site
1984 Trench and Trench Extension. Squares 39 and (8-VO-202), Florida (1930s mined area can be seen in the
59 were Excavated in 1985. background). Screened Material (Bottom) is Moved to a
Table and Sorted into Categories.
. :/. .. Honto.-)n Island
1984 Trench and Test Pit ' :':1985 Excavation Units
4M THE WATER TABLE
EXCAVATED FIRST 2M THREE LEVELS ONLY
-0. ~ -- -L LEVELS ELW THE
E 1985 EXCAVATIONS
C COLUMN SAMPLES 0]
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 Figure 4. Depth of Excavation in Each Square of
Where One of the Column Samples was the Trench and Trench Extension at
Taken. the Hontoon Island Site (8-VO-202), Florida.
Many "postholes," fire hearths, and shell concentrations The organic deposit at the bottom of the midden in the were noted. The exact vertical and horizontal location western portion of the trench reveals that this part of of preserved plant remains were recorded on the the midden has been continuously wet since the drawings. The plant material would not have survived material was deposited. Since the test unit on the if the watertable had fallen below this point. We west side of the mined area has sand below the shell secured wood from this area for radiocarbon analysis midden, rather than orcianic material, the people living and pottery for thermoluminescence dating (Purdy, this there must have deposited shell, gradually building up issue). the dry sand rise, and eventually expanding the midden
On the east end of the trench (toward the lagoon),toadheslowagnar. organic material was preserved at 2.5 m below site While still in the field, the materials recovered datum, as predicted from the auger samples, and from the trench underwent preliminary sorting, continued until culturally sterile sand was reached at conservation, and study. The pottery was dried,
3.7 m below datum. An auger sample was taken in the placed in major style categories, counted, weighed, westernmost unit of the trench to determine if plant and recorded. The bone was washed, dipped into a remains had been preserved and to define the base of weak solution of polyvinyl acetate (Elmer's glue), and the midden. Organic preservation was encountered at the species identified and recorded. The wood and 2.75 m below datum and sterile sand at 3.5 meters other plant remains were identified and sorted into below datum. Later, fragments of preserved plant categories (nuts, berries, pinecones, fiber, fungi, material were recovered in the western unit (square 25 etc.), counted, weighed, charted, and stored either of the trench; see Figure I for location) during waterlogged or in a preservative to prevent excavations at 2.5 to 2.75 m below datum. This unit degradation (polyethylene glycol [PEG] for the wood was not completed because the trench walls collapsed and formalin acetic acid [FAA] for other plant parts). after several days of heavy rain and there was not The woods, especially the softwoods, have preserved enough time remaining to reexcavate the area. With the well using 60-80% PEG 540 Blend. The wood knowledge that organic material was preserved at the recovered from the test pit in 1980, for example, was west end of the trench, a test unit 42 m west of, but on removed from PEG in 1981 and appears to be in stable line with, the trench, across the mined area, was condition. Experiments are now underway using PEG opened to gain additional information about site along with freeze drying in an effort to solve development and organic preservation. No organic problems that sometimes exist with preserving the preservation was found in this square. Culturally hardwoods. This topic is discussed in another paper sterile sand was encountered at 2.5 m below datum. (LeCompte, this issue) as is the flotation method All levels in the trench contained abundant ceramics, used to separate the material contained in the but no ceramics were recovered in this square. The column samples (Newsom, this issue). only artifacts from this unit were two bone points. RyM ce
Bruce K. Nadine
Department of Anthropology
University of Florida
Gainesville, Florida 32611
WETSITE EXCAVATION STRATEGIES AT IIONTOON ISLAND (840O-202), FLORIDA 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 three types of pumps, to evacuate water from a trench
water-saturated archaeological sites are becoming whose depth in places was more than two meters below increasingly apparent. Associated with these benefits, the water table. (An attempt to use well points in however, are problems of excavating below the water excavating the 1980 test pit failed and was not tried table and of processing large volumes of degraded again for the 1984 trench although we are aware that organic material. well points have been used successfully at other
Each wetsite is unique and so is the problems to be
solved. At Hontoon Island we devised a system, using The largest of the pumps used at Hontoon Island was a
Volume 40 Number THlE FLORIDA ANTHROPOLOGIST March, 1987
FLORIDA ANTHROPOLOGICAL SOCIETY PUBLICATIONS NUMBER 13
Pace brand centrifuge pump with a 7.5-cm hose The intake hoses for the pumps should be rigid in order attached to an 8-hp Briggs and Straton engine, Its not to collapse under the suction from the pumps, and special characteristics make it well suited for high they should be as short as possible. It is difficult for volume evacuation of water. It has nylon impellers the pumps to draw water up over a long distance. It is which are inexpensive to change if there is too much better for the pumps to take water in easily and then grit in the water. It is self-priming and can run push it out over whatever distance is required. The without water. This feature is important because the intake lines should have only enough screening on the pump can be run faster than necessary and, when the ends to keep out rocks, weeds, and other solids. water is down to the desired level, it will idle until the
level rises. Most pumps will burn out if they are The Pace pump can pump about 1710 liters/mmn. It was allowed to run without water. the main drainer of the trench. Its intake was put in a
basket that was placed in the bottom of the lowest
A second type of pump used at Hontoon Island was a unit. The intake hose had a .625-cm mesh around it. mud hog which is a moderately high-volume pump with The output for the main drainer should be short and
7.5-cm hoses and a 5-hp engine. It is a diaphragm wide open so the pump will have no output pressure. pump commonly used to clean septic tanks. It can pump mud or anything else that will fit in the intake. The mudhog was used for solids that accumulated in the bottom of the sump area. As with the Pace pump,
A third type of pump used at Hontoon Island was a the intake of the mudhog should be short and the output Sears centrifuge pump having a 5-cm hose and 1.5 hp wide open. engine. Several of these pumps were used for
excavation and for screening. The Sears pumps were not allowed to pump to capacity.
Hose trees were put on the outputs of these pumps so
All pumps need routine maintenance. Oil must be that the water could be pressurized into garden hoses. changed after every 20 to 40 hours depending on the The inputs were put near a supply of clean water. We type of pump and the conditions of use. If pumps are found it easiest to repump the water from the output of used at different times, it is desirable to keep a log the Pace pump and, therefore, we had to be careful that book to record when pumps are started and shut off. the Sears pump did not run dry when the Pace pump shut When possible, oil should be changed when the engine is off. It is best to run some of the hoses from each warm. Spark plugs for the various kinds of pumps small pump to the excavation area and to the screen should be kept on hand to avoid excavation delays. Oil area because if one pump runs out of gas or breaks may get into the cylinder and gum the plug if the pump down, it is possible to run on other pumps. 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. The effective channeling of water to a pump that can A wire brush and a plug gapper can be used to clean handle it is an important factor in excavating below the spark plugs. If the area is dusty, it is necessary to water table. The strategy was to use square 33 of the have some detergent around to clean the filters and trench as a sump unit and pumping station, and drain recoat them with a little oil. We did not have a all other units to it. This square was excavated, by problem with dust at Hontoon Island. conventional means, to the water table. Using augers,
shovels, and posthole diggers, we then dug a 50-cm
Pumps may operate for several hundred hours without diameter hole in one corner to approximately 75 cm much maintenance. The Sears pumps lasted until the below the excavated floor. The mud hog intake was put last week of the four-month excavation period. It is in the basket in the hole, the unit was drained, and we wise to have back up pumps or access to a good continued excavating with trowel, skim shovel, and repairman. water.
It cost about fifty dollars a week to operate the There were problems with the stability of the walls as pumping system. We found that the pumps used 3.8 the sump unit got deeper. From this experience, we liters of gasoline every 4 to 6 hours depending upon the concluded that the sump unit should be taken down only pump size and the speed of operation. We used about 91 as far as necessary to drain the adjacent units. We liters of gasoline a week. Since water would fill the learned also that, at Hontoon Island, it was more trench rapidly when the pumps shut down, we learned important to prevent the matrix from eroding than to to distinguish by tone when they were about to run out attempt to stop the water. of gasoline and we were ready to fill them
immediately. One soon discovers the most efficient In the beginning we used sand bags but the water and economical speed at which to run the various eroded around them and made the situation worse. Next pumps. we used hardware cloth and metal poles but the
hardware cloth was not rigid enough to permit us toBrcK.Ndn lower it along the wall when the excavation unit got Beprucen of NAdnhoolg deeper. Stainless steel grids used in conjunction with Dprmn fAtrplg 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-in distance was sufficient. We excavated a tostep" 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.
REFUSE DISPOSAL AND MIDDEN FORMATION AT HONTOON ISLAND 19
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 innovations in refuse disposal practices. The cases people's garbage but rarely address refuse disposal as particularly important to investigating refuse disposal a primary focus of investigation. Studies of hunters processes are those in which the disposal techniques and gatherers indicate that these people simply move do not follow changes in settlement pattern, away from their refuse before accumulation reaches suggesting that variables other than settlement, such excess. In contrast, there are many studies of modern as the type of resource exploited or social traditions, urban centers that attest to the vast problems may have a primary influence upon disposal. associated with present-day disposal. These bodies of
literature represent the extremes in refuse behavior, The prehistoric shell midden located at Hontoon Island i.e., moving habitation away from refuse and moving may be considered such an anomalous or "lag behind" refuse away from habitation. The most fruitful arena case because the disposal practices appear to be for producing and testing models of refuse disposal techniques most often used by nomadic people (i.e., behavior lies between these extremes but there have kitchen midden or convenience dumping) while the been few investigations of discard behavior among settlement at Hontoon appears to be sedentary or nonindustrial sedentary people or semi-sedentary semi-sedentary. The inhabitants at Hontoon Island people. exploited a wide variety of resources in their fresh
water riverine environment including the fresh water
In writing about human residues, Gould (1980) snail Viviparus georgianus. addresses and formalizes the quest for predictive Findings indicate that the discard of massive volumes statements concerning a people's discards: of this snail's shell and other materials may have actually produced the land on which the Indians lived.
At the simplest or most basic level... [w]hen and Furthermore, it is probable that at Hontoon Island the
under what conditions do people go to the trouble disposal practices do not reflect merely the local
of removing their rubbish and other discards instead settlement pattern. The need to dispose of incredible
of simply moving away when these residues volumes of non-biodegradable refuse may also have had
accumulate to excess? Perhaps the answer... lies a primary influence on places available for habitation,
in the degree of nucleation and permanence of thus influencing the settlement pattern.
settlement of a community, or perhaps it has more to
do with the nature of resources beino exploited The above assertions were facilitated by using a model
(Gould 1980:114). of midden formation produced by May (1982). I altered
May's model for use on the Hontoon Island midden by
Schiffer (1972:162) postulates that "with increasing adding information supplied by Meehan (1982) who lived site population (or perhaps site size) and increasing with coastal Australians for extended periods during intensity of occupation, there will be a decreasing 1972 and 1973. correspondence between use and discard locations for
all elements used in activities and discarded at the May began his study by searching the HRAF for cultures site." which exploit shellfish. From the HRAF he obtained a
sample of twenty cultures which he used to produce a
Investigations specifically oriented to refuse disposal model for midden formation. He then tested his model (e.g., Hayden and Cannon 1983, Murray 1980) attest to using a trend surface analysis. May's strategy was to the complexity of refuse disposal processes. For predict (from his model) what a processing area would example, samples obtained from the Human Relations look like, making it possible to predict what a discard Area Files (HRAF) indicate the association between area would look like (May 1982:246). May was able to refuse disposal complexity and settlement permanence identify areas, or excavation units which were is statistically significant (Nodine 1986). respectively discard and habitation areas Furthermore, the distribution produced by this (1982:237-238) but he was not able to delimit association shows that innovations occur first in the individual habitation episodes (1982:241). This settlement sphere and are usually followed by situation is understandable for, as the model to be
Volume 40 Number I THE FLORIDA ANTHROPOLOGIST March, 1987
FLORIDA ANTHROPOLOGICAL SOCIETY PUBLICATIONS NUMBER 13
presented below predicts, the processes leading to the other side of the mound and level the "new" area midden formation are very complex, leaving, of course, (May 1982:133-134). The above descriptions from May a very complicated archaeological record, and Waselkov are from Df~ othniograp~ic sources.
The dynamics of midden formation can be best Meehan presents firsthand information:
understood through ethnoarchaeology. Betty Meehan
studied the collecting and discard practices of the There is a predictably complicated pattern for Gidjingali of coastal Australia. From her disposing of food debris in home base camps ....
investigations she produced an excellent book (Meehan Various parts of base camps are continually being 1982). moved, distances ranging from a few metres to
several hundred metres. Debris continues to be
Using new information from Meehan, I altered May's deposited on the periphery of each hearth complex. model for a test on the midden at Hontoon Island. From .. At regular intervals, every week or two, the excavations and core samples, it is evident that the entire camp area is cleaned up with rakes, sticks, or base of the shell midden at Hontoon Island was first feet. The rubbish is dumped in various places around deposited on a flat area with a small rise compared to the periphery of the hearth complex, usually in areas the surrounding swamp. Starting with this that are unimportant for use and access. Eventually,
information, I now present my understanding (to be quite large banks of debris form, and these are tested below) of what people living at Hontoon Island occasionally burnt if a lot of grass is growing on for over 1000 years did about volumes of garbage. them or if they are attracting large numbers of flies (Meehan 1982:114).
When first inhabiting Hontoon Island, the Indians would
discard shell or other refuse anywhere around them. This activity, although systematic, produces a very This would produce a pattern much like the map complicated archaeological record. With the Indians presented by Meehan (1982:116) of a home base shell moving around on the mound, leveling old piles of fishing settlement. Once there was some shell refuse, burning sections and living on top of their deposition, this area would be even more elevated and garbage, the identification of individual habitation desirable than other areas nearby. Soon a small midden episodes is virtually impossible.
woul devlop.May also had problems identifying individual habitation
To emphasize this point, I made a transparency of the episodes (1982:241). He states, however, that map presented by Meehan. By rotating the "interpretation of the distribution of ubiquitous midden
transparency, I began to achieve the effect of the variables resulted in the discovery of habitation and accumulation of repeated depositional patterns. Figure activity areas in the central part of the midden and I is the result of one accumulation. Figure 2 is the discard areas at the periphery" (1982:250). result of two accumulations combined. Figure 3 is the
result of four accumulations. Here, we can see the This is the crux of the model: the processes presented emerging pattern: a midden is beginning to accrue with will produce substantially more refuse on the most of the hearths clustering near the center. Figures periphery and more hearths (activity areas) near the
4 and 5 represent seven and nine accumulations, center. This statement is preceded by May's respectively. If Figure 5 could be produced on the presentation of this configuration, Meehan's discussion ground, I believe most shell midden archaeologists of it in her book, and the exercise utilizing would call it a mound, not a shell scatter, but it was transparencies from Meehan's map. The next step is to produced by an accumulation of shell scatters. As this see how well this configuration is represented in the idealized midden shows, there are more hearths near ground at Hontoon Island. the center and more freely associated refuse near the
edge. It is interesting to note that in the late According to the model, the relative locations of nineteenth century, Walker (1880, 1883, 1885) proposed associated hearth and refuse areas could be in any a similar evolutionary framework for the development configuration at any given time. For this reason, the of mounds. most effective way to test for the modeled processes
is to use information from deep intensive excavations.
Perhaps the Meehan configuration is not so closely This will not bring to light the daily or even yearly followed as the midden accumulates. Perhaps the process but will illuminate the result of over 1500 inhabitants would begin to throw their refuse down the years of depositional processes. slope more often because they could get the garbage
further away with the same or less energy than before, This test will complement May's work which utilized thus adding to convenience or economy of effort of extensive surface information. As mentioned above, disposal (Hayden and Cannon 1983). In this case the May looked for activity areas which helped identify contrast between the central "hearth" area and the discard areas. I will look for discard areas which will outer "refuse" area would be even more pronounced. As help define activity areas. Whereas May used two types one side or area of a midden became too cluttered, the of shell and sandstone fragments, I will use ceramics, people would clear it, leveling the undulating piles of bone, and features as variables to test for the refuse (May 1982:133; Waselkov 1 981:59) or move to predicted con figuarion.
Hearths are show thus
rhe rubb,sh heaops suroundg 111n areas are indicated ths .
and here the middens were dense and up to 0 5 m thick Elsewhere the .
mrdderns became thin with sDarse scatters of shells and other food remains, indicated thus
-, ut" :'.- --2e ----Figure I. Result of One Accumulation of Shell Disposal Figure 2. Result of Two Combined Accumulations of Shell Disposal
(from Meehan 1982). achievedd by rotating a transparency of Meehan's 
.. .. ., e-fj
- ,' -_. ., ..r -rr-r &; .".t"o,, -i-" -,r.
- .... ..,. r
V:, o . --.-:
.-- ..-.a& .- -/ ,frs. ...*4 o
Figure 3. Result of ForCombined Accumulations of Shell Disposal Figure 4. ReutofSenCobedAcmltnsfShlDip al
(achieved by rotating a transparency of Meehan's  (achieved by rotating a transparency of Meehan's  map).
V Ilk 4e,
Figure~~~~~~~~~~~ ~ 5. Reuto ieCmbndAcmlton fSelDsosl(civdb
25157 .13 2.36 .48
26 ~ ~ ~ ~ -,~ 103.8 79631 8
296. 1.4 44 .83 ,6
30 ~ ~ ~ ~ 5. 4.9 61044511
Fiure 5hc R e l f Ninel Combin5)rreedt Accu ua s of ic Sh el sosal (acieed bysh frttin leeas trasarenlucy ofmeerhslp w1982 map). ti enrlyuhefrtee'
Square1.Numberaof Pottery Be Potter Densit y SqoneDest
HOTOSAN 29 8 FEATURES B Y SQUARE
10 0* U-0-V.
CcrChacoa Velcl c eHor ia cnto I'r'na ScS W E~
-------- 15 05 57
1 2 M Figure 7. Features by SquareanElto in the 1984 Trench Excavated at the HnonIln Hoto sadSite (8-VO-202), Florida.
The deep trench excavated at Hontoon Island provides that differential preservation is not a factor here is 2.5 nearly continuous information of a cross section of the further supported by the high correlation of the bone middlen. Even more important to the testing strategy is density to the pottery density: r2 = .94, corr. coef. the nature of the deposits at the lower end of the .97 with 6 degrees of freedom. trench: here the shell and other deposited debris is
mixed with peat. Because of the nature of peat The second factor is slumping, or the oozing effect formation (i.e., it must remain constantly saturated and which may occur on inclined sites (Rick 1976; Rowlett virtually undisturbed) these units have not been dry or and Robbins 1982). Some dlownslope movement disturbed since the deposit began accumulating about probably occurred but the stratigraphy is intact and 2000 years ago (Purdy and Newsom 1985:567). discrete. This can be seen easily when considering the
quantity and condition of the features encountered
This point is significant because it helps set up the during the excavation. Ruling out differential differences between the activity areas and the refuse preservation and gross post-depositional movement, areas at Hontoon Island. Activity areas could be the exponential increase of pottery and bone density anywhere on the dry part of the midden. Areas above when moving towards the lagoon presents a good case the water table should have attributes of both activity for the identification of the periphery of the midden as and refuse while areas which have been constantly the place most often used as refuse areas. saturated should have attributes mostly of refuse Features (hearths, post molds, pits) represent a deposition. discrete location of previous activity (Schiffer
1976:115). According to the model, these items should
I will use three variables to test the predications be far more prevalent on the upper part of the midden. resulting from the model presented above: density of Figure 6 shows the location and types of features pottery, density of bone, and presence/absence of encountered along the trench. This figure depicts the features. I will first use the pottery and bone to define features in their actual elevational locations. The the peripheral refuse area and then bring in features to lowest location of hearths occurs at or near the same show the contrast of activity areas. Moreover, if the elevation in several units. This elevation is about I mn model is accurate, the relative location of feature above the elevation previously mentioned as the lowest areas should move generally away (up and out) from the level of the water for at least the last 1500 years. core of the midden as it accumulates. This type of Surely, the water level fluctuated within certain movement may be necessary for two reasons: (1) bounds and the activity areas were moved accordingly; vertical movement is necessary to stay on top of the it appears, however, that there are no activity areas accumulating refuse, and (2) horizontal movement is below the low water mark. Following the model, the necessary to stay in close proximity to the water as features at the periphery of the midden should be the midden "grows"~ out. younger than those in the core of the midden even if
they are at the same elevation. This should be true if,
The density of pottery in each unit along the trench is for no other reason, hundreds of years of deposition is calculated by dividing the total dry weight of sherds necessary on the periphery before it is dry land on from each unit by that unit's volume. The density of which an activity area could be placed. bone is similarly calculated for each of the eight units.
(Note that the bone used is shattered bone and bone This process can be demonstrated by depicting the debitage, not bone artifacts). I believe that the areas features as distance below surface instead of actual where the density of pottery and bone are highest were elevation (Figure 7). This is the "up and out" pattern more often considered refuse areas. It does not matter discussed earlier. The interpretation of pottery style if these materials represent primary or secondary changes and radiometric dates indicate that the refuse. A broken pot or a splinter of bone may have distance below surface on this section of the midden is been used as convenience tool and may have lain around highly representative of the age of the level. This on an activity area for a while. Eventually, procedures diagram, then, represents the features by level were sometimes taken to move these materials to the coevally. It is apparent that the older features are periphery (Table 1). more to the "core' of the midden. As the model
Usin th poterydenityas te dpenent aribleand predicted, while the midden accumulated vertically and
Usin th poterydenityas te dpenent aribleand horizontally, the features (activity areas) followed a the distance from the top of the trench as the smlrpten independent variable, pottery increases exponentially smlrpten with r2 = .73, corr. coef. =.85 with 6 degrees of Summary
freedom.The results from the Hontoon Island study support the There are several factors which may cause this type of presence of the predicted configuration: refuse areas configuration other than the processes mentioned. The on the periphery and activity areas clustering near the first is differential preservation. If the bone were to center. The presence of this configuration in turn decay more rapidly on the terrestrial part of the increases the validity of the model of midden midden, bone density would increase down the trench. formation. The processes which the model describes In fact, the bone in all areas of the trench was in were confirmed by Meehan using ethnographic excellent condition. Also, the pottery is not subject to information, tested for and confirmed by May using the decay processes which may act on bone. The notion surface extensive information, and tested and
confirmed on the Hontoon Island midden using deep people are permanently settled so their general
intensive excavations. settlement pattern is established. They should, at
some point, begin to use a discrete village dump. Although the model accounts for the midden's general However, the attributes of the refuse itself along with
configuration, there are some aspects concerning the social variables intercede to dictate a disposal
site for which the model offers no prediction. There is no practice different from the expected. The seemingly
conclusive finding to indicate how many people were simple process of throwing away refuse from an
living at Hontoon Island. This is an important point in evening meal becomes a complicated issue.
assessing the "intensity" aspect of the refuse situation. Furthermore, the above confirmed This case study approach to refuse disposal at Hontoon
configuration could have been produced by several Island yields information about the interaction of
types of settlements on top of the midden; e.g., (I) the refuse variables and settlement variables. On the level
people could have lived on one half of the midden and of archaeological method and theory, this approach
thrown their refuse on the other side, or (2) the people brings to light how modeling for past processes can
could have been spread evenly across the midden. help understand past behavior. And on the level of
Although the model does not specifically account for shell midden archaeology, we gain another example of
this difference, from observation of the matrix changes how a shell midden is produced and interpreted.
at Hontoon Island, I believe the latter to be the case. If the former had been the cause of the configuration, one References Cited
would expect more consistent layers in the GodR.A
stratigraphy (i.e., thick layers of refuse above and ,198o Archaeolo. CambridgeUniversiyPress. Cambridge
below thick layers of feature areas). Hayden, Brian and Aubrey Cannon
198: Where the Garbage Goes Refuse D isposa in the Maya Highlands
Journal of Anthropo Archaeo 2:117-163.
The verified configuration also does not confirm or May, J Aldn
1982 Midden Formation Modeling Using Ethnographic and -Archaeolegical
refute the general settlement pattern. (By general Cat: A Trend Surface A asis of MtPdden De s iseato ihe
Carlslon Annis Site (151315),Kentucky. PhD. dissertation,
settlement pattern, I mean the year-by-year Departmen of Anhropology, University of Mio iColumbia.
University Microfilms International, Ann Arbor.
movements of the population.) The statement that the Meehan, Bety
population was permanently settled at Hontoon Island 982t.e. Canberra. ustralianinsfiteoAoriginal
(Purdy and Newsom 1985) comes from faunal (Wing Melos,. Martin V. (ed)
1981 Garbage In the Cities: Refuse Reform and 1he Enirnm n
1984) and floral (Newsom 1984) information. The fact T80-980. TexA&MUeniviyPress Colleg Statonn.
that the population was permanently settled is not Murray. Priscilla
t980 Discard Localion: The Ethnographic Data. 6mrcn A.........g..
integral to the above model: it does not matter if the 980 r Location: TheEthnographicData mrin Aniquiy
described occurrences took place every month Newsom. Lee Ann
1984 Archaeological Plant Remains from Honloon Island. Paper
throughout the year, or just annually when the pe d in the symposium. H ot..n Islan haeological
Wetsite. Southeastern Archaeological Conference, Penscoa
population returned for a season. The disposal strategy Nodine.BruceK.t
remans te sa e. ,985 Society and the universal problem of refuse disposal: an
remains the same. anlhropological analysis. M.A. thesis, Department of Anthropology,
University of Florida. Gainesville
The fact that the population was settled permanently 1985 Setlement Plans, Environmenta Constraints, and Reuse
Patterns. Paper presented at the 501h Annual Meeting of 1he
at Hontoon Island becomes important when discussing Society forAmericanArchaeology.Denver.M
the social dimensions of the refuse strategy. In Pno A AiceoW.chaologalVieofmanSelmeSpace
1981 A Microarchaeological View of Human Settlemen d Space
addition, the population's permanence makes the ndunion n Modernaeria C .. R.A. GouldandMB
Schfler (eds.). pp. 213-224. Academic Press. New York.
Hontoon Island situation interesting from a refuse Purdy, Barbara A. and Lee Ann Newsom
point of view, i.e., permanent population utilizing t" a tifi G o rca ic s Fo5rida Example.
convenience and kitchen midden techniques. This is Rick, John W
when "intensity" comes into the picture. Melosi Analysis. pAmerican Ani Intrasie33Spatial.
(1981:3) states: "it is the modern industrial society, Rowlett, Ralph M. and Michael C. Robbins
. 1982 Estimating Original Assemblage Content to Adjust the not the ancient primitive society, that has experienced Posdepositional Veriical Artifact Movement. World Arcaeol g
the most intense refuse problem." Melosi also makes 14:7i 1383. B.
Schffer, Micbael 8.
the point that one of the major factors which helps 1972 Archaeological Context and SystemicContext. American
deal with refuse problems is that the problem must be 1976 Behavioral Archaeology. Academic Press. New York.
perceived by those affected by it (1981:3). If people Walker, S.T.
are acutely aware of their refuse production and 880 Report n lhe Sei Heapsoan Tpa Bay, Florida. 3n.
Reor of he a S iho an ntiution t879, pp, 413-422,
disposal strategy, the refuse problem will not be 1883 TheAboriginesof Florida. Annual Report of the Smithsonian
intense. I examine this issue at greater length in Instt on 188, PP 67780.
1089 Mounds and SheJ Reaps on the West Cst t of Florida. Annal
Nodine (1986) and discuss models used by Portnoy Rep on one Smhsoan onstiltCom 1pp. 8.4-89.
(1981) and Oetelaar (1985). At Hontoon Island there Waselkov,Gregory
1982 Shellfish Gathering and Shell Midden Archaeology. Ph D.
must have been social problems pertaining to refuse dissertation, University of North Carolina, Chapel HilJ
disposal that were solved eventually by fixed rules. WingES.
1984 Faunal Remains from Honloon Island In Sympostum on
Hontoon Island, an Archaeological Wetsite Southeaslern
The above discussion not only indicates how social Archaeological Conerene. Pensacola.
models can be used in conjunction with the Bruce K. Nodine
archaeological record, but helps clarify many of the Department of Anthropology
aspects of refuse disposal. In the Hontoon Island case, Univeristy of Florida
Gainesville, Florida 32611
HONTOON ISLAND, FLORIDA (8-VO-202): ARTIFACTS
Barbara A. Purdy
Introduction Artifact Categories
The typical and atypical artifacts recovered during A simplified "lumper" approach to the classification of excavations at site 8-VO-202 on Hontoon Island, the artifacts excavated at Hontoon Island is presented Florida and changes that occurred through time in the in Table 1. All numerical and other data pertaining to artifact assemblage, are described and discussed each artifact has been compiled at the University of below. From all excavated units, totalling Florida and is available for inspection and study by
approximately 132 mJ (McGee and Nodine, this issue), interested scholars. we recovered a total of 87 marine shell tools, 119
pieces of stone (most of which were utilized), 17 shark agn teeth, 113 bone tools plus hundreds of bone fragments In addition to the 113 bone artifacts listed in Table I, that could have been utilized, 48 wooden artifacts there were tens of thousands of unmodified or plus thousands of adzed wood chips and naturally splintered food bones recovered at Hontoon Island. deposited wood debitage, 37 historic period artifacts, Much of this discarded scrap, especially the fractured 34 pieces of red or yellow ochre, and more than 70,000 bone, may have served briefly as convenience tools pottery sherds. The overwhelming majority of the leaving little or no evidence of utilization. Thus, the pottery, >99%, is St. Johns paste. The decoration bone artifact count would be increased measurably if consists of predominantly St. Johns Check Stamped in these fragments, lacking formal shapes and sustained the upper levels. Plain sherds occur in various usage, had been detected. The bone artifacts have not proportions throughout but are dominant in the lower been studied minutely for methods of manufacture but levels (St. Johns 1) (Goggin 1952; Milanich and a few comments can be made. Fairbanks 1980).
The bone artifacts classified as pins were highly
Bone artifacts were made primarily from deer and polished, symmetrical, and often had some carving or varied considerably in size. Some of the bone artifacts incising at the proximal end. The utilitarian objects, were very delicate while others were sturdier and on the other hand, were dull and asymmetrical with no might have functioned as weapons. We also recovered decoration. The five ulna awls are a good example. two bone beads and three bone pendants, two of which Polish on the decorative objects could have been were manufactured from deer scapula. Shell achieved by rubbing with sandstone, shark skin, or
implements, made from marine clams or conchs, through use. Some of the pins are so highly polished included scrapers, a drinking cup, and a variety of other that no other manufacturing marks are still visible. tools and utensils. Wooden artifacts consisted of a Most of the objects, however, retain some striations fire starter, an atlatl or notched stick, a wedge, a possibly made with a shark tooth, a flint flake, or a number of burned "posts," small pointed objects sandstone abrader. perhaps pegs or projectiles, and other objects
unidentifiable as to function. The most common stone Two methods appear to have been used at Hontoon point was Pinellas which fits in well with the pottery Island to manufacture bone pins, and both were utilized chronology and the radiocarbon dates received for the throughout the time represented in our excavations, site. A Duval point was recovered in Level 10 of one that is, from about A.D. 0 to A.D. 1760. In one technique 2-in section of the 1984 trench which verifies the the bone was split lengthwise, then the outer surfaces increasing antiquity with depth. Duval should predate were tapered and polished but the channel was not Pinellas according to already established removed. These pins are very uniform in size and differ
classifications (Bullen 1975:13). We found a stone primarily from one another in that one has a hole plummet or net weight in a permanently saturated drilled in the middle of the top of the head (possibly to level of the trench. It still retains pitch or resin attach something such as a feather) and another is permitting an examination of the kind of cordage that slightly incised (Figure 1). One of the pins in this bound it at the neck. The plummet, a metate made of category is fossilized and might have been an old coquina, a mano, and several broken or fragmentary artifact that was found and brought to the site. These stone celts were all manufactured of nonlocal stone. pins are highly polished but the polish on some of them There were numerous pieces of sandstone recovered, has eroded partially leaving blotchy dull areas. Upon many of which had grooves or otherwise showed signs examination of these artifacts, it is very apparent that of use. Two pieces of sandstone may have been this situation did not result from differential "sculpted" into art forms although their shapes may be polishing. These are called expanding head pins in the fortuitous. literature (Goggin 1952).
Volume 40 Number 1 THE FLORIDA ANTHROPOLOGIST March, 1987
FLORIDA ANTHROPOLOGICAL SOCIETY PUBLICATIONS NUMBER 13
Stone I 2 3 4 5 6 7 8 9 10 11 12 13 14 15 No Wood I 2 3 4 5 6 7 8 9 10 11 12 13 14 15 No
Flint Cypress (I~odiulosp.)
Point I I 13 13 7 2 1 1 1 2 Shaft I
Utilized flake 4 6 6 3 4 1 1 I I Firestarter
Waste flake I I Plank fragment
Tool 1 1 I 2 1 Impressed design
Heat exploded 2 Point 1 1 2
Ounflint2 I Post
Sandstone abrader3 1 6 2 1 1 2 Unknown function I 2 1 1
Celt frags & hammers2 3 3 I 1 Pine ( i i sp.)
Coquina metate I Shaft
Limestone plummet2 I Lighterknot 3
Ochre 3 10 11 1 6 2 I Atletl(?)
Post 1 2
Willow (Salix Gcal1iionae)
Bone 1 2 3 4 5 6 7 8 9 10 11 12 13 14 5I No Peo. cut
Provenienc Ash (fra ius.sp)
Bay (prsc sp)
Pins 4 11 7 8 8 1 3 5 4 I Staggerbush(Lyoni sp)
Beads 2 Plug(?)
Pendants 3 Function unknown
Utilitarian 3 10 6 6 6 3 2 3 1 1 Oak (Qucussp.)
Exotic/fossil I 2 I Bowl preform
Miscellaneous 1 2 2 2 1 Elm (Ulmuams)rican)
Shark teeth Post
Drilled I I I 1 Button bush (Ceoo.luthju
Used I c rJI Its)
No alteration deected 5 3 1 2 1 Plug(?)
Cedar (JuniDu sp)
Shell 1 2 3 4 5 5 7 8 9 10 11 12 13 14 IS No Firestarter
Provenience Persimmon (DinsyCOS
Marine clam Wedge
Scraper4 3 3 2 1 1 Unidentified hardwood with
Unknown function4 I 4 I I 1 incised lines
Unused5 Species & function unknown 4 1 2 3
Decorative 1 Two points made of silicified coral were found, one in level 3 end the other in level 4 Both weste
Pendant flakes were silicified coral
Bead 2Non-Floridastone Ounflint listedalsowithhistoric artifacts.
Utilitarian 3The sandstone abrader found in level 7 is very smooth and the shape resembles an animal
Adze 3 I 1 4 I 2 2 1 2 1 4Someare osndisu
Hammer 1 5 Includes very small whole marine shells like Crithium abuneum that are sometimes found with a
Oouge 2 I 2 3 I I drillholeat sitesinFlorida
Unknown function 2 2 I 5 3 1 1 1 2
Fragment 1 2
Oyster (unused) I I
Species & function not
known 1 1 4 2
Historic 12 3 4 5 6 7 8 9 10 1i 12 13 14 15 No
Copper 1 2
Iron 8 13
Majolica 5 I
Glass beod 1
Table 1. Nonceramic Artifacts from Hontoon Island
The other method was to cut out a section of the long be exerted in its manufacture. If, however, the bone or bone and then shape it until it was perfectly round in stone object is a tool to make a tool, e.g., an awl or a mid-cross section with no channel remaining. As an scraper, then the main concern of its manufacture will alternative, after the bone was split, the channel may be its ability to perform a specified function. have been removed by extensive sanding or by some
other method. We have only one artifact that shows Bone artifacts were recovered from every excavation how the long bone was cut to remove a portion of it for unit and to a depth greater than 3 m. As shown in a pin. The proximal ends of these pins tended to have Table 1, significantly larger numbers of bone artifacts more incising than the first type and often they have occur in the upper levels. This situation might be peg tops of various shapes (Figure 2) probably for the explained by an increase in population, a change in attachment of decorations. Some of the incising could activity, or both. In the last section of this paper, I have been functional to provide purchase for a line tied explore these possibilities further. around the circumference.
A third method, of which there is only one example,
was used to make a "dagger," which was classified as a Forty-one of the 81 chert artifacts found at Hontoon utilitarian object (Figure 3). Here the leg bone has not Island were points, and 33 of the remaining 40 chert been split but has been modified into a very effective artifacts were utilized flakes or tools. This situation
weapon.suggests that (1) chert tools were not being
weapon.manufactured at the site, (2) chert was not readily
A common technique of manufacture was observed on available, and (3) materials other than chert, such as many utilitarian bone artifacts, especially those that bone or marine shell, performed tasks that usually probably functioned as awls, fid, perforators, or required chert. A far-fetched additional explanation is gravers. Durable but very sharp distal ends were that few tasks were performed at the site that formed by cutting at a dihedral angle (Figure 4). The required sharp, durable materials. That chert was hard result is nearly identical to what has been described to come by is demonstrated by the fact that in nearly for stone tools shaped by directing a burin blow to all levels, some stone had been reworked because old either side of the tip. This same technique was used weathered surface had been partially removed by also in shaping the pointed ends of many of the bone reflaking. Of the remaining 60 pieces of stone, ochre pins but evidence of it was largely obliterated by and sandstone represent the largest categories, subsequent finishing, totalling 48. The only metate recovered from the site
was made of coquina.
The ends of some small bone artifacts are flat and
rounded like little spatulas. One of the bone pins has The greenstone celt fragments, the coquina metate, the uniform rotary marks encircling the proximal end for a Plummet made of non-Florida limestone, as well as the distance of several millimeters. chert are interesting because they suggest that at
least occasionally Hontoon Islanders had some contact
Bone artifacts range in weight from 0.20 g to 36.65 g, with outsiders. From the distribution of the stone in in length from 1.30 cm to 27.10 cm (many are broken), the deposit, it is apparent that this contact was in width from 0.49 cm to 6.50 cm, and in thickness or accelerated after historic contact (see Table I and diameter from 0.18 to 4.92 Cm. The primary species Discussion section). was deer including deer scapula which was used for
two of the pendants. Bird bone was used for a bead and Thirty-four of the 41 stone points were recovered from for one of the elaborately incised objects. The leg bone the historic levels of the site. Twelve of the 41 chert of a crane, Grus canadensis or, perhaps, whooping points and several stone tools belong to the crane, Grus americanus was used to manufacture the Preceramic Archaic period and thus they are out of artifact shown in Figure 5, and one of the pins was place chronologically. It is interesting that all of the made of alligator bone. Archaic Period spearheads and two of the tools were
recovered in the Historic Period levels. This situation
Fifty-nine bone artifacts are listed as decorative in suggests that (1) the Indians may have uncovered old Table 1. Thus, the decorative items, unexpectedly, chert points when they cleared land for the newly exceed slightly the utilitarian category making even introduced idea of cultivating crops, or that (2) the more convincing the suggestion that unmodified source of chert was cut off as a result of changing fractured bone scrap might have served as utilitarian relationships after historic contact and this forced the objects. The bone pins were probably used as hair and Indians to become relic hunters in order to increase the clothing ornaments. There is, of course, a value number of chert points needed for defense or for judgment involved in classifying objects as bone pins hunting (see Discussion section). One observation that since some of these might have served other functions verifies that these are really old points and not and might, in fact, have been manufactured for reproductions is the fact that none of the Pinellas utilitarian purposes. But I believe the same principle points, believed to be in place chronologically (Bullen can be applied to bone as has been mentioned 1975 ; Purdy 1981), are as weathered as the Archaic previously (Purdy 1981) for stone implements: if the Period points. finished product is the objective, e.g., a bone pin or a
stone spearhead, then greater care and precision will Twenty-seven of the remaining points are Pinellas 29
(some nicely made, some crude). A Duval point was than bashing, it is more accurate, aesthetically recovered in Level 10 of one 2-m section of the 1984 pleasing, and conserves the shell. In examining the trench and is in correct chronological position based on shell tools recovered at Hontoon Island, it appears that radiocarbon analysis and on ceramic styles. The other both bashing and cutting methods were used. I have point is an extremely small, unclassified, stemmed concluded, however, that little or no local production point. of shell implements occurred at Hontoon Island, at
least in the areas excavated, because large amounts of
There is quite a variety in the color and texture of the shell debitage should have been found if the first points recovered. This may suggest that the Indians technique were utilized and one would expect were obtaining the stone from several locations, quantities of flint debitage if the second technique Although this statement should be accepted with were utilized. caution because of the great diversity that may occur
even within a single quarry source, it is somewhat Prentice also conducted experiments on the same whelk substantiated by the fact that four artifacts of to determine the length of time and amount of flint silicified coral were recovered at the site, two from needed to remove the columnella and to cut small historic and two from prehistoric levels. Outcrops of sections in order to produce beads. "After 1391 minutes silicified coral occur in the Tampa Bay area and the (slightly more than 23 hours) of cutting time and the Suwannee River as well as a few additional locations, use of 191 flakes, the Busycon had been reduced to one Other kinds of chert also occur in these areas and in cup, three columnella pieces, and 17 bead blanks" numerous quarries in the central highlands. (Prentice 1983). He then proceeded to drill holes in the bead blanks with a flint microdrill and discovered that
The stone artifacts range in weight from 0.35 g to heat-altered blades produced from Florida chert were 830.00 g, in length from 1.20 cm to 15.02 cm, in width superior to unheated blades for drilling. After the from 1.00 cm to 12.00 cm, and in thickness or diameter holes are drilled, the blanks are strung on a line and run from 0.12 cm to 4.25 cm. through a grooved sandstone abrader in order to make
them round and uniform in size (Prentice 1983;
Shell personal communication 1986). For other useful and
comprehensive descriptions of the manufacture and use Tools and ornaments were produced from various of shell tools, see Holmes (1883), and Phillips and species of marine clams and conchs including Dosinia Brown (1978). discus, Mercenaria mercenaria, Dinocardium robustum,
Chione cancellata, Busycon sp. and Melongena sp. Nearly The functions of the shell tools have been broadly three times as many tools were made from conchs than assigned to adze, scraper, or gouge categories (Table were produced from marine clams. Two oyster shells I). One shell cup, made from Busycon sp., was (Oystrea sp.) and one Murex fulvescens were recovered recovered. No celts or hoes were found. Many of the also. These probably were fossils although the Murex implements placed in the unknown category were had a ragged hole in it that might have been manmade. fragments and, thus, their function could not be A shell hook (Figure 6) was made of Elliptio sp., the determined. It is very probable that some of the shell freshwater clam. It was the only shell artifact made tools were multipurpose. The major problem, of from a freshwater species recovered at Hontoon Island course, is to determine the actual use of an implement. during our excavations. It is one thing to note that a gouging, scraping, or
adzing motion was involved and quite another to
The technology involved in producing implements and specify what material was being worked or what kinds ornaments from marine shell species is not understood of objects were being produced. We are virtually thoroughly (Prentice, personal communication 1986). certain that wood was processed with shell tools and According to Prentice, in areas where large numbers of we have even been able to match up the marks on some specimens were available, the Indians probably found it of the wooden artifacts with the working end of the expedient to bash the conchs in order to fracture them. shell implements. Comparative studies and replicative If the columella was to be used as a tool or a gorget, experiments, similiar to those conducted by Prentice, the outer portion of the shell would be hit until the desperately need to be carried out in Florida to solve column could be removed easily. If the wall of the the problem of use wear on many classes of artifacts. conch was to be used, the desired sized piece of the
fractured shell would be chosen and then ground to a In contrast to the abundant decorative bone artifacts, working edge. only three ornaments were made of shell: a bead and a
Prentice conducted an experiment to determine if it pendant were manufactured from marine species and
Preniceconucte anexprimet t deermie i itthe shell hook from Elliptic sp. No shell gorgets were was possible to cut a knobbed whelk (Busycon carica found. [Gmelin]) with flint tools in order to manufacture a
shell cup. After 386 minutes (nearly 6-112 hours), Significantly larger percentages of shell tools were exhausting 60 flakes in the process, Prentice was able found in the squares excavated near the lagoon than in to detach a large triangular section from the body of more terrestrial units. This situation suggests that the shell (Prentice 1983). While this method is activities, like woodworking, occurred on the periphery considerably more time, energy, and flint consuming 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), Florida.
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 utilitarian.
S' x "/j.
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.
that broken or exhausted shell implements were tossed Historic Artifacts toward the lagoon rather than being discarded onto the
main habitation area (see Nodine 1986). The position of the European artifacts (Figure 8) in the
Shel atifctsrane i weghtfro .3 g o 37.0 gtoptre levels ot the deposit (Ta1le 1) ~uppOQ5 QUf inel lengath roma.0cmtge .m in w idth from 1 t 4.0g belief that the stratigraphy in the location of our to legt3o .85 mt 81 cm,an in thcnsso imfrec from 3 excavation units at Hontoon Island has not been 0.to 8.905 cm. nntikesorcrufrnefo disturbed by activities occurring in the twentieth
0.20 o 8.9 cm.century. The three historic period artifacts that were
Wood recovered from Level 4 were from the most lagoonward
- excavated squares (39 and 59) where greater organic
Along with the other botanical remains from Hontoon deposition was present. It is important to point out Island, the wooden artifacts, debitage, and ecofacts that all but two of the 37 historic period artifacts Supply information about environment and culture that were recovered from squares fringing the lagoon where is seldom preserved. Large carvings like the owl, found lessctrampin and ersooocredatrtbetswr in 1955 (Bullen 1955), and the pelican and otter, found dsaddo ot in 1977, were not recovered during our excavations Tecpe on(iue9,mjlc n lv a
since 1980 but the 48 wooden artifacts listed in Table I sherdspproabl thge las beadic and oe nariat
and houandsof iece ofdebtageproide lue toan estradiot hilt, support our conclusion based on other the wood species available and selected by the evidence that the historic artifacts were introduced aboriginal inhabitants. fairly early in the contact period. The other evidence
consists of limited radiocarbon and
Cypress and pine appear to be the predominant woods thermoluminescence analyses, stratigraphic position, utilized but many other species were used also and the fact that no European animal bones or plants suggesting that, in some cases or for some were recovered. Another bit of evidence that supports
woodworking projects, availability of a species might the conclusion that the European artifacts were have been just as important as wood properties. One introduced at Hontoon Island early in the contact period example pertains to the artifacts classified as posts. is Rouse's (1951:216) statement that "only in the earlier These were made of cypress, pine, elm, ash, and bay, periods ..dd the Spaniards have any real difficulty Long, straight pieces about 20 cm or more in with looting" of shipwrecks because during the later
circumference were selected and were sharpened on periods, "the Spaniards had only to send one or two one end by using a shell or stone adze. Fire may have soldiers from their outposts further north to recover been used to facilitate the shaping process. Only the loot. ." Yet, contrary to the conclusion reached by portions of posts driven into the ground were Rouse (1951), Goggin (1952) and others, that "the coming
recovered, because the tops had been burned off or of the Spaniards caused no appreciable changes in the eroded away. Indian cultures of the area" (Rouse 1951 :257), the
extreme changes that took place at Hontoon Island
Tool marks and manufacturing techniques can be suggest that the European artifacts represent more identified on many of the wooden artifacts. Future than the addition of a few exotic baubles to the experiments are expected to specify the types of assemblage. implements and processes utilized. Shark teeth marks,
for example, can be seen on the canoe paddle made of I am grateful to Jonathon Leader for his identification, red cedar (Figure 7), stone or shell adze marks can be descripton, and classification of the historic period seen also on this specimen as well as on many others, artifacts from Hontoon Island. Leader (1985,1986) and a shaft or pole of pine exhibits cross grain cutting. presents a comprehensive discussion of metal artifacts Most of the wooden artifacts have some charring thatfoninFrda indicates the use of fire either in the manufacturingfoninFrda process or through use. Charring is present on many of
the thousands of flat debitage pieces and may have Ceramics occurred from utilizing a fire hollowing technique in
the production of canoes. Differential burned areas on From the 1984 Trench (Squares 24-36, Table 2), a total portions of a single piece of wood contribute of 66,164 pottery sherds was recovered. They weigh
significantly to problems of preservation. 397,237 g, averaging 6 g per sherd. In 1980, 3153
sherds were recovered from the 3-meter test pit and
Determination of wood species and studies of wood 143 sherds were recovered from the 1982 column properties, technology, and artifacts are in progress at sample whose dimensions were 30 x 30 x 140 cm. The the University of Florida. The species listed in Table I sherds from the 2-meter test pit excavated in 1982 were identified by Lee Ann Newsom. were not counted and the sherds from the 1984 column
samples, trench extension, and the 1985 excavations
The wooden artifacts range in weight from 1.30 g to have not been counted yet. Greater than 99% of the 491.10 g, in length from 3.30 cm to 48.90 cm, in width pottery is St. Johns Plain and Check Stamped. An from .75 cm to 23.70 cm, and in thickness or analysis of the pottery from Hontoon Island will be the
circumference from .60 cm to 9.30 cm. subject of a master's thesis (McGee, in preparation).
Square 24 Square 25 Square 26 Square 29 Square 30 Square 31
Avg. Avg. Avg. Avg. Avg. Avg.
Level No. Wt. Wt. No. Wt. Wt. No. Wt. Wt. No. Wt. Wt. No. Wt. Wt. No. Wt. Wt.
(g) (g) (g) (g) (g) (9) (g) (9) (g) (g) (g) (9)
1 87 662 7.61 327 1497 4.58 573 1988 3.47 467 1241 2.66 1113 4038 3.63 1831 6976 3.81
2 118 1509 12.79 1066 4972 4.66 1287 3981 3.09 1207 5088 4.22 1541 7818 5.07 951 4777 5.02
3 93 1132 12.17 783 3192 4.08 554 3140 5.67 708 3977 5.62 596 3664 6.15 655 3126 4.77
4 72 986 13.69 531 4412 8.30 615 2987 4.86 358 2032 5.68 804 4252 5.29 887 4082 4.60
5 386 1767 4.58 439 2212 5.03 324 2009 6.20 445 2512 5.64 320 1601 5.00 709 4423 6.28
6 332 3997 12.03 635 3949 6.22 452 3241 7.17 278 1208 4.35 238 2026 8.51 321 1756 5.47
7 389 2910 7.48 659 4915 7.45 562 4559 8.11 241 1126 4.67 134 1441 10.75
8 430 3357 7.80 436 2168 4.97 483 2801 5.80 9 540 2529 4.68 530 2699 5.09 667 4023 6.03 10 611 3545 5.80 587 2870 4.89
HI 481 2265 4.71
12 171 370 2.16
13 162 1001 6.18
14 75 375 5.00
15 26 192 7.38
Total 2445 18849 7.70 6932 37764 5.45 5542 31639 5.70 3704 17184 4.64 4612 23399 5.07 5488 26581 4.84
Square 32 Square 33 Square 34 Square 35 Square 36
Avg. Avg. Avg. Avg. Avg
LeveI No. Wt. Wt. No. Wt. Wt. No. Wt. Wt. No. Wt. Wt. No. Wt. Wt.
(g) (g) (9) (g) (g) (g) (g) (g) (g) (g)
I 115 576 5.00 17 65 3.82 1026 5374 5.24 148 495 3.34 2 4 2.00
2 IOll 4452 4.40 485 4222 8.71 3501 16589 4.74 4974 18152 3.65 466 2549 5.47
3 1643 8759 5.33 778 6629 8.52 1364 5873 4.31 2240 10509 4.69 5010 20478 4.09 4 595 4040 6.79 923 4714 5.11 1014 7475 7.37 117 7968 7.13 1572 6572 4.18
5 473 4104 8.68 797 4678 5.87 721 5118 7.10 851 6191 7.27
6 215 2414 11.23 704 3760 5.34 417 3038 7.29 357 2424 6.79
7 766 4357 5.69 487 6330 13.00 350 5697 16.28
8 1333 12044 9.03 401 6838 17.05 697 14408 20.67
9 670 6127 9.14 661 11851 17.93
10 488 7320 15.00
II 488 7320 15.00
12 62 597 9.63
13 14 15
Total 4052 24345 6.01 7449 61833 8.30 9592 66486 6.93 10734 65844 6.13 5635 23688 4.20
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 once there, they were not stepped on and reduced to and supports our inference that through time the mound smaller sizes. Another explanation for the large grew toward the lagoon and actually increased the size increase in average weight per sherd with depth may be of the island. If one compares, for example, the number related to the thicker walls of the vessels in the of check stamped versus plain St. Johns sherds from earlier time period represented at the site (i.e., St. terrestrial square 26 of the 1984 trench (Williams Johns I). However, this does not seem to be borne out 1984) with those of the 1980 test pit excavated near by the average weight of sherds from the lower levels the lagoon (Newsom 1981), it is apparent that the of squares 25 and 26 which are correspondingly older cultural material from the bottom of the 1980 test pit than the lagoonward squares. Also, even though all of is comparable to only levels 3 to 4 of the more the ceramics were washed and dried before they were terrestrial portion of the 1984 trench (Table 3). It also counted and weighed, it is possible that the sherds demonstrates that the terrestrial or living portion of from the saturated levels of the trench remained the mound was eroded from trampling and other somewhat waterlogged and this may account partially activities and from the compaction that occurred when for the weight differential (Newsom, personal the botanical materials decomposed. communication, 1986).
One might conclude that the number of sherds versus In an effort to understand the meaning of 66,164 sherds sherd weight may give an indication of the amount of weighing 397,237 g I weighed and measured an Indian activity that has occurred on a site after a broken pot pottery bowl from the collections in the Department of has been discarded. All other factors being equal, for Anthropology, University of Florida. The bowl weighs example, if the average weight of sherds decreases or 1110 g, is 27 cm (approximately 11 inches) in diameter, increases, it may be correct to assume that there was 12 cm (approximately 5 inches) high, and has a wall more or less activity on the site following deposition thickness of .6 cm which falls within the thickness but prior to burial of a living floor (see Table 2, square range for the Hontoon Island sherds (McGee, n.d.). If 25, level 4 and square 26, levels 6 and 7). It is this were an average sized vessel being produced at interesting to note that in all of the lagoonward Hontoon Island, then sherds from 358 vessels were squares, beginning at approximately square 31, there is recovered from our excavations of the 1984 trench. A often an increase in number of sherds and a significant total of ninety-three, 25-cm levels in eleven 2 x 2-in increase in average weight of sherds with depth. This squares was excavated (93 in3) in the area of the 1984 situation probably exists for two reasons: broken pots trench where the sherds were recovered. This averages may have been thrown toward the lagoon (Nodine 1986) 8.5 levels per square and represents at least 1500
1980 lest Pit(Q sq. m) 1984, Sq. 26 (2 sq. m)
Level Plain % Check Stamped Plain3 Z Check Stamped Z
1 0 ster ile 1 0 0 414 73 157 27
2 0 sterile1 0 0 975 76 303 24
3 143 37 245 63 406 74 139 26
4 384 27 1013 72 586 95 31 6
5 577 52 531 48 417 98 10 2
6 202 89 2 23 10 452 100 0 0
7 0 sterile 2 0 0 562 100 0 0
8 0 sterile2 0 0 483 100 0 0
9 0 sterile2 0 0 667 100 0 0
10 0 sterile2 0 0 587 100 0 0
1 Accumulation of decomposed organic vegetation since site abandonment.
2Organically 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 River but is present within 25 miles of Hontoon Island. forty-two vessels were produced for each 175 years of The small size and evidence of use on most of the occupation. Admittedly, I have taken liberties with chipped stone implements suggest that flint may not these figures, but I believe this calculation suggests have been easily obtainable. The material of the that what originally seemed to be immense quantities coquina metate, the plummet (Figure 10) and celts were of ceramics, turns out to be a very low density for the not available locally but the distances they travelled to time and area involved. It is possible that broken pots get to the site varied considerably. The marine shell were used for something away from the site thus also had to be brought in from the coast although we reducing the amount of ceramic debitage. There are have not ruled out the possibility that the Indians were ethnographic descriptions of such a practice (Moseley, utilizing fossil shell beds that occur close by on the St. personal communication 1986). Johns River. The shark teeth may also fall into this
category. It is unlikely that any of the imported
To gain additional support for the accuracy of my material was essential for the maintenance of the calculations pertaining to the number of vessels group except, possibly, the marine shell which was represented by 66,164 sherds, Ann Cordell, manufactured into implements to work wood and other
Archaeologist at the Florida State Museum, weighed materials. and measured ten chalky paste vessels for me from the
museum collections that were recovered from burial We have positive evidence of local manufacture of sites. These vessels were reasonably complete except wooden objects because of the quantities of debitage. for the presence of kill holes in most of them. I Bone and pottery industries probably were present also figured the weight of the sherd missing from the kill at or near the site. hole of each pot, added it to the weight of each vessel,
and obtained an average of 837 g for each pot. If this Except for the marine shell drinking cup and some large were the average size for vessels produced at Hontoon pieces of wood, most of the artifacts are small. Many Island, then 475 vessels are represented. However, artifacts are broken. We found more complete from preliminary evaluations of rim sherds from specimens in the water-saturated squares. If people Hontoon Island, McGee (personal communication, 1986) were working along the edge of the water, they could tends to believe that vessels, larger than those from easily lose objects if they were dropped and settled the museum collections, are represented at the site. quickly into the organic deposit. There are many good Still, even if the 475 number is correct, only fifty-six opportunities at Hontoon Island to construct theories vessels were recovered for each 175 years of about site formation processes, and this situation occupation at Hontoon Island. applies to the artifacts as well.
In addition to St. Johns Plain and Check Stamped, a few With very few exceptions, such as the nicely made and Dunns Creek Red, Little Manatee Zoned Stamped, and decorated bone artifacts and a shell hook (Figure 6), fiber-tempered sherds were recovered at Hontoon there is nothing spectacular to hint at the artistic Island. There were also three sherds that had a design accomplishments we know existed prehistorically on the inside, a number of vessel feet and handles, and along the St. Johns River. Perhaps all of the elaborate the shapes of at least two sherds exhibit European carvings, other evidences of craftsmanship, and influence. expressions of the belief system were placed in graves
which we did not find at the site even though they were
Summary and Discussion mentioned by Wyman (1875:28).
All in all the artifacts are quite ordinary and not very Because we did not recover any European objects in the abundant when one considers the enormous quantities 1980 and 1982 squares or in the 1982 column sample, of flora and fauna present. The recovery of the wood is we attempted to explain in other ways the abrupt unique but most of it represents byproducts of changes that occurred after A.D. 1500. The most manufacture, perhaps for canoes or structures for the obvious change toward the top of the deposit is the top of the midden that have long since disappeared. disappearance of freshwater shell. About 10 cm of Certainly, we did not find carvings like the owl, pelican cultural deposit lies ,and otter (Purdy, Overview and Chronology, this issue)
that were discovered near Hontoon Island in the past. above the shell and contains abundant remains of There are a number of objects, such as a Clearfork pottery, bone, wood, and other items. The mystery is Gouge, Preceramic Archaic points, and several pieces of why the Indians abandoned a major food resource, fossilized bone (including Pleistocene horse and tapir freshwater snails and mussels in the early 1500s, but teeth) that lead us to conclude that these people were continued to live at the site. We have tentatively ruled relic collectors because these objects are out of place out an invasion by another Indian group as an chronologically. There are, however, other possible explanation because the pottery style is identical in explanations for their presence. the shell zone with that of the nonshell zone although
the shapes of some sherds in the nonshell zone suggest Some materials were brought into the site because an alien influence. There is a chance that a they do not occur locally. There is, for example, an catastrophe occurred that destroyed the shell beds for occasional piece of pottery that is not typical of the we verified in the 1984 trench and again in 1985 an area. Flint rock does not outcrop along the St. Johns observation made during the 1980 and 1982 excavations:
the uppermost portion of the shell deposit is tools, in the contact zone. Thirteen of the 17 shark
predominantly freshwater mussel (Elliptio sp.) that teeth and 34 of the 41 stone points, for example, came
seems to lie like a frosting over meters' thick from the upper horizon. These absolute numbers
predominantly freshwater snail (Viviparus sp.) We become more significant when one considers that the
have no explanation yet for the change from snail to volume of material removed from the prehistoric
mussel. It seems unlikely that either species was levels far exceeds that of the contact zone. The
affected by a climatic event because both can tolerate increase in artifacts at the site during a time when one
a wide range of temperatures (Thompson, personal would expect a decline in material culture resulting
communication 1984). The disappearance of the shell from population decimation, may indicate that Hontoon
may have occurred if the Indians simply did not desire Island was a refuge area for displaced people because
to eat it any more. This situation may have existed if of its favorable location along the St. Johns River. In
there had been a sudden emphasis on cultivated crops, contrast only about 20 of the 87 marine shell
such as squash and corn. Other changes that took place implements were found in the upper horizon suggesting,
at this time in the flora and fauna are discussed in as one possibility, that metal may have become
other papers in this issue and in Purdy and Newsom available.
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 References Cited
Spanish maravedi coin (Figure 9) minted at the Santo Domingo mint in the early to mid-1500s probably during Bullen, Ripley
1955 Carved Owl Totem, Deland, Florida. The Florida Anthropologist the reign of Carlos V and his mother Joanna (Deagan, 8:61-73.
personal communication 1985; Lyon, personal 1975 A Guide to the Identification of Florida Projectile Points.
communication 1985). The blue on white majolica ware Revised 'edition. Kendall Books, Gainesville.
also was manufactured during this period. The coin, Deagan, Kathleen
the majolica ware, the estradiot hilt, the absence of 1985 Personal communication. Dr. Deagan is curator in archaeology
any European plants or animals, plus supporting at the Florida State Museum. She is an historic archaeologist.
radiocarbon and thermoluminescence dates verify that Goggin, John M.
1952 Space and Time Perspective in Northern St. Johns Archeology, the Spanish material predates St. Augustine and thus Florida. Yale University Publications in Anthropology, No. 47.
may have been brought in as shipwreck material (Smith and Gottlob 1978:12) traded inland from coastal Indians. Holmes William H.
1883 Art in Shell of the Ancient Americans. Bureau of American
The changes we have noted, however, indicate that the Ethnologv, 2nd Annual Report, 1880-1881.
impact of the Europeans was greater than one would Leader, Jonathan M.
expect from the introduction of a few exotic items. 1985 Metal Artifacts from Fort Center: Aboriginal Metal Working in
the Southeastern U.S. M.A. thesis, Department of Anthropology,
University of Florida.
One example pertains to the bone artifacts recovered at 1986 Metalwork Among the Florida Indians in the Early Contact Period.
Hontoon Island. These were undoubtedly used for many Symposium on Spanish and Native American Encounters in Sixteenth
purposes including awls, hairpins, and weapons. In the Century Florida. Archaeological Institute of America and the Institute
prehistoric levels these are usually undecorated or of Early Contact Studies. Florida Slate Museum, Gainesville.
contain only a small amount of incising. We continued Lyon, Eugene
1985 Personal communication. Dr. Lyon is an historian whose specialty to find bone artifacts in the historic zone but a is the early contact period in Florida.
dramatic change occurs in decoration on some of the bone pins and on a piece of deer scapula (Figure 11). It McGee, Ray M.
1986 Personal communication. Mr. McGee is in the master's program, is interesting to note also that all three of the bone Department of Anthropology, Univers iof Florida. He is
pendants were recovered in the historic levels. It is studying the pottery from Hontoon Island.
possible that the new art style was introduced by Milanich, J.T. and C.H. Fairbanks
Indians from other areas as they were displaced by the 1980 Florida ArchaeologyI. Academic Press, New York.
Europeans, or it may represent a traditional design that Moseley, Michael
could not be executed on bone until metal tools were 1986 Personal communication. Dr. Moseley is Professor of
Anthropology, Department of Anthropology, University of
available. Perhaps, as an alternative, the new art style Florida.
was influenced by people from a different hemisphere Newsom, Lee Ann
and it may have been accompanied by changes in 1981 Preliminary Report on Ceramics from Hontoon Island. Manuscript
worldview and symbolism. The change in decoration on on file in Purdy's office, Department of Anthropology, University of
the bone pins is the topic of another paper (Purdy, in Florida.
preparation). 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, As mentioned above, we noted changes in the shape of particularly the botanical specimens.
some pottery sherds, suggesting European influences, Nodine, Bruce K.
and most of the unusual pottery was found in the 1986 Society and the Universal Problem of Refuse Disposal: An
Anthropological Analysis. M.A. thesis, Department of Anthropology,
shell-less zone. University of Florida.
Phillips, Philip and James A. Brown
Another interesting observation is that there was an 1978 Pre-Columbian Shell Enqravingqs from the Craig Mound at Spiro
increase in all artifact classes, except marine shell 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 Technoloqy. University Presses of Flor Gnesville.
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 Geographic Research 1(4):564-69.
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 Archeoloaical Evidence, 1500-1763, pp. 1-18. In: Tabc1. Edited by Jerald T.
Milanich and Samuel Proctor. -University Presses of Florida,
1984 Personal communication. Dr. Thompson is malacologist and curator in Natural Sciences, Florida State Museum, University of
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.
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
40 PRELIMINARY STUDY OF THE ANIMAL REMAINS EXCAVATED FROM THE
HONTOON ISLAND SITE
Elizabeth S. Wing and Laurie McKean
Introduction sediments and to stabilize the organic remains once they
were exposed to the air. A description of the excavation
Our knowledge of the past uses of plant and animal procedures can be found in the paper by Purdy and
resources is dependent upon the preservation of the Newsom (1985). The excavated bone deteriorates rapidly
remains of these organic materials and their careful once it is exposed to the air even when it is allowed to dry
excavation from archaeological sites. The Hontoon Island slowly. Therefore, all bone was soaked in a dilute solution site provides an extraordinarily enhanced view of a past of Elmers glue long enough for the glue to penetrate the economy because of the remarkably good preservation of bone. It was subsequently air dried, sorted to class, and
organic remains from this site. The conditions that are identified to the lowest taxon possible. responsible for this excellent preservation are permanent
submergence in water. A portion of the accumulated All identifications were made by comparison to series of
debris of the site is deposited in permanently saturated reference specimens. Standard methods of quantification peat and muck in the marsh and lagoon at the edge of the were used. These include counts of identified specimens, site. Anaerobic conditions retard the natural processes of estimates of minimum numbers of individuals, and weights deterioration which often eliminate or reduce the floral of identified remains. and faunal component of an archaeological deposit.
Identifications were made in the Zooarcheology
The Hontoon Island site is not only exceptional for the Department of the Florida State Museum in 1981 by fine preservation of organic remains but also valuable as a Laurie McKean and in 1986 by Tamara Markowitz and documentation of the changes in human uses of biological Elizabeth Wing. The sample identified by McKean is from resources. We have biological remains accumulated over the pit excavated in 1980, while sample numbers 135 and 1800 years of occupation at this site. The time span of this 145, both from the 1985 pit number 59, were identified by occupation extended from 0 A.D. to 1750 A.D. which Markowitz and Wing respectively. The two samples (135
witnessed the changes to aboriginal life in Florida wrought and 145) from the 1985 pit differ very little from one by the advent of Spanish exploration and settlement. another and are therefore combined for this analysis. A
total of 7753 identified fragments representing 38 species
The Hontoon site is located on the northern end of forms the basis of this analysis. This represents over 31
Hontoon Island which is midway along the St. Johns River. kilograms (glue impregnated dry weight) of vertebrate The site is situated in mixed hardwood forest and at the remains. Details of the faunal composition and edge of a freshwater marsh. The interior of the small, 1650 quantification of the finds are found in Tables 6 through 8. acre, island is predominantly pine flatwoods. These data appear in summary form in Tables 1 through 5.
Excavatations at Hontoon began in 1980 under the Five zones were defined from the first excavation at
direction of Dr. Barbara Purdy of the University of Florida. Hontoon Island: This paper discusses the vertebrate animal remains
recovered from thl first 3 m2 pit excavated at the site and Zone 1 (0-60 cm) is culturally sterile. a portion of a 2 m pit excavated in 1985. All of these
materials were recovered using a 0.6 cm gauge screen. Zone 2 (60-75 cm) consists of pottery of St. Johns 1I
Excavations in 1982, 1984, and 1985 included the removal period, adzed wood, and bone.
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 Zone 3 (75-85 cm) consists primarily of the freshwater results of the analyses of these excavation units are mussel (Elliptio), abundant bone, pottery, and
reported elsewhere (Newsom 1986; Wing in press) and are adzed wood.
in part still under investigation.
Zone 4 (90-120 cm) has a shell matrix consisting of diverse species but dominated by the aquatic snail
Materials and Methods iviparus georgianus. A few individuals of the
apple snail (Pomacea paludosa), a spiral snail
Investigations at Hontoon required special (Ganiobasisfloridensis), freshwater mussel
techniques to recover the material from submerged (Elliptio spp.), marine quahog (Mercenaria
Volume 40 Number THE FLORIDA ANTHROPOLOGIST March, 1987
FLORIDA ANTHROPOLOGICAL SOCIETY PUBLICATIONS NUMBER 13
campechiensis), marine clam (Chione cancellata), environments (Pennak 1978:723, 744-748). The Viviparus
marine cockle (Dinocardia robusta), and marine snail can be collected by hand in areas where colonies exist.
venerid clam (Macrocalista sp.) are also present. Colonies may be somewhat separated from one another
All the marine shells showed some sign of having and over exploitation might cause at least temporary
been altered. extirpation of a colony. Due to the prolific nature of the
snail, however, it is unlikely that populations would
Zone 5 (120 cm unknown depth) was not completed undergo any permanent decline as a result of human
because of limited time. At the completion of predation (Cumbaa 1976). The mussel was probably also
excavation a reduction in the amount of shell was an easy prey though they have to be dug up from the sand.
noted as well as the replacement of the fiberous
peat by organically stained sand. The vertebrates identified from the Hontoon samples
reflect a fauna similar to that which exists around the
Zones 2 and 3 became known as the Elliptio or historic island today, perhaps with the exception of the gopher
levels and Zones 4 and 5 are called the Viviparus or tortoise (Gopheruspolyphemus) and the wild turkey
prehistoric levels (Purdy and Newsom 1985). The 1980 pit (Meleagris gallopavo) which prefer generally drier terrain. includes both prehistoric and historic components whereas The turkey is known to roost on branches overhanging the the 1985 pit includes only historic remains. The historic water which affords the animal some protection against sample is slightly more than three times the size of the nocturnal assailants (Schorger 1966) but the tortoise prehistoric sample in this analysis. remains almost exclusively in the sandy areas in which it
makes its burrows (Carr 1952).
Studies of the biological remains from column samples
have shown how important the remains of small organisms, Evidences of Seasonality lost in the 0.6 cm screening procedure, are to a more
complete understanding of the exploited fauna. We realize Among the species represented by the bone material from a number of small species and smaller individuals of Hontoon, only one would be available during only one
species represented by larger specimens are absent from season. A fragment of the right tibiotarsus of a Canada the samples analyzed here. Despite this loss of information goose (Branta canadensis) was identified from zone 4 of from the samples, they can still provide insight into the the 1980 pit. This species presently occurs as a migrant or past uses of animals. winter resident in the middle St. Johns region.
Results and Discussion Fishes were probably caught all year around but are known
to be easier to catch in drier winters when the populations
Faunal Composition are more concentrated in the deeper pools and lakes.
Mullet become increasingly abundant in freshwater
Eleven species predominate in these samples. They are the drainages in late summer, fall, and early winter, particularly two freshwater molluscs, mussel (Elliptio buckleyi) and on sandy bottoms in well-rooted main channel vegetation snail (Viviparus georgianus), and five freshwater (McLane 1955). Both the striped mullet (Mugil cephalus)
vertebrates, gar (Lepisosteus sp.), catfish (Ictalurus sp.), and the white mullet (Mugil curema) were probably present bass (Micropterus salmoides), mullet (Mugil sp.), and slider in some numbers during all seasons. Most of the other turtle (Pseudemys spp.). Four primarily terrestrial fishes represented are year around residents of the St.
vertebrates are also promenent in the faunas and these Johns River.
include gopher tortoise (Gopherus polyphemus), ducks
(Anatidae), rabbit (Sylvilagus sp.), and deer (Odocoileus Reptiles are more easily available during some seasons. virginianus). Twenty-nine other vertebrate species were Gopher tortoises retire to deep burrow during the cold identified but are considered relatively less important winter weather. Alligators, likewise, are not frequently numerically, encountered during the winter.
Although the remains of the two dominant molluscs were Environment and Exploitation by Humans
not quantified in these samples, they were extremely
abundant. They, in fact, appeared to form the matrix of One of the key species represented in the historic horizon the cultural deposit and as mentioned before they are material is the gopher tortoise. Gopher tortoises are not
associated with the two cultural horizons at the site, the found on Hontoon Island today, nor is there evidence of mussel with the historic levels and the snail with the their presence on the Island in the past. The absence of
prehistoric levels. Little light is shed on the reasons for tortoises on Hontoon Island may be related to the Island's this dissimilarity of the two horizons based on the ecology soils which are unsuitable for the extensive burrows made of these two dominant species. Both I4viparus and Elliptio by the tortoises. About a mile away and to the west of the are found in sandy bottoms of shallow parts of rivers and Island are sand ridges which are a prime area for colonies lakes and are less frequently seen in the muddier of the tortoises. They can be collected when found outside
of their burrows. During extreme temperatures, either compared with the relative abundance of aquatic turtle,
extremely hot or cold, they retire to the more equitable primarily pond turtle, the two historic samples are temperatures of their burrows from which they are difficult consistent in having significantly more terrestrial turtle to extract (Taylor 1982). remains than is found in the prehistoric sample. On the
other hand when the relative abundance of all of the turtle
The pond turtles of the genus Pseudemys are found in is compared with all of the fish, the two samples from the
various aquatic habitats particularly lakes, sloughs, and 1980 pit are similar to each other and differ significantly rivers. They may congregate to sun themselves on logs from the 1985 sample which has much more fish than
sticking out of the water. When they are disturbed during turtle (Table 4 and 5). their sunning they will often slide into the water. People
who catch turtles take advantage of this behavior by 4) Catfish is the most abundant fish in all of the samples.
hanging traps in the water under a log used by the turtle They are followed in relative abundance by mullet and bass for sunning so that they slide off into the trap. In the or gar and bowfin. Many more remains of sunfishes are absence of archaeological finds of traps, we have no way of recovered from the finely screened column samples and it knowing whether such traps were used by the inhabitants may be assumed that they were of significance in these
of Hontoon. The evidence from the faunal assemblage samples as well. When the relative abundances of the
indicates that turtles were frequently caught. Such three large important fishes are compared in the three
consistent success in catching pond turtles suggests that an samples it is evident that the prehistoric 1980 sample and efficient technique, such as a trap, had been developed the 1985 sample are more similar in this respect than either
based on a knowledge of the behavior of the prey species. is to the historic 1980 sample (Table 5).
This same knowledge of animal behavior would of course 5) The data from the column sample taken in 1982 and to
have been applied to the consistent capture of any animal. be published elsewhere (Wing in press) can only be The most abundantly represented fishes are catfish, bass, compared with the information reported here with great and mullet. Today catfish and bass are most frequently caution because of the differences in the recovery
caught with hook and line. Catfishes are bottom feeders techniques used in collecting the samples. Each of the pit
and therefore are most successfully caught with a weighted samples analysed here is 40 times larger than the single hook near the bottom. Bass feed closer to the surface of column sample. It is unlikely that everything encountered the water and are most frequently caught in the upper in the 1982 square will be represented in the adjoining
layers of the water. Mullet eat plants and are usually column. We also know that most of the small remains,
caught in nets. Netting was found at the water logged those smaller than 0.6 cm were not recovered by the
deposits at Key Marco (Gilliland 1975) and a small piece of procedure used to excavate the pits. This gap in very similar two-ply cordage was found at Hontoon. A information provided by the two excavation procedures
shell fish hook has also been found at Hontoon. It is likely can only be filled by analysis of further samples. However, that both of these basic fishing techniques were used at in many respects the trends in change seen between Hontoon. prehistoric and historic samples are similar but differ in
Comparison of the Faunal Assemblage Represented in the
Three Samples 6) The overall breakage of bone is similar in the three
samples ranging from an average fragment size of 3 gra to
Based on these data some differences can be seen between 5.3 gm. However, differential breakage between key species
the predominant groups of vertebrates represented in the is evident. For example, the gopher tortoise fragments are
two horizons of the 1980 pit and between the two historic three times larger in the 1985 sample than in the 1980 horizon samples. The differences and similarities that are samples while catfish fragments are half as large in the evident are as follows: 1985 sample as the other. The average bone size in the
1982 column sample is of course much smaller (0.4 gm)
1) A comparison between the prehistoric and the historic because fine gauge screen was used to recover the horizons of the 1980 pit shows a substantially greater biological remains.
relative abundance of terrestrial vertebrates in the historic
horizon. Terrestrial turtles, primarily gopher tortoises, are Conclusions
responsible for this change (Tables 1, 2, and 3).
Despite the numbers of detailed studies of biological
2) A comparison between the two historic samples shows a remains from Hontoon our knowledge of the uses of plants difference in respect to the relative abundance of and animals by the site's inhabitants is still preliminary
terrestrial vertebrates in the samples and that the 1985 (Purdy and Newsom 1985; Newsom 1986; Wing in press). historic sample appears to be more similar to the Some characteristics of the faunal and floral assemblages
prehistoric sample than to the historic sample from 1980. seem to be consistent in the samples from different locations within the site studied thus far. They may well be
3) When this is analysed in more detail and the relative characteristic of the site as a whole but further research abundance of terrestrial turtle, primarily gopher tortoise, is must be conducted on biological remains to verify this. 42
Table 1. Fragment count of major vertebrate groups identified from the three samples. 243
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
[err. 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 83.7
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 73.9
Total 1200 1955 3083
Table 3. Weight (in gins) of identified fragments of the major vertebrate groups represented in the three samples.
Vertebrate 1980 1980 1985
Group Prehistoric Historic Historic
Weight Percent Weight Percent Weight Percent
Mammal 210.9 5.5 980.3 8.6 717.4 711
Bird 107.2 2.8 59.9 0.5 149 1.5
Terr. Turtle 1065.1 27.6 7896.6 69.5 4658.6 45.3
Total Terr. 1383.2 35.9 8936.8 78.7 5525 53.7
Aquatic Turtle 2054.4 53.2 2166.4 19.1 1095.5 10.7
Fish 420.8 10.9 253.7 2.2 3664.8 35.6
Total Aqtuatic 2475.2 64.2 2420.1 21.3 4760.3 46.3
Total Fauna 3858.4 11356.9 10285.3
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
Turtle 3119.5 88.1 10063 97.5 5754.1 61.1
Fish 420.8 11.9 235.7 2.3 3664.8 38.9
Total 3540.3 10316.7 9418.9
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
SPECIES NUMBER % MNI % WT
Sigmodon 2.0 0.1 2.0 0.8 0.3 0.0
Sylvilagus 43.0 1.1 5.0 2.0 133.7 1.1
Odocilagus 6.0 1.6 5.0 1.2 30.8 51 Table 7 Faunal list for 1980 Prehistoric horizon
Odocoileus 60.0 1.6 3.0 1.2 670.8 5.7
indet.mammal 14.0 0.4 0.0 0.0 22.6 0.2 SPECIES NUMBER MNI
indet. bird 180.0 4.7 16.0 6.5 153.2 1.3 S MNI
Alligator 2.0 0.1 1.0 0.4 4.5 0.0
indet. snake 1.0 0.0 1.0 0.4 1.3 0.0 Didelphis 1.0 0.1 1.0 0.9 1.7 0.0
Terrapene 19.0 0.5 4.0 1.6 122.3 1. 0 Sigmodon 1.0 0.1 1.0 0.9 0.4 0.0
Gopherus 646.0 16.8 19.0 7.7 4536.3 38.5 Sylvilagus 10.0 0.6 3.0 2.7 8.1 0.1
Chelydra 1.0 0.0 1.0 0.4 0.7 0.0 Procyon 3.0 0.2 1.0 0.9 17.3 0.2
Kinosternon 11.0 0.3 2.0 0.8 8.2 0.1 Odocoileus 5.0 0.3 1.0 0.9 123.0 1.6
Pseudemys 113.0 2.9 5.0 2.0 1037.2 8.8 indet. mammal 23.0 1.4 0.0 0.0 60.4 0.8
Trionyx 14.0 0.4 2.0 0.8 49.4 0.4 Phalacrocorax 1.0 0.1 1.0 0.9 0.6 0.0
indet. turtle 560.0 14.6 0.0 0.0 1372.2 11.7 Branta 1.0 0.1 1.0 0.9 3.2 0.0
Anura 1.0 0.0 1.0 0.4 0.2 0.0 Podilymbus 5.0 0.3 2.0 1.8 3.6 0.0
Ardeidae 1.0 0.1 1.0 0.9 0.1 0.0
Lepisosteus 92.0 2.4 4.0 1.6 71.8 0.6 Ardea 3.0 0.2 2.0 1.8 15.5 0.2
Amia 59.0 1.5 6.0 2.4 41.0 0.3
Ictalurus 1390.0 36.1 100.0 40.7 741.3 6.3 Casmerodius 1.0 0.1 1.0 0.9 1.3 0.0
Lepomus 35.0 0.9 13.0 5.3 18.2 0.2 A ramus 4.0 0.2 2.0 1.8 94.4 0.1
Micropterus 255.0 6.6 17.0 6.9 188.3 1.6 Cathartes 6.0 0.4 2.0 1.8 249.5 0.3
Mugil 348.0 9.0 44.0 17.9 118.9 1.0 Buteo 1.0 0.1 1.0 0.9 1.3 0.0
indet. fish 0.0 0.0 0.0 2485.3 21.1 Buteo 1.0 0.1 1.0 0.9 1.3 0.0
Meleagris 5.0 0.3 2.0 1.8 16.3 0.2
indet. bird 23.0 1.4 0.0 27.4 0.4
Total 3846.0 246.0 11777.7 Alligator 95.0 5.6 7.0 6.2 3718.4 47.6
Crotalidae 3.0 0.2 2.0 1.8 0.5 0.0
indet. snake 1.0 0.1 1.0 0.9 0.3 0.0
Table 8 Faunal list for 1980 Historic Horizon Terrapene 23.0 1.4 4.0 3.5 59.4 0.8
Gopherus 198.0 11.8 7.0 6.2 1005.7 12.9
SPECIES NUMBER MNI WT % Chelydra 54.0 3.2 3.0 2.7 379.6 4.9
Kinosternidae 25.0 1.5 2.0 1.8 20.4 0.3
Sylvilagus 11.0 0.5 2.0 1.3 9.6 0.1 Kinosternon 1.0 0.1 1.0 0.9 2.2 0.0
Lynx 1.0 0.0 1.0 0.6 4.2 0.0 Sternotherus 7.0 0.4 2.0 1.8 6.3 0.1
Procyon 1.0 0.0 1.0 0.6 4.2 0.0 Pseudemys 223.0 13.3 12.0 10.6 1383.6 17.7
Odocoileus 94.0 4.2 6.0 3.9 962.3 8.2 Cheloniidae 10.0 0.6 2.0 1.8 86.4 1.1
Phalacrocorax 1.0 0.0 1.0 0.6 2.6 0.0 Trionyx 36.0 2.1 5.0 4.4 175.9 2.3
Podilymbus 1.0 0.0 1.0 0.6 1.3 0.0 indet. turtle 283.0 16.8 0.0 0.0 232.1 3.0
Anatidae 8.0 0.4 5.0 3.2 6.7 0.1 Rana 2.0 0.1 2.0 1.8 0.6 0.0
Ardea 1.0 0.0 1.0 0.6 5.3 0.0 Siren 2.0 0.1 2.0 1.8 0.6 0.0
Butorides 1.0 0.0 1.0 0.6 0.2 0.0 Lepisosteus 52.0 3.1 5.0 4.4 78.9 1.0
Casmerodius 4.0 0.2 2.0 1.3 7.8 0.1 Amia 43.0 2.6 5.0 4.4 31.6 0.4
Aramus 1.0 0.0 1.0 0.6 1.1 0.0 Ictalurus 141.0 8.4 11.0 9.7 106.4 1.4
Meleagris 3.0 0.1 2.0 1.3 30.3 0.3 Centrarchidae 12.0 0.7 5.0 4.4 5.0 0.1
indet. bird 3.0 0.1 0.0 0.0 4.6 0.0 Pomoxis 3.0 0.2 1.0 0.9 1.7 0.0
Alligator 3.0 0.1 2.0 1.3 139.1 1.2 Lepomis 2.0 0.1 2.0 1.8 3.3 0.0
Terrapene 30.0 1.3 6.0 3.9 241.7 2.0 Micropterus 30.0 1.8 5.0 4.4 39.9 0.5
Gopherus 1343.0 60.4 40.0 25.8 7654.9 64.9 Mugil 34.0 2.0 4.0 3.5 15.8 0.2
Chelydra 2.0 0.1 1.0 0.6 23.9 0.2 indet. fish 306.0 18.2 0.0 0.0 138.2 1.8
Pseudemys 238.0 10.7 12.0 7.7 1925.4 16.3
Trionyx 34.0 1.5 3.0 1.9 217.1 1.8 Total 1683.0 113.0 7810.9
indet. turtle 136.0 6.1 0.0 0.0 301.3 2.6
Lepisosteus 22.0 1.0 5.0 3.2 63.8 0.5
Amia 10.0 0.4 3.0 1.9 11.2 0.1
Ictalurus 97.0 4.4 35.0 22.6 92.7 0.8
Centrarchidae 3.0 0.1 1.0 0.6 1.1 0.0
Lepomis 1.0 0.0 1.0 0.6 0.8 0.0
Micropterus 9.0 0.4 2.0 1.3 10.0 0.1
Mugil 63.0 2.8 20.0 12.9 48.6 0.4
indet. fish 103.0 4.6 0.0 0.0 25.5 0.2
Total 2224.0 155.0 11797.3
Further research is required to understand fully the nature remains from the unit and compare this with the faunal of the changes in the uses of plants and animals by the remains recovered by identical screen size from the
Indian inhabitants of the site brought about by the advent column. If the faunal assemblages are similar then one of the Spanish entrance upon the Florida scene. More could be secure in the belief that the column represents
samples must also be studied to document variability the fauna of the larger unit. Furthermore, the faunal
throughout the site and to gain an understanding of the component that fell through the larger gauge screen and
causes of this variability, could be identified from the column represents the fauna
lost in the larger excavation unit. By this device, the gap Certain characteristics of the biological remains seem to be that exists between the data from the excavation unit and well established by the ethnobiological studies thus far. the column sample could be bridged. These data may then The predominant species in the faunal assemblages from permit more accurate interpretation of past economies and
the different parts of the site remain the same. The change site activity areas. from a predominance of snails in the prehistoric horizon to mussels in the historic horizon appears to be constant Acknowledgments
throughout the site. Three groups of vertebrates
represented in great abundance in all samples (consituting We are grateful to Dr. Barbara Purdy for the
32, 54, and 82 percent of these samples) are the terrestrial opportunity to study these most interesting samples of turtles primarily the gopher tortoise, the aquatic turtles animal remains from the Hontoon site. Thanks are also primarily the pond turtles, and fishes which includes due to Tamara Markowitz who spent the major part of the
primarily catfishes. If comparison between the relative summer of 1986 identifying the remains from subsample
abundances (based on weight of identified remains) of the number 135 of the 1985 sample. Lee Newsom who has three primary species is made a trend towards increased worked intensively on the botanical remains from Hontoon
abundance of the gopher tortoise in the historic horizon is has helped us in many ways throughout our research on clear. the animal remains for which we are profoundly grateful.
A more general trend towards an increase in terrestrial Ctr rheF
vertebrates in the historic horizon is less obvious because C 19r 0 Arh A F. tiuint il eptlg f lrd.Uie
of differences in the faunal assemblages of the different of Florida puitCion. Biological Scientce Series No. 1 1 1
samples. For example, the weight of alligator remains
idenifid i th prhisori 198 saple costiute 47Cumria, Stephen Lidetifed n te pehitorc 980samlescontittes471976 A Reconsideration of Freshwater Shellfish Exploitation ill
percent of the assemblage and the weight of unidentified tile Florida Archaic. Thle Florida Anithrop~ologyist 29(2):49fish remains is 21 percent in the 1985 sample whereas these59 are negligible in the other samples. G illitand, Marian S.
1975 The Material Cultuire of Key Marco Florida. University
Presses of Florida. Gainesville 266 tsp.
Slender evidence points to year around occupation of the M~nWlimM
sit. Gphe totoies ndallgatrs re oreeasly1955 The Fishes of tile St. Johns River System. Ph.D.
caught during the warmer months. The Canada goose and Dissertation University of Florida.
some ducks are available during their winter migrations to Newsont, Lee A.
1986 Plants, Human Subsistence, and Environment: A Case
the south. Evidence from the ethnobotanical. research Srudy from Ilontoon island (8-VG-202), Florida. Masters
augments these data. Thesis University of Florida 152 pp.
Peitnak, Robert W.
Thefaual vidnceindcats hatthrughut he istry1978 Freshwater tnvertebrates Of the United States (second
of occupation at the site the inhabitants engaged in a eiin ie-nesineNwYr
variety of subsistence activities. These included initially Purdy, Barbara A. and Lee A. Newsom
gaterig sail an laer iggng usslscathin pod1985 Significance of Archaeological Wet Sites: A Florida
gathrin snilsandlatr dggin musel, ctchng ondExampile. National Geographic Research ](4):564-569. turtles possibly with traps or by hand, gathering gopher Schorger, A. W.
tortoises, and fishing with both nets and hook and line. 1966 Thte Wild Turkey its History and Domestication.
These activities were of course integrated with plant University of Oklahoma Press, Norman OK 625 pp.
gathering and cultivation (Newsom 1986). Taylor. Robert W. Jr.
1982 Human Predation on the Gopher Tortoise (Gopherus
polypltenrus) in North-Central Florida. Bulletin Of tile
More comparable samples are needed in order to Florida State Museum 28(1-4):79-102.
understand the differential distribution of species and Wing, Elizabetht S.
diffrenes i th frgmen sie thougoutis press Integration of Floral and Faunal Data from Hoittoon dfeecsithfrgetszthogotthe site. island. Florida. Arcliaeozoolonia.
Column samples should be increased and examined in such a way so that one can be confident that they reflect the Elizabeth S. WingLareMen
faunl asembageof he ajacnt xcavtio uni. Oe Forid Stte useu an Zlrierment faunal ~ ~ ~ ~ ~ ~ ~ ~ nvest asebaeooheajcnfxavto nt n Florida SttNuem ar ooloh Darntenvrst
wytdotiwolbetidniyadaayethe animal Gainesville, FL 32611 Raleight, NC 27695
ANALYSIS OF BOTANICAL REMAINS FROM HONTOON 47
ISLAND (8V0202), FLORDA: 1980-1985 EXCAVATIONS
Lee A. Newsom
This paper summarizes the analysis of plant remains from a submerged shell midden at Hontoon Island (8Vo2O2), Florida.
The deposit has yielded an exceptional array of well preserved 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-gatheringfishing 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 saturated deposit that the
located in western Volusia County, majority of the excavation units
Florida. It is situated on the were placed. The archaeobotanical
St. Johns River, nearly midway component of this wet deposit is
along the river's 315 mile (504 extensive and diverse. The
km) course. Several archaeo- species identified represent a
logical deposits are on the lush river basin and hammock
island, including two very large environment characterized by great
shell middens, two middens of much species richness, and upon this
smaller extent, and at least one base the island's inhabitants made
shell field or scatter. The mid- their living. Through ceramic
den on the northern end of the chronology and absolute dating,
island has been the focus of Uni- the deposit has been demonstrated
versity of Florida excavations for to range from the early St. Johns
five years (1980-1985) under the I Period, or earlier (Purdy, this
direction of Dr. Barbara A. Purdy. volume; Milanich and Fairbanks
The northeastern and eastern edges 1980), through the Spanish Contact
of this midden slope off into a and Mission periods of the sixlarge, low energy lagoon and teenth and seventeenth centuries
continue for some distance under (Ehrmann 1940; Deagan 1978)-the surface of the water, there- roughly A.D. 0 through A.D. 1750.
fore affording unusually good
conditions for organic preser- Two distinct cultural expressions
vation. It is in this water- have been identified in the work
Volume 40 Number 1 THE FLORIDA ANTHROPOLOGIST March, 1987
FLORIDA ANTHROPOLOGICAL SOCIETY PUBLICATIONS NUMBER 13
at Hontoon Island. The earliest headwaters in southwest Brevard
is associated with an aboriginal County, and then north to approxioccupation of long duration, mately Lake Harney; the Middle
having existed approximately one Basin, beginning with Lake Harney
thousand years. The evidence and going north to the vicinity of
suggests an extensive habitation present-day Palatka; and finally,
area occupied by a people who had the Lower Basin, from Palatka to
established a complex economy the mouth of the river near Jackbased upon resources procured from sonville (Figure 1). Hontoon
the river basin and flanking Island is in the center of the
uplands. In addition, their diet Middle Basin. The distance from
may have been supplemented by Hontoon Island to the coast is
gardening or the encouragement of approximately 40 km.
at least two cultivated plants.
Early in the historic period The evidence suggests that the
change occurred, and there is Hontoon Islanders conducted most
evidence of more extensive use of of their plant collecting/harvestcultivated plants along with ing activities within close proxicertain other adjustments to the mity to home base. The greatest
subsistence base. The observed intensity of collections appears
changes in the archaeological to have occurred within 5-10 km of
record at Hontoon Island may the site, with the majority of the
reflect voluntary and involun- plants identified being representary disruptions in the aboriginal tative of the river basin and
culture arising with the initial flanking uplands. Coastal
presence of Europeans in the resources were also exploited,
region. Further stimulus for though perhaps on an occasional
change may have come about dir- basis, being represented in much
ectly through influences exerted lower frequencies than remains
by the Spanish mission system in from nearby habitats.
Within the intensive use area,
THE BIOCLIMATIC SETTING being that within 10 km of the
site, are five principal ecoloHontoon Island is located in an gical communities: freshwater
area of environmental diversity, marsh, hardwood river swamp, hardVegetational complexes in this wood hammock, pine flatwoods, and
region are governed less by soil sand pine scrub (Laessle 1942;
fertility than by texture and Puckett 1982; U.S.D.A. Soil Conmoisture holding capacity or servation Service 1985). On the
permeability (Snedaker 1971:25). whole, these reflect a gradient
The climate is relatively mild due from wetter conditions and lower
to close proximity to the Atlantic elevations along the river's edge,
Ocean. to progressively better drained
soils and higher elevations on the
Along the St. Johns River's north- nearby sand ridges. The diversity
erly course the river channel, its in topography and soil/edaphic
floodplain, and adjacent topogra- conditions is reflected in the
phy, undergo several changes in associated plant assemblages, and
character. Because of this, the the species of plants identified
river has been described as having in the archaeological record
three distinct segments: the Upper closely mirror the present day
Basin, beginning with the river's vegetation.
- 4 6
Matanzas Inlet Pa Iatka 5 7
7 Daytona Beach
HONTOON New Smyrna 8 10
I SLAND Beach
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 Figure 1. Northeast-central Florida including the St. Johns River Basin. Segments.
METHODS The column samples were excavated
in 10 cm segments. Each such
The recent excavations at Hontoon segment was bagged in its entirety
Island include a 3-in square, exca- and sent directly to the laboravated in December 1980; a 2-rn tory for processing. The indivisquare plus column sample, exca- dual segments then underwent a
vated in February 1982; a 26x2-m series of fine screening and chemtrench with column samples, exca- ical flotation (nested screen
vated during an archaeological sizes are: 4mm, 2mm, 1mm, and 125
field school in the spring of )-km; see Newsom 1986 for a complete
1984; and finally, a pair of 2-in description).
squares with two column samples,Afe coptinfthiiia
in May 1985.Afecopeinoth iiia
flotation and separation process,
the plant remains were further
Thisrepot fcuse oninfoma-sorted and identified with the aid
tion obtained from two of the o iscigadtasitd
column samples: one from the pro- light microscopes. All quantififile of the 1982 square, and the cations were made using volume as
other from the profile of 1985 a unit of relative mass since
squae 5. Th corse-crenedmeasurements based on weight were
sqarea 59. m the coaresreeneda deemed unreliable due to the
mtial froms thel rer oxcava waterlogged condition of the
tiomar uni mets wl reev onplmn material. The faunal material
summaryoratmnt aso ah souplmnt from the heavy fractions of the
tomle inorain from thre sclumn flotations underwent further prosmssi nce thed large scree5] n cessing at the Zooarchaeology
size5 use (.27] cm tr[0.5e] and Laboratory of the Florida State
great a bias against smaller sized carredm out unde the dirvecto ofs
plant remains. Both of these Darie Eliabth S.de Wing Flrcia o
column samples were excavated from Stat Muszaem WigFlrd
fully water-saturated deposits andSteMuum
preservation can be considered Seed identifications were made
roughly equal. The two columns with the aid of comparative colwere excavated and processed in lections belonging to the Florida
identical manner, although it must State Museum Herbarium and to the
be noted that they are not equalauhranwihteseo
in sze.The198 colmn as alfseveral reference works (e.g.,
as wide as the first column (the Martin and Barkley 1961; Hitchcock
1982 column measured 30x30 cm, 1971; Landers and Johnson 1976).
while the 1985 column was 30x15 Wood and charcoal identifications
cm) (Figure 2). The principal were made using anatomical keys
investigator believed a smaller (Record and Hess 1943-1947; Record
size was necessary to facilitate and Hess 1943; Panshin and de
and shorten the time spent in Zeeuw 1980) and by reference to
processing the samples. One comparative specimens in the Herfurther difference is that the barium's wood collection, along
1985 column was not excavated as with samples collected at Hontoon
deeply as the 1982 column due to Island, Florida. In the interest
problems with water control. These of brevity, an analysis of the
differences are considered in the wood and charcoal samples will be
discussion on the plant remains, reserved for a separate report.
Certain features of the shell mid- each (excluding sterile overden must be made clear before con- burden) (i.e., approximately 45-50
tinuing on to the description and cm thick; the underlying snail
analysis of the plant remains, shell midden was not encountered
Earlier it was mentioned that two before the water level precluded
distinct components/deposits were further excavation). Thus, in the
uncovered at the site. This is 1982 column both deposits are
evidenced most clearly by changes represented, while in the 1985
in shellfish composition within column only the historic period
the midden which has resulted in midden was encountered. Also, in
the identification of two separate association with the shell midden
"macro-strata" or perhaps, more proper, there is a discontinuous
accurately, two separate middens. shell-free midden overlying the
The first is a very extensive mussel shell levels. Since, other
midden containing evidence of a than the absence or near-absence
wholly aboriginal material cul- of shell, it is identical in plant
ture. Freshwater snail (Viviparus and animal remains to the mussel
georgianus) forms the primary shell midden, for these purposes
building material of this deposit, it is considered one and the same.
which dates from ca. A.D. 0 (the
early St. Johns I period) through No attempt will be made here to
approximately A.D. 1500. This rank taxa according to their rela"snail shell" midden was tive importance. This is being
encountered in all excavations at delayed until more data are availthe site except the 1985 units. able, even though the present samThe second deposit is much smaller ple is thought to be representain extent and is situated strati- tive.
graphically superior to the
original midden. It is distinc- PLANT REMAINS
tive in being composed primarily
of freshwater mussel shell A total of 3754 seeds and plant
(Elliptio buckleyi). These parts were identified from the
superficial or "mussel shell" 1982 column sample (not including
levels yielded a number of the fragment counts of nut remains
European artifacts, and radio- and gourd/squash rind). In addicarbon dates obtained on these tion 1224 wood fragments were
layers range from ca. A.D. 1540 to examined and identified. The
the 1750s (Purdy, this volume), identity of 95 seeds and fragments
remained uncertain. From the 1985
The differences between the two column, 2880 seeds were identified
deposits or middens are being (again, exclusive of nut remains,
emphasized as necessary to an cucurbit rind, and also corn cob
understanding of the following fragments), along with 320 wood
analysis since marked changes in samples. Unidentified seeds and
the botanical and faunal compo- fragments from the 1985 column
nents occurred with the change in totaled 232.
shell matrix noted above. The
mussel shell layer comprised the Tables 1 and 2 show the division
uppermost segments (Levels 5-8, of the plant remains from the two
ca. 40 cm thick) of the 1982 column samples into gross catecolumn sample and excavation unit. gories. "Plant residue" repreIt likewise occurred in the 1985 sents unidentifiable, largely
units and adjacent column samples, fragmentary plant remains (e.g.,
where it comprised the whole of 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
'1UT 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
ZEA MAYS (COB)
PLANT RESIDUE 525 1295 1070 2550 1170 905 920 1510 350 451
'Volumes are in cubic centimeters.
**"Oter" 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.
*""Other" includes pine cones and scales, cane stem, and fungi.
The combined identifications from latter observations were partithe two column samples and the cularly important to the first
larger excavation units yielded a European explorers since it often
total of 97 vascular and meant the difference between life
nonvascular plants, representative and death as supplies from their
of 82 different species of seeds homelands were not reliable or
and miscellaneous plant parts were not forthcoming at all. Reitz
(e.g., cane stem and pine cone and Scarry (1985) and others (cf.
scales), and 30 different woods. Sturtevant 1962:68) have found
Table 3 lists these taxa. The that sixteenth century Spanish
identifications from the snail colonists came to rely almost
shell and mussel shell deposits exclusively on New World plants
closely paralleled each other and animals primarily for these
where native (wild) plants are reasons. Thomas Hariot, an
concerned, that is, in terms of Englishman and "planter," came to
what species were actually repre- "Virginia" (now coastal North
sented. Differences were observed Carolina) around 1585 (Hariot
though, in their frequency and 1951). As a farmer, he paid pardistribution through time. Marked ticular attention to the soils,
differences between the two depo- climate, what was being sown and
sits were found in the presence gathered by the Indians, their
and relative abundance of culti- processing methods, and more.
vated plants. Corn was found only Because many of the same plants
in the historic period mussel and associated human traditions
shell levels and, in addition, are common to the Southeastern
noteworthy changes in the pumpkin/ Coastal Plain and range into censquash remains were discovered. tral Florida, his work is most
enlightening. Table 4 lists the
Wild Plant Resources plants mentioned in a few selected
sixteenth century descriptions of
At least 90 wild or indigenous Florida and the southeastern
plants have been identified; among United States.
these are bracket fungi, aquatic
plants, grasses, sedges, herbs, Laudonniere's (1975) and Ribaut's
vines, shrubs, and trees. With (1964) accounts have more direct
the aid of ethnohistoric and bearing on this study since most
ethnobotanical literature we can of the Indians they had contact
begin to identify which plants with were members of the Timucuan
might have been economically tribe--the parent group with which
valuable as food, oils, and the inhabitants of Hontoon Island
condiments to the human inhabi- are believed to have been affilitants of Hontoon Island (medicinal ated (Newsom 1986:12-22). A numand other uses, such as dyes, will ber of the plants mentioned by the
not be considered here). explorers and listed in Table 4
have been found in the Hontoon
The early accounts of exploration Island samples, including: hicand attempts at colonization in kory, bay, oak, grape, persimmon,
the New World are replete with elderberry, sorrel, and more.
references to plants, both in an Table 5 lists the plants identieffort to describe natural vegeta- fied from the archaeological
tion and to relate how and which deposits at Hontoon Island which
among the native plants were being might have been utilized for their
used by Native Americans. These 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 "ornamental gourd"
C. pepo, type two squash/pumpkin
C. pepo, type three squash/pumpkin
C. pepo, (?) type four squash/pumpkin
Lagenaria siceraria L. bottle gourd
Zea mays L. corn
? Pyrus malus L. domestic apple
WILD PLANT RESOURCES:
Amaranthus sp. amaranth; pigweed
Amaranthaceae amaranth family
Ambrosia sp. ragweed
Ampelopsis arborea (L.) Koehne peppervine
Asteraceae [inflorescence] sunflower family
Berchemia scandens (Hill) K. Koch rattan vine
Brasenia schreberi Gmel. water shield
Capsella bursa-pastoris (L.) Medic. shepherd's purse
? Carex sp. sedge
Carva aquatic (Michx. f.) Nutt. water hickory
C. glabra (Mill.) Sweet pignut hickory
Celtis laevigata Willd. [leaves and seed] sugarberry
Cephalanthus occidentalis L. buttonbush
Chenopodiaceae goosefoot family
Cornus florida L. flowering dogwood
C. foemina Mill. swamp dogwood
Cyperaceae, type one sedge, unidentified
Cyperaceae, type two: tublar shape sedge, unidentified
Cyperus sp. nutsedge
Dioda virginiana L. buttonweed
Diospyros virginiana L. persimmon
Fomes sclerodermeus (Lev.) Cooke [thallus] bracket fungus
Gaylussacia sp. huckleberry
Hypericaceae [inflorescence] St. John's-wort
Ilex opaca Ait. American holly
? Ipomoea sp. morning-glory
Lyonia sp. staggerbush/
L. ferruginea (Walt.) Nutt. rusty Lyonia
Magnolia grandiflora L. southern magnolia
M. virginiana L. sweet bay
Myrica cerifera L. wax myrtle
Naias guadelupensis (Spreng.) Magnus pondweed; naiad
N. marina L. spiny naiad
Nuphar lutea Sibth. & Small spatterdock
Table 3 continued.
Nymphaea sp., prob. N. odorata Ait. waterlily
Nyssa sylvatica Marsh. black gum
Passiflora incarnata L. may pop
Persea sp. bay
P. ? palustris (Raf.) Sarg. swamp bay
Physalis sp. groundcherry
Phytolacca americana L. pokeberry
Pinus clausa (Chapm.) Vasey [cones] sand pine
P. elliottii Engelm. [cones] slash pine
P. palustris Mill. [cones] longleaf pine
Polygonum sp. smartweed/knotweed
P. densiflorum Meisn. smartweed/knotweed
P. hydropiperoides Michx. waterpepper
P. punctatum Ell. dotted smartweed
Polyporus hydnoides Swartz ex Fries [thallus] bracket fungus
P_. sulfureous (Bull.) Fries [thallus] sulphur shelf
Portulacca oleracea 1. common purslane
Prunus caroliniana Ait. Carolina laurel
Quercus lrata Walt. overcup oak
Q. michauxii Nutt. swamp chestnut oak
. nigra L./Q. laurifolia Michx. water oak or
. virginiana Mill. live oak
Rhynchospora sp. sedge
Rubus sp., type one blackberry
Rubus sp., type two blackberry
Rumex sp. dock
Sabal palmetto (Walt.) Lodd. ex Schultes cabbage palm
Sambucus canadensis L. elderberry
Scirpus sp. bulrush
S. validus Vahl soft-stem bulrush
Serenoa repens (Bartr.) Small saw palmetto
Setaria sp. bristlegrass
Sida sp. [seeds and fruit] broomweed;
Smilax sp. [leaves] greenbrier/catbrier
? Sparganium sp. bur-reed
Tillandsia usneoides (L.) L. [stems] Spanish moss
Ulmus americana L. [buds] American elm
Vaccinium sp., prob. V. corymbosum L. highbush blueberry
Vitis sp. (? cultivar; Watlington 1985) grape
V. aestivalis Michx. summer grape
V. cordifolia Lam. frost grape
V. munsoniana Simpson muscadine grape
unidentified charred rhizomes ? waterlily
unidentified thorns, prob. Rubus spp. blackberry
Table 3 continued.
Part II: Woods
Arundinaria gigantea (Walt.) Chapm. switch cane
Sabal palmetto (Walt.) Lodd. ex Schultes cabbage palm
Juniperus sp. red cedar
Pinus sp., section Diploxylon hard pines
Taxodium sp. cypress
Acer sp., prob. A. saccaharinum L. silver maple
Carva sp. hickory
C. aquatica (Michx. f.) Nutt. water hickory
Cephalanthus occidentalis L. buttonbush
Diospyros virginiana L. persimmon
Fraxinus sp. ash
F. caroliniana Mill. water ash/pop ash
Gordonia lasianthus (L.) Ellis loblolly bay
Liquidambar styraciflua L. sweetgum
Lyonia sp. staggerbush/
Magnoliaceae magnolia family
Nyssa sylvatica Marsh. black gum
Persea sp. bay
Prunus serotina Ehrh. black cherry
Quercus sp., red oak group water or laurel oak
Q. virginiana Mill. live oak
? Rhus sp. sumac
Salix caroliniana Michx. Coastal Plain
Sapindus marginatus Willd. soapberry
Ulmus americana L. American elm
Ulmaceae, ? Celtis sp. sugarberry
Vitis sp. grape vine
Unidentified hardwood no. 1, 1984 ? Bursera sp.
Unidentified hardwood no. 2, 1985 tropical wood
Unidentified hardwood no. 3, 1985 tropical wood
Table 4. Plants Recorded in Early Historic Period Documents.*
Plants described as foods:
1. Plants managed by the Indians.
maize (corn) CV, SM, F, R, TH, D, L,
JR, H, KE
squash . . . . . . KE, TH, JR
pumpkins . . . . . . . CV, KE, R, TH, H, D, JR, L
gourds TH, D, JR, L
kidney beans .. ... R
beans . . . . . . . CV, KE, TH, H, JR, L
"melden" (possibly Polygonum spp.) . TH, ? L
"wickonzowr"/"peaze" (planted peas). TH "plantafolis" (planted with corn) . TH "uppowoc" (tobacco) . . . . TH
2. Plants gathered from the wild.
roots, water plant . . . . . CV, TH
blackberries . . . .. . .. CV, KE, R, H, L
walnuts . . . . . ... CV, KE, R, TH, JR
prickly pears . . . . . . CV
mulberries . . . . .. . .. CV, KE, TH, JR, R
cocoplums . . . . . . SM
palmetto berries, "roots" . .. SM, D, L
acorns, five types . . . . . SM, KE, R, TH, L
arrowroot/coontie (Zamia pumilia) . F, L
root (prob. Apios sp., groundnut). . F, ? TH
plums . . . ............... KE, H, L
"medlar"/"ameixas" (persimmon) . . KE, TH, H, L
grapes . KE, TH, H, JR, L
pecan/hickory . . . . . . KE, TH
chestnut KE, TH, R
onions . . KE, R
peas, native TH
roots, general JR, L
"opeepenauk," a dry ground root/tuber. TH "kaishucqenauk," a root/tuber . . TH "tfinaw," catbrier (Smilax sp.) . TH "coscushaw," a root . . . . TH "habascon," a root . . . . . TH "metaquesunnauk" (? cactus) . .. TH
"sacquenummener," berries from a
water plant (? waterlily) . . TH
reed with seeds (prob. switch cane) TH
blueberries .. H, L
"graynes and herbes" . . . . H, JR
cassina drink ........ . H, D, L
"maste" (probably nut foods) . . H
"simples" (? plums, cherries) . . H, JR
Table 4 continued.
"mil" or "mill," "graines of Mill" . H
palm berries, roots (cabbage palm) . D, L
"peasen" (?) JR
"sorrel-like plant in Indian fields"
(? goosefoot/amaranth) . . . L
"pinocqs," waterlily seeds . . . L "chinaroot bush" L
Plants mentioned in descriptions of "the country:"
walnut and hickory trees . . . CV, TH, H, L
"laurel" (probably bay tree) . . CV "liquid-amber" (sweetgum) . . . CV
cedars CV, TH, H, L
"savins" (?) CV
evergreen oak (live oak) . . . CV, L
red oaks CV
oak, unspecified . . . . . TH, H, L
pines CV, TH, H, L
"palmitos" (palmetto) . . . . CV
sassafras TH, SM
mulberry trees . . . . . TH, H, L
fir trees TH
"rakiock," (for canoe manufacture) . TH
cypress trees .. TH, H, L
maple trees TH
witch-hazel . TH
holly .. TH, H, L
willow ............... TH
reeds TH, L
beech . . .* TH
ash o..... . . TH
elm . . . . ....... TH
"ascopo" (probably Zanthoxylem sp.) TH
black cherry . . . . H, JR, L
"lentiskes" (?) H
chestnut trees . . .......... . H, L
bays . . . . . . . . H, L
palm trees . . . . . . .. H, L
magnolias L grape vines . .. ..... L
Spanish moss . . ........ L
"chinaroot bush" (?) . . . . L
*Key: SM=Sols 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).
Table 5. Wild Plants Having Potential Subsistence Value.
Plant Species Part Used
Acer sp. (maple) sap
Amaranthus sp. (amaranth) seeds, greens
Amaranthaceae (amaranth family) seeds, greens
Arundinaria gigantea (switch cane) seeds, shoots
Brasenia schreberi (water shield) roots, tubers, leaves
Capsella sp. (shepherd's purse) greens, seedpods
Carya spp. (hickory) nuts
Celtis laevigata (sugarberry) fruit
Cenopodiaceae (goosefoot family) seeds, greens
Cyperus sp. (nut sedge) tubers
Diospyros virginiana (persimmon) fruit
Gaylussacia sp. (huckleberry) fruit
Nuphar lutea (spatterdock) tubers, seeds
Nymphaea sp. (waterlily) tubers, seeds
Nyssa sylvatica (black gum) fruit
Passiflora incarnata (may pop) fruit
Persea borbonia (red bay) leaves
Physalis sp. (groundcherry) mature fruit
Phytolacca americana (pokeweed) young shoots and greens
Pinus spp. (pine) young shoots, young male
cones, inner bark
Polygonum spp. (smartweed/knotweed) seeds, whole plant
Polyporous sulfureous (sulphur shelf) tender edges of thallus
Portulacca oleracea (common purslane) stems, leaves, seeds Prunus serotina (black cherry) fruit
Quercus spp. (oak) nuts
Rhus sp. (sumac) fruit
Rubus spp. (blackberry) fruit, young shoots, leaves
Rumex sp. (dock) greens, stems
Sabal Rgalmeto (cabbage palm) fruit, buds
Sambucus canadensis (elderberry) fruit
Sapindus marginatus (soapberry) fruit
Scirpus spp. (bulrush) seeds, shoots, tubers
Serenoa repens (saw palmetto) fruit, terminal bud
Setaria sp. (bristlegrass) seeds
Smilax sp. (greenbrier) young shoots, leaves, tubers
Vaccinium sp. (blueberry) fruit
Vitaceae (grape family) 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 Hontoon Island as it was among
Table 5 occur naturally in the other Southeastern Indians
marsh/river swamp ecotone where (Swanton 1946:265, 286). Persimmon
the midden is located, care must was mentioned by several early
be exercised in interpreting them explorers (see Table 4), being
as food remains as their pres- referred to as "medlars" or
ence in the archaeological record "ameixas," and perhaps also "simmay be more incidental than from ples" by the French. A medlar is
deliberate use, given the quality a European fruit in the rose famof preservation and natural seed ily--Mespilus germanica (Tutin et
rain. The fact that most of the al. 1968), which is very similar
seeds show no evidence of human to the native persimmon. Hariot's
manipulation or alteration, such description leaves little doubt as
as being charred, broken (or to which fruit he is referring:
ground), or clustered in caches,
tends to support a conclusion of Medlars a kind of verie
their natural intrusion in the good fruit, so called by
site, although it does not pre- us chieflie for these
clude the possible use of certain respectes: first in that
of these plants (e.g., amaranth they are not good until
and spatterdock) for greens, tub- they be rotten: then in
ers, or other plant parts (cf. that they open at the head
Peterson 1977). These parts would as our medlars, and are
have been consumed in their about the same bigness:
entirety, leaving the seeds as otherwise in taste and
unused residue. In the absence of colour they are farre
strong contextual data, much of different for they are as
the preliminary interpretation has red as cheries and very
been based on analogy with ethno- sweet: but whereas the
historic and ethnobotanical infor- cherie is sharpe sweet,
mation, keeping in mind certain they are lushious sweet.
problems with doing so (cf. Gould (Hariot 1951:D)
1978; Gould and Watson 1982;
Cordell 1977). In addition to ethnohistoric
documentation, evidence for the
Wild Plants Utilized for Food. utilization of those plants menThe wild plant resources for which tioned above comes in the form of:
direct evidence of usage (subsis- (1) the condition of the remains
tence) exists are: hickory, oak, themselves (i.e. charring, breakmay pop, grape, saw palmetto, and age); and (2) in their relative
cabbage palm. In addition, at quantities, keeping in mind that
least thirteen more plants are certain plants are more prolific
believed to have contributed to than others at a given time. Since
subsistence, though the evidence no firehearths or features were
is less clear. Each of these is defined and excavated as separate
described below, units, precise contextual evidence
as to utilization is presently
Persimmon seeds and portions of lacking.
pericarp were recovered in the
1985 unit excavations and column Tables 6 and 7 show frequencies
sample (one seed). Though not and/or volumes of the most confound in abundance (7 seeds spicuous plant remains identified
total), persimmon was undoubtedly in the column samples (excluding
valued as a food resource at cultivated plants which will be
Table 6. Most Conspicuous Plant Remains (Seeds, Nuts, Misc.) From the 1982 Column Sample.
LEVELS 5 6 7 8 9 10 11 12 13 14
Oak 1.6 3.4 4.0 1.2 3.0 4.6 14.4 38.0 7.0 7.1
Hickory 5.0 11.5 32.0 50.0 10.0 2.0 8.0 72.0 16.0 50.0
Cabbage palm 4 207 71 27 15 20 57 35 25 98
Saw palmetto 0 0 3 2 9 0 11 0 2 0
Persinmmon 0 0 0 0 0 0 0 0 0 0
Huckleberry 0 0 0 0 1 0 0 0 1 1
Blueberry 0 0 0 0 1 0 0 0 0 0
Elderberry 8 2 3 1 1 5 0 1 1 4
Blackberry 1 0 0 0 2 2 0 0 1 0
lay pop 10 3 0 3 0 1 0 1 1 2
Grape 53 38 8 52 18 2 2 5 9 28
Peppervine 0 54 26 28 26 8 0 1 4 28
Black gum 2 36 130 37 2 0 1 0 0 1
Groundcnerry 0 0 0 0 0 0 1 0 0 0
Amaranth 0 0 0 0 0 3 4 10 0 0
Amaranthaceae 0 0 0 0 12 15 8 30 9 3
Chenopodiaceae 0 0 0 0 0 0 0 0 0 0
Bristlegrass 0 0 0 0 0 4 2 0 4 0
Pokeweed 0 10 5 1 11 13 8 19 20 22
Broomweed 0 0 0 0 0 0 0 0 0 0
Portulacca 0 0 0 0 0 0 0 0 0 0
Smartweed/ 0 0 2 1 13 27 10 33 17 48
Bulrush 0 0 1 0 9 28 18 30 12 30
hIut sedge 0 0 1 0 11 16 11 0 20 3
Buttonbush 0 0 0 0 3 0 0 0 0 0
Water shield 0 0 0 0 9 35 17 3 13 35
Spatterdock 6 2 8 34 40 94 78 46 14 52
*Nut remains are represented by volume only (in cubic centimeters); other plants by numerical count.
Table 7. Most Conspicuous Plant Remains (Seeds, Nuts, Misc.) From the 1985 Column Sample.
LEVELS 7 8 9 10
Oak 1.4 61.7 97.0 15.0
Hickory 3.1 10.0 29.0 34.0
Cabbage palm 1 219 268 138
Saw palmetto 1 2 15 7
Persinunon 0 1 0 0
Huckleberry 1 34 18 6
Blueberry 1 14 8 2
Elderberry 1 6 4 2
Blackberry 0 5 6 0
May pop 0 0 0 1
Grape 7 4 1 1
Peppervine 0 0 0 0
Black gum 0 0 0 1
Groundcherry 2 7 3 3
Amaranth 0 23 25 64
Amaranthaceae 3 475 59 90
Chenopodiaceae 0 2 12 2
Bristlegrass 22 181 178 56
Pokeweed 7 13 33 5
Broomweed 0 38 11 0
Portulacca 0 1 2 4
Smartweed/knotweed 492 5 4 10
Bulrush 209 7 2 2
Nut sedge 2 22 27 18
Buttonbush 216 0 0 0
Water shield 16 5 0 7
Spatterdock 3 2 0 0
*Nut remains are represented by volume only (in cubic centimeters); other plants by numerical count.
described separately). Nut region) red oaks. This is because
remains, specifically oak and acorns from white oaks can be
hickory, are represented by volume eaten directly, whereas those from
since their high degree of frag- red oaks, which have a high tannin
mentation renders a numerical content, must first be subjected
value meaningless. The rest of to a long leaching process (Larson
the plant remains are represented 1980:188). These species curby absolute number. rently range north of Hontoon
Island, making their presence in
Nut remains were among the most the samples especially interabundant wild plant resources in esting. Two (or three) additional
all tests from the site. Acorns oak species were identified:
appear to have been more inten- Quercus virginiana (live oak),
sively used in later time periods which is also in the white oak
corresponding with the mussel group, and either, or perhaps
shell midden. Acorn had a volume both, Q. laurifolia and Q. nigra
of about half that of hickory in (laurel and water oaks)--their
the lower snail shell strata acorns being very difficult to
(Levels 10-14, 1982 column) but distinguish. These latter two
increased to more than twice as species are members of the red oak
much as hickory in the mussel group. Hull fragments from live
shell layers (in particular the oak and laurel/water oak acorns
1985 column and larger excavation were recovered throughout all
units) (Tables 6 and 7). Hickory temporal levels of the site.
does not seem to follow as clear a
trend as acorn and was probably a Acorns were often mentioned in
less important food. At least a ethnohistoric accounts (cf. Sol~s
third of the hickory remains are de Meres 1964:175, 226; Hariot
from the bitter-tasting water 1951:DI-D2; Ranjel 1922:107). They
hickory (Carya aquatica, and were used by the Indians in a
these show no sign of use, indi- variety of ways. Acorns were
cating deposition under strictly boiled to extract their oil-natural conditions. The apparent Hariot (1951:Dl-D2) recorded three
increase in acorn remains from species which he says were preHontoon Island runs contrary to a ferred for this purpose. The nut
general pattern outlined by meats were dried on "hurdles of
Yarnell and Black (1985:97) for reeds" over fires so that they
other roughly contemporaneous could be stored for later consumpcultural areas in the southeastern tion, at which time they were
United States, although Alexander soaked in water and then either
(1984a:22-24) documented a similar eaten directly, or pulverized and
increase in acorn usage at three made into "loaves or lumpes of
Fort Walton Period sites in Leon bread" (Hariot 1951:DI-D2).
County, Florida. Hickory nuts and walnuts were used
in similar fashion: that is, for
Two oak species occurred exclu- direct consumption, in a bread, or
sively in the mussel shell strata: for oil content (Cabeza de Vaca
Quercus michauxii (swamp chestnut 1871; Hariot 1951:11; Knight of
oak), and _. lvrata (overcup oak). Elvas 1922:74, 221; Swanton 1946:
These species bear exceptionally 365 -366).
large acorns and are members of
the white oak group (Sargent 1965: Most of the acorn recovered was in
240), which would make them pref- the form of uncharred, fragmented
erable to the common (in this hull, with the exception of one
intact charred nut meat of the and 1697, were cabbage palm (and/
laurel/water oak type, and a few or the saw palmetto). The new
charred hull fragments from the shoots of the palm are also edible
1985 column sample. These, we (Swanton 1946:284).
might infer, were accidentally
charred, perhaps during the drying Plants Associated Primarily With
process described above. Very the Mussel Shell Levels. At
little hickory shell was charred. least ten other plants underwent
the same temporal pattern of
After acorn and hickory, cabbage increase and decline as palm and
palm was the next most abundant acorn. These include: amaranth
wild plant resource recovered. (Amaranthus sp.), and an addiLike acorn, cabbage palm under- tional member of the amaranth
went a substantial increase in the family (Amaranthaceae), bristlesuperficial levels. It averaged grass (Setaria sp.), nut sedge
47 fruits per column sample level (Cyperu sp.), huckleberry (Gayin the lower five levels (snail lussacia sp.), blueberry (Vaccishell midden), whereas in the nium sp.), pokeweed (Phytolacca
mussel shell strata (evidenced americana), and at least three
most clearly in the 1985 column), more--Chenopodiaceae (goosefoot
palm averaged 208 fruits per family), groundcherry (Physalis
level, with the exclusion of sp.), and elderberry (Sambucus
uppermost Level 7, where it under- sp.)--though these latter three
went a marked reduction in number were far less abundant than the
(Tables 6 and 7). Correcting for others. All of these are potenthe smaller size of the 1985 tial food plants (see Table 5).
column as compared to the 1982 The amaranths and Chenopodiaceae
column, this increase in abundance provide both edible seeds and
is even more dramatic--averaging greens, nut sedge has edible
416 fruits per level in the mussel tubers, pokeweed has edible greens
shell layers. This increase when they are young, and fresh
could, in reality, be more a fruit could be had from the
factor of the particular location huckleberry, blueberry, groundor placement of the excavation cherry, and elderberry (Peterson
units since cabbage palm is an 1977). Blueberry is mentioned dirabundant producer of fruit and ectly in ethnohistoric documents
perhaps a tree(s) was once near (e.g., Hakluyt 1909:451), and the
the location of the 1985 units. An descriptions could apply to both
acultural explanation does not blueberry and huckleberry since
appear to be indicated though, they are closely related plants
since at least eleven additional and are very similar in appearplants, including oak, underwent ance. There are vague references
an identical pattern of increase to weedy plants growing in agriduring the historic period, fol- cultural fields (Hariot 1951:C2lowed by a sudden marked reduc- C3; Laudonni~re 1975:122) which
tion in the early-middle eight- could apply to the amaranths and
eenth century. The fruits of goosefoot, but also to a few other
cabbage palm are edible, and can plants which frequent tilled soil.
be eaten fresh (Larson 1980:204- "Roots" or tubers (from nut sedge,
205; Swanton 1946:283). It may be for example) were often described
that the "palm berries" consumed as foods by early explorers or
by Dickinson and his party (1945: settlers (Hakluyt 1909:451; Ribaut
61, 70) during their unfortunate 1964:73; Laudonniere 1975:122).
trek up the Florida coast in 1696 Hariot described at least five
such "Rootes," and in addition went an increase during the period
related their Indian names, habi- in which the mussel shell midden
tat, and how they were used was accumulating, and for some
(Hariot 1951:C4-C5). species, this increase was very
pronounced. Further, following
The identification of bristle- the expansion, each plant undergrass (etaria sp.) was made by went a simultaneous and perhaps
Dr. David Hall, Florida State rather sudden, decline or almost
Museum Herbarium, who is an expert complete disappearance by Level 7
in the grass family. This plant (the uppermost cultural stratum of
is not presently known to have the mussel shell midden, 1985
been used as a food resource by column). The fact that each one
southeastern Indians, but similar of these plants followed the same
grasses played very important pattern of abundance and expanroles among these people-- sion, followed by a marked reducespecially Phalaris spp. (may tion in number, is a good indicagrass), which has recently been tion that their presence in the
placed in a quazi-cultivar cate- samples might have been due to
gory (cf. Asch and Asch 1985; human agency and not simple coinCowen 1978; Yarnell and Black cidence.
1985:99). Setaria sp. has been
found in abundance in human copro- In addition, two other events
lites recovered from archaeolo- occurred in conjunction with those
gical sites in the Tehuacan Valley outlined above: (1) bone, charin Mexico (Bender 1975:188). There coal, wood, and ceramics also
its use as a food was apparently underwent a pronounced decrease in
long in duration. It too has been Level 7; and, (2) a second group
proposed as a possible candidate of plants underwent a reverse
for having been cultivated (in trend, which is a notable increase
Mexico) (Callen 1965). in Level 7 (see below). To illustrate the former, faunal remains
Of the eleven plants just men- had a volume of 1103 cc in Level 8
tioned (cabbage palm, amaranth, of the mussel shell midden (1985
Amaranthaceae, bristlegrass, nut column sample), but only 17 cc in
sedge, huckleberry, blueberry, Level 7. Likewise, nut shell
pokeweed, Chenopodiaceae, ground- dropped by a similar percentage,
cherry, and elderberry), the seeds and wood and charcoal dropped by
of only three--cabbage palm, poke- at least a half (see Table 2).
weed, and Chenopodiaceae--were
found charred and then only in Plants Indicating a Return to
very small percentages. But since Natural Conditions. In contrast
the subsistence value of these to those plants and other bioloplants rests primarily in their gical remains which decreased in
fresh fruits and greens, direct Level 7 of the 1985 column, at
contact with fire seldom should least three plant species
have occurred. increased in abundance. These are
buttonbush (Cephalanthus occidenStronger support for their infer- talis), knotweed (Polygonum spp.),
red function as subsistence items and bulrush (Scirpus spp.). The
(versus a strictly environmental habitat occupied by buttonbush is
inclusion in the samples) can be quite limited. It is a component
had with a consideration of their of the river swamp and occupies
temporal distributions and habi- the outer fringe of the swamp--the
tat. All of these plants under- zone in closest proximity to the
freshwater marsh and river. As east. The whole plant is edible
observed on the island today, (cf. Murray and Sheehan 1984:282buttonbush, along with Coastal 298). In limited support of this,
Plain willow (Salix caroliniana), is that some of the seeds from
forms dense thickets between the Hontoon Island are charred, but
lagoon and more open vegetation of this plant too, is interpreted
the inner, or landward, swamp (also) as an indicator of a return
zone. Neither plant is known to to natural swamp vegetation since
have any subsistence value, and it it became far more common by Level
is conceivable that they might 7. Additionally, a greater variety
have been partially cleared away of knotweed species appeared
during the occupation of the site (going from one species in the
to facilitate access to the lower levels to at least three in
lagoon. Adzed chips and/or cop- Level 7).
piced stems of both woods have
been identified in historic and
prehistoric levels of the site. To reiterate, an inverse relationSeeds were not recovered at all, ship was documented between twelve
except for the abundance of but- native plants (oak, cabbage palm,
tonbush seeds in Level 7 of the amaranth, Amaranthaceae, bristle1985 column sample (Table 7) and grass, nut sedge, huckleberry,
contemporaneous levels of the blueberry, pokeweed, Chenopodiaadjacent units. Subsequent to the ceae, groundcherry, and elderabandonment of the area by humans, berry) which were common in the
buttonbush (and willow) would have accreting stages of the mussel
quickly regained its position bor- shell midden and which were then
dering the lagoon, a likely reason drastically reduced in number by
for the abundance of buttonbush the final stages, and those plants
seeds in Level 7. Willow seeds are (buttonbush, bulrush, knotweed)
very small in size, plumed and which were relatively inconspiclight-weight for wind dispersal. uous or absent in the lower
Consequently, the chances of their levels, but then underwent an
being present are minimal, increase when the others declined.
Like the first group of plants,
The situation is slightly differ- other categories of remains were
ent for the other two plants, also abundant prior to Level 7 and
knotweed and bulrush, which then experienced the same decline
increased as buttonbush did. Like --all of which is taken as evibuttonbush, these plants inhabit dence that these plants were being
hydric environments--the river concentrated at the site during
swamp and marsh respectively--but the time prior to Level 7, after
unlike buttonbush, both have which the evidence for them is
edible parts. The seeds and shoots meager as the site saw a drastic
of bulrushes are edible, and their reduction in usage, or was abanpossible use by aboriginal inhabi- doned altogether. The fact that
tants of the island has been twelve plant candidates for human
considered, but the plant, like subsistence underwent a substanbuttonbush, appears to have under- tial reduction in number along
gone a recolonization or radiation with other subsistence remains
after, or as, humans left the (fauna) and items of material cularea. Knotweed was probably used ture, is an indirect indication
as a food at Hontoon Island, both that these particular plants were
in prehistoric and historic times, being used by the Indian occupants
as it was throughout the South- of the site.
There is a possible alternative soil and natural vegetation may be
explanation which could account the primary factor accounting for
for the more abundant presence of the progressively increased abunat least seven of these plants dance of these seven plants as
during site occupation, and site occupation continued. Then,
likewise, their subsequent decline logically, as humans quit the
as human presence lessened. Unlike area, and natural conditions began
oak, cabbage palm, blueberry, to prevail, their niche would
huckleberry, and elderberry, which begin to fade and a reduction in
normally occur in the hammock/ their numbers would be expected,
river swamp environment surround- as was observed for Level 7 and
ing the site, the other seven the overlying, culturally sterile
plants--amaranth, Amaranthaceae, layers.
bristlegrass, nut sedge, pokeweed, This second hypothesis, which
Chenopodiaceae, and groundcherry-- emphasizes life cycles and biology
characteristically inhabit of individual plants is just as
disturbed sites and two even have of in plants jutha
speces hic ted tobe oun unlikely an explanation for the
species which tend to be found on changing frequencies of the seven
shell middens (i.e., Setaria plants in question, as the first
macrosperma and Cyperus plani- hypothesis of direct human usage.
folius tWunderlin 1982:91, 1041). Even if it is eventually shown
What this means is that though that the second proposed hypothey might have occurred in the thesis better explains the plants'
vicinity of the site anyway, with thesis bette plistplns
the resnce f hman ther hbi-presence and temporal distributhe presence of humans their babi- tion, it must be noted that this
tat would have been greatly does not preclude their having
enhanced and expanded, and they been used as foods or otherwise by
might, therefore, occur in greater the Indian occupants. Their close
numbers. proximity and ready availability
would have made them attractive
This is a phenomenon known to candidates for use. Rindos (1984:
botanists as the "door-yard" or 138-143) and others (e.g.,
"dump-heap garden" effect (Harlan Fanr1 968 he potlae
and de Wet 1963:19; Rindos 1984: Flannery 1968) have postulated
128-34) Huan fot raficthat this is a likely avenue of
128-134). Human foot traffic, early plant domestication--taking
deliberate and/or unintentional efrm plant osoildisurbaces anda hmanthe form of a human-plant cosoil disturbances, and a human evolution. With this in mind,
induced concentration of nitrogen otrs h c to the nci on
and phosphates in the soil favor tht it i obl o simpl
growth of certain plants (weeds) that ay no and
which would otherwise be less able potnidmest amon the
to cmpee. uman unon-potential domestics among the
to compete. Humans uncon- suhatr nin r lnso
sciously (in a passive role) widen southeastern Indians are plants of
and strengthen their niche by this type (see for example,
alindtng tenteil nchet- bYarnell and Black 1985). Analysis
eitorsnt byhurning a d co t- of human coprolites would contriitors and by churning and enhanc- bt ra eltwr eovn
ing the soil in a way for which bute a great deal toward resolving
these plants are suited (or have questions surrounding the plants
become adapted). Thus, the
"anthropogenic environment" Other Potentially Utilized
(Rindos 1984:134) created with the Plants. Nine additional plants
formation of the midden and the also might have been utilized,
concomitant disturbances to the though they occurred in much lower
frequencies than the plants men- quencies (Tables 6 and 7). All
tioned above. Three occurred three plants have edible parts:
almost exclusively in the mussel the roots, tubers, and leaves of
shell levels, while three others water shield; mature fruit and
were more prominant in the snail terminal bud of saw palmetto; and,
shell midden. Finally, the last mature fruit of may pop (Peterson
three were quite scattered 1977). Water shield is a fairly
throughout all levels. The three common plant in the freshwater
which are associated with the marsh and its presence in the
superficial layers are Sida sp. samples is just as likely inci(Indian hemp or broomweed), black- dental as it is the result of
berry (Rubus spp.), and purslane human utilization. The seeds of
(Portulaca oleracea) (Tables 6 and saw palmetto though, were always
7). Those associated with the found charred, which is a good
underlying, snail shell midden are indication that it was used. The
grape (Vitis spp.), black gum "ripe berries on the palm shrubs"
(Nyssa sylvatica), and spatterdock [emphasis added] eaten by Jonathan
(Nuphar lutea), and lastly, those Dickinson and his fellow shipwreck
which are more evenly distributed victims during their travels along
are saw palmetto (Serenoa repens), the east Florida coastline
may pop (Passiflora incarnata), (Dickinson 1945:70) might have
and water shield (Brasenia come from the saw palmetto. The
schreberi) (Tables 6 and 7). members of the first Menendez
expedition ate palmetto berries
Both purslane and blackberry are (Solis de Meras 1964:127), and
food plants (Table 5), and occur Laudonniere (1975:123) mentioned
in disturbed habitats, including having eaten the "roots" (probably
human living and activity areas. young shoots) of palmettos while
Purslane was found exclusively in in Florida. Charred saw palmetto
the historic period Levels 8-10 of seeds have been identified from a
the 1985 column. Blackberry number of archaeological sites on
occurred in both historic and the Southeastern Coastal Plain,
prehistoric contexts (more in the including the Jungerman Site on
upper levels), though never in the Indian River just south of
abundance. Broomweed was found Cape Canaveral (approximately 112
only in Levels 8 and 9 of the 1985 km southeast of Hontoon Island)
column sample. Its potential (Jordan et al. 1963:17), and at
value, if any, is not known, St. Augustine, 112 km to the
though its alternative common northeast (Reitz and Scarry
name--Indian hemp--would suggest 1985:72).
that it could possibly have provided fibers for twine or the May pop is well known as a food
like. Since it too characteris- item among southeastern Indians,
tically inhabits disturbed sites particularly during the Misand hammocks, its presence alone sissippian and historic periods.
is not a viable indicator of Yarnell and Black (1985:99) have
importance to the Indians. suggested it be placed in a quazicultigen category. It is used as
Saw palmetto (Serenoa repens), may a fresh fruit, having a citruspop (Passiflora incarnata), and like pulp, but can be eaten and
water shield (Brasenia schreberi) processed in a variety of ways
were identified in both pre- (Peterson 1977:94-95; Spencer
historic and historic period con- 1985). Many of the seeds from
texts in roughly the same fre- Hontoon Island were charred,
suggesting perhaps that mature are very good meate . .
seeds were being expectorated into Coscushaw . it groweth
the fire during consumption of the in very muddie pooles and
pulp. In the immature state the moist groundes. Being
seeds are edible. dressed according to the
countrey manner, it maketh
The last three wild plants--grape a good bread, and also a
(Vitis spp.), black gum (Nyssa good sponemeate, and is
sylvatica), and spatterdock used very much by the inhab(Npha_ lutea), seem to have been itants: The juice of this
more important during the pre- root is poison, and therehistoric period, being more fore heede must be taken
abundant in the snail shell before any thing be made
midden. Spatterdock went from an there-withall: Either the
average of 54 seeds per column rootes must bee first sliced
sample level in the prehistoric and dried in the Sunne, or
midden, to approximately 5 per by the fire, and then being
level in the superficial mussel pounded into floure will
shell strata. This is a member of make a good bread . .
the water lily family (Nympha- (Hariot 1951:C4)
ceae), and both its tubers and
seeds are edible (Peterson 1977: The seeds of spatterdock are what
60-61). has actually been positively
identified in the archaeological
The use of the "roots" of water samples. Interestingly, keeping
plants by southeastern Indians was in mind Hariot's foregoing
mentioned by Thomas Hariot, and by description of the Indians'
Cabeza de Vaca of the Narvaez methods for preparing rootstock
expedition. Cabeza de Vaca (1871: for consumption, charred rhizomes
167, 103) recounted the use of have been recovered although they
"roots," some walnut-sized, and have not been securely identified
some smaller, that were "taken as to which plant they belonged.
from under the water" and eaten. If inclusion of spatterdock seeds
This could apply to spatterdock or in the deposits was a result of
to several other water plants, human activity (versus natural
including the nut sedge (Cypreus deposition from the marsh), then
sp.) mentioned earlier. Hariot we might account for their pres(1951:C4-C5) was more specific, ence if they were gathered uninthough again, it is difficult to tentionally during procurement of
determine to which water plant he rootstock or, more likely, both
was referring. He described six seeds and rootstock were used.
"rootes," at least two of which
were from wetland or marsh plants: Black gum occurs in wet environments including bayheads and river
Openauk are a kind of swamps, such as that in which the
roots of round forme, midden is located. This tree is a
some of the bignes of prolific producer of fruit and
walnuts, some far greater, tree rain could account for its
which are found in moist seeds in the samples. There is
and marish grounds growing nothing to suggest that its fruits
many together one by another were utilized at Hontoon, but
in ropes, or as thogh they ethnographic accounts suggest such
were fastnened with a string, evidence would be difficult to
Being boiled or sodden they recognize. According to Swanton
(1946:360), the fleshy pericarp some other purpose. Cones from at
was eaten fresh and the seeds then least three species have been
ejected, so that only by chance, identified (Table 3).
would the seeds be subjected to
fire, and in any case, carbon- Another unusual component in the
ization would not in itself be a deposits were bracket fungi, three
reliable indication that the fruit species of which have been idenwas actually used. Since the tified (Table 3). The use of
fruits are edible and occur in these fungi at the site, if at
abundance, it is likely that both all, is not understood, but they
cultural and acultural factors are known from ethnographies to
attributed to their inclusion in have been used as paints, tinder,
the samples. hallucinates, and more (Findlay
1982; Ison 1985). One species,
Grape was undoubtedly utilized by the "sulphur shelf," is edible
prehistoric and historic period (Peterson 1977:238). Shelf fungi
inhabitants of the island, though such as these have been identified
more seeds were recovered from from at least two other archaeoprehistoric levels. Four species logical sites in eastern North
were identified (Table 3) America. They occurred at a rock
(Watlington 1984, 1985). At least shelter in southwestern Missouri
four early chroniclers mentioned (Gilmore 1931:92) and in Archaic
grapes (Knight of Elvas 1922:82, Period deposits from the Windover
221; Hakluyt 1909:451; Ribaut Site near Titusville, Brevard
1964:72, 95; Hariot 1951:B2,Dl). County, Florida (Kimborough 1986).
They were eaten fresh or dried
over a fire for future consumption
(Speck 1909:45). Accordingly, a Also among the miscellaneous matenumber of grape seeds from Hontoon rials were Spanish moss (TillandIsland were charred. sia usneoides), and bundles of
fiber and pediole fragments from
Brief mention of a small group of the cabbage palm. A small amount
miscellaneous plant remains will of two-ply cordage was recovered
finish this discussion of the wild in the 1985 excavations which
plant species. Many of the sam- appears to be made from twined
ples from the columns and larger leaf fibers of cabbage palm and/or
units contained charred and a palmetto.
uncharred stem fragments of the
switch cane (Arundinaria gigan- Cultivated Plants
tica). This plant had many uses.
According to Hariot (1951), its At least three genera of cultiseeds were ground into meal, and vated plants have been identified
the cane was used for building in the archaeological samples from
material, arrow shafts, fish Hontoon Island (Table 3). The
weirs, drying racks, basketry, and seeds and rind fragments from
as a pipe (Ehrmann 1940:13). members of the family Cucurbitaceae (pumpkins, squashes, and
Pine cones and loose scales, char- gourds) are among the most volumired and uncharred, were another nous plant remains from the site.
common component of the archaeo- More than 1300 seeds and an abunlogical samples. This may indi- dance of rind fragments of Cucurcate the use of pine nuts for bita pepo have been recovered from
food, or perhaps the cones them- the column samples (Table 8). Many
selves were desired for tinder or more have come from the larger
70 excavation units. The other posi- and contemporary levels of the
tively identified cultivated larger units, came evidence of the
plants were bottle gourd (Lage- first two cultivated plants-naria siceraria) and corn (Zea bottle gourd and a small ornamay, mental gourd-like form of Cucurbita pepo (Figure 3). The bottleSeveral interesting temporal gourd seeds are quite small comtrends with regard to domesticated pared to modern analogues, and
species have been discovered, and closely resemble seeds from the
in particular, within the family Phillips Spring Site in Missouri
Cucurbitaceae (which includes both which date from around 4300 B.P.
C. pepo and Lagenaria siceraria). (Kay et al. 1980), and seeds from
From prehistoric levels at the Key Marco in Collier County, Florsite, bottle gourd and a small- ida (Cushing 1896; Cutler, in
seeded variety of C. pepo were Gilliland 1975:255-256) of roughly
identified. While in the historic the same time frame as Hontoon
period mussel shell layers, not Island. The small pepo, labeled
only were these earliest two cul- C. p Type One for convenience,
tivated plants found, but also two also possesses relatively small
to three more varieties of C. seeds and very thin, smooth-surpepo, corn, and possibly one addi- faced rind. Similar ornamental
tional non-native domesticated gourd-like pepo and bottle gourd
species (apple). remains have been identified from
other sites in Florida and the
Beginning with the earliest incep- Southeastern United States. The
tion of cultural material (ca. earliest date to the Late Archaic
A.D. 0) in the 1982 column sample and Early Woodland periods (cf.
Figure 3. Bottle Gourd (Lagenaria Siceraria) (top) and Cucurbita
pepo Type One (bottom). Small Squares on Grid Equal 2 mm.
Newsom and Decker 1986; Yarnell to length ratio. They have been
and Black 1985; Watson 1985:101- classified on the basis of numeri103; King 1985:76-97). cal analysis with cucurbits such
as the scallop and acorn squashes
The small, hard-shelled gourds (see Decker and Newsom 1987;
have been interpreted as (most Newsom and Decker 1986). Seeds
likely) weedy forms, probably more possessing a similar morphology
camp followers or semi-encouraged and dimensions were recovered from
plants than fully obligate culti- Key Marco (Newsom and Decker
gens (Yarnell and Black 1985:99, 1986). They also compare with a
104). Their use is believed to sample of larger (than the
have been more for a container ovifera-type) seeds from Salts
than for the edible seeds and Cave, Kentucky (refer to King
bitter flesh (Speck 1941; Heiser 1985:80 for discussion on similar
1985:66). This might explain the finds from the Southeast).
occurrence of the great numbers of
uncharred seeds from Hontoon At least one, probably two, addiIsland, if indeed, upon maturity tional Q. pep varieties appeared
the seeds and pulp were being at Hontoon Island coincident with,
separated from the rind and dis- or slightly later than C. pepo
carded. All of the rind fragments Type Two and likewise, are found
from Hontoon Island were from in the historic period levels
mature fruits, and in addition, (Figure 4). One of these, desigall but one seed were fully nated C. pepo Type Three, is dismature. Also, only two seeds (out tinctive in having a relatively
of more than 2000) were carbon- long, narrow seed. Ninety seven
ized. The bottle gourd was pro- such seeds have been recovered.
bably also used as a container or _. pepo Type Four is also large in
utensil, although its seeds, like size, but much wider than Type
those of the ornamental gourds, Three. The possibility of their
are edible and high in oil content belonging to the same variety has
(King 1985:77). Gourds as con- not been ruled out since Type Four
tainers were described by several could simply represent the extreme
early explorers including at least for width. Eleven Type Four seeds
one who visited Florida (Ribaut have been recovered. The seeds of
1964:73). types Three and Four could have
come from a pumpkin-like fruit(s)
The Cucurbita spp. seeds from the since they are similar to the
site clustered into four distinct seeds of modern pumpkins and since
groups based on size and overall they fit within Decker and
morphology. Seed measurements and Wilson's (1986) "pumpkin" genetic
analyses of the size groups are class based on numerical analysis
reported by Newsom (1986), and (Newsom and Decker 1986). Large
Newsom and Decker (1986). peduncles (stems), which also fall
within the pumpkin size range,
Seeds from a second form of C. have been recovered from the same
pepo, referred to as C. pepo Type contexts. Dickinson (1945:69)
Two (Figure 4), were recovered observed pumpkins growing in the
only from levels dating to the vicinity of Cape Canaveral, FlorContact Period and perhaps some- ida, only 64 km southeast of Honwhat earlier. These seeds are toon, during his coastal trek in
similar to those of the Type One 1696-1697. Large seeds much like
cucurbit, but are consistently Type Three from Hontoon Island
larger and have a different width were recovered from two late his-
. . . .. . + i ..
4 tor~ 1j
Figure 4. (a) Cucurbita pepo Type One (top) and C. pepo Type Two
(bottom); (b) C. pe.2 Type Three (top) and C. pepo Type
Four (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 pe 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
? Pyrus malus 1
*C. pepo Type Four was only identified in the larger excavation units.
toric period sites occupied by seeds in large quantity, from the
Missouri and Osage Indians dating fleshier, squash/pumpkin variefrom the mid-1600s to late 1700s. ties (as is the ease at pontoon
These are believed to have Island). In contrast to the large
resulted from, or been influenced quantities of seed and rind fragby, a Mexican variety of pumpkin/ ments from the small, ovifera-like
squash (Blake 1986:6). Type One cucurbit, which may have
been used as a container rather
No thick "pumpkin-like" rind has than for edible parts, the other
been discovered in association two (three) cucurbits (types Two,
with these latter two pepo seed Three, and ? Four) are represented
types. This could be a matter of by much less seed and apparently
preservation, but it is unlikely no rind or mesocarp.
because bottle gourd rind and the
thin rind (believed associated Another important southeastern
with Type One) were present. More cultivated plant, Zea mays (corn),
likely, this is a reflection of was identified in Hontoon Island
the manner in which the Indians samples. Like the larger pumpkin/
prepared and utilized the squash seeds, it occurred exclu(assumed) fleshier fruit(s). sively in the historic period
Pumpkins, "pompions," or squash horizon. Eighteen cobs or cob
were mentioned in virtually every fragments were recovered, along
ethnohistoric account describing with two kernels and a number of
the southeastern Indians (Table loose cupules and cupule fragments
4). The Knight of Elvas, in his (this is for the site as a whole,
description of the "fruits encoun- only 26.10 g total). All were
tered in Florida," made reference thoroughly carbonized, with the
to a fruit which might be inter- exception of two partially charred
preted as similar to modern cobs. Less than half of the cobs
"yellow squash": "everywhere in were judged adequate on which to
the country a fruit, the produce base measurements so that taxoof a plant like liqoacam, that is nomic placement is tenuous, but a
propogated by the Indians, having beginning since little is known
the appearance of the royal pear, about corn from this region. Half
with an agreeable smell and taste of the cobs have a row number of
. ." (1922:221). According to eight, and the other half are tenethnographic information, pepos row. The ten-row cobs averaged
were normally prepared by the 13.6 mm in diameter, and the
southeastern Indians in two ways. eight-row cobs averaged 11.8 mm
"Summer" or immature squash (e.g., (all measurements were taken from
yellow crookneck) was boiled and cob midsections). Based on cupule
eaten in its entirety much as is size, kernel morphology and size,
done today, while "winter" or and row number, the Hontoon Island
mature varieties (e.g., acorn corn falls within the range for
squash) were generally cut into the variety known as "Eastern
strips, the rind removed, and the Flint" (Reitz and Scarry 1985:59flesh then dried for consumption 60). A few of the cobs and one
at a later date (Swanton 1946:269, kernel were examined by Margaret
275, 360). The Knight of Elvas Scarry, who has spent a great deal
(1922:134) reported on the "roast- of time studying southeastern and
ing" of pumpkins as a preparative non-local varieties of corn. In
technique. Based on the foregoing her opinion, the Hontoon Island
discussion, we would not neces- samples were typical of Eastern
sarily expect to recover rind, or Flint. Both kernels exhibit the
cresent shape and smooth surface subsistence complexes throughout
characterizing this variety, the southeastern United States,
especially during the MissisCorn was very important to south- sippian Period (ca. post 900 A.D.)
eastern Indian subsistence. Like (Yarnell and Black 1985). Together
the cucurbits, it was mentioned in with beans (Phaseolus spp.), these
virtually every ethnohistoric domesticates comprised what has
account (Table 4). Processing been designated the "eastern
methods included grinding into Mexican crop complex" (Watson
meal and flour, and then making 1982). Bottle gourd and a "weedy"
bread or corn cakes (Ranjel 1922: form of Cucurbita pepo seem to
87; Hariot 1951:Cl; Knight of have made the earliest entries
Elvas 1922:38); the kernels were into the east, at least as early
sometimes parched and eaten as the late Archaic. Corn, beans,
(Ranjel 1922:96), or processed by and possibly other varieties of C.
"seething them whole until they be pepo--"the fleshy cultigen squash"
broken" (Hariot 1951:C); the known ethnohistorically (Yarnell
"floure" was boiled in water to and Black 1985:99)--appear in the
make a "pappe" (Hariot 1951:Cl); Southeast at much later dates.
and it was also dried and stored
for later use (Ranjel 1922:86; At first glance, the Hontoon
Knight of Elvas 1922:46). Island data appears to mirror this
sequence of domesticated plant
The presence of another possible entry and inferred usage (though
cultivated plant in the historic beans are missing) in the Southperiod samples is interesting. One east, but certain inconsistencies
thoroughly carbonized seed from a- have been recognized. As in the
plant in the rose family (Rosa- Southeast proper, the small, weedy
ceae) came from Level 9 of the cucurbit and bottle gourd were the
1985 column sample. It is in poor first to occur, and they may or
condition but has been tentatively may not have appeared at roughly
identified as cultivated apple the same time as elsewhere.
(Pyrus malus), an Old World Following these, corn began to
species. Since apples were a appear in other areas of the
likely cargo on ships bound for Southeast, during the Late Middle
Spanish Florida in the sixteenth Woodland, and became a very importhrough eighteenth centuries, the tant crop by Mississippian times.
presence of a seed at Hontoon It continued to dominate the subIsland is tenable, especially sistence base during the protogiven the nearness of the coast historic and historic periods.
and St. Augustine, and later, of Where Hontoon Island differs is
the missions. that there is no evidence for corn
until the middle to late historic,
In summation, at least five, pos- the Mission Period (being found
sibly as many as seven, domes- only in the middle to upper levels
ticated plants were identified. of the mussel shell midden and
These are corn, bottle gourd, at overlying shell-free midden). The
least three forms (possibly four) same situation was found for corn
of Cucurbita pp (pumpkin/squash/ at Mount Royal, another long term
gourd), and finally, a tentatively village site on the middle St.
identified domestic apple. Corn, Johns River (Jones, personal combottle gourd, and "pumpkin/squash" munication, 1986). In contrast,
are well known as having been corn has been recovered at least
important components of aboriginal as early as Weeden Island times
from other areas in Florida to the naturally occurring mix of
west (e.g., A.D. 955 + 85, Syca- exploited plants and animals was
more Site, A.D. 1000-1300, Patton adequate and dependable. This is
Leslie Site, Curlie Site (Milanich certainly true for areas further
1974:15; Cutler and Blake 19761). south of Hontoon, where the overThis conforms more with the rest all environment becomes increasof the Southeast. Beans are known ingly more dissimilar (Siglerto have been part of the south- Lavelle and Russo 1983:17). Given
eastern Indian diet by the Missis- the lushness and high productivity
sippian Period (Yarnell and Black of the natural environment within
1985:103), and at least as early the river basin, along with the
as the late Fort Walton Period nearness of coastal resources,
(ca. A.D. 1300-1500) in north- corn cultivation may not have been
western Florida (Alexander 1984; desirable or advantageous. These
Cutler and Blake 1976). This people had long since developed a
again conforms well with evidence fine-grained adaptation to the
from the core southeast. There- aquatic and terrestrial resources
fore, at Hontoon Island, and by of the basin.
inference east-central Florida in
general, the absence of corn prior The Middle River Basin is bisected
to the middle or late historic by Larson's (1980:213) inferred
period (even though it was being southern limit for agriculture
grown in other parts of the state (just south of Lake George), and
much earlier), along with the Hontoon Island is approximately 24
absence of beans, is perplexing. km south of this imaginary line.
Beans though, are difficult to The soils in the area are genertrace because processing and con- ally acidic and poorly suited to
sumption often entailed their agriculture, being mostly either
being made into gruels and eaten excessively drained, nearly
entirely. sterile sands or excessively wet
mucks with high organic content
Some possible answers are (U.S.D.A. Soil Conservation Sersuggested. (1) One or both of vice 1980). Indians are not known
these members of the corn-beans- to have fertilized their fields,
squash trinity may have been pre- which would probably have been
sent at earlier dates, but were necessary to grow corn in this
missed during testing due to the area on the scale that Indians to
vagaries of sampling, preserva- the north and west of the Middle
tion, and aboriginal processing River Basin did. Even if small
methods; or (2) they were never garden plots were maintained, the
present at all; that is, they may chances of producing enough corn
not have been a component of the for the maintenance of a large
subsistence practices employed by human population are minimal so
people inhabiting the Middle St. that the investment in time and
Johns River Basin. East-central energy may not have been warFlorida is both peripheral and rented. It is thus probable that
distinctive in terms of environ- full scale plant cultivation was
ment and culture from the South- not a particularly feasible econoeast as a whole, and even from mic choice for prehistoric people
other areas of Florida. Cultures residing in the Middle St. Johns
in east-central Florida should not River Basin. The question still
necessarily be expected to conform remains open. Based on ethnoor pursue the same spectra of historic documents (e.g., Fontanresources, especially if the eda 1854; Dickinson 1945) and the
archaeological record, limited as sixteenth century at least two
it is, it does not appear that missions were located close to
corn was under cultivation. Hontoon Island (Newsom 1986). The
missionaries in eastern Florida
The cucurbits present a different were urging the Indians to cultipicture. These plants, especially vate corn and other produce to
the early varieties, are much benefit the missions and to cause
easier to grow, requiring little the Indians to settle in one place
or no cultivation. They have (making the Indians more accesoften been described as weeds or sible and therefore, more easily
"camp followers" and will volun- Christianized). The missionaries
teer or sprout independently in also encouraged the importation of
mulch piles and abandoned gardens foodstuffs from other areas of
(King 1985:78-79). Unlike corn, Florida, especially the Apalachee
they will tolerate wet, organic Province (Sturtevant 1962:64). An
soils and have been observed grow- established trade of corn to these
ing untended in swamp hummocks missions presents the possibility
(Stephens, personal communication, that Hontoon Islanders did not
1985). Appending cucurbits to the cultivate corn at all, especially
aboriginal inventory of utilized considering the small quantities
wild plants would have neces- recovered archaeologically at
sitated little effort or change in Hontoon and nearby Mount Royal
subsistence or other activities, (Jones, personal communication,
and indeed Cucurbitaceae were 1986). Further, in 1602 the
present at least as early as Freshwater Timucua were described
A.D. 0 at Hontoon Island. by the Governor of Florida as
primitives who only subsisted on
Corn does occur, along with the fish (and by inference, did not
larger pepos, at Hontoon in strata cultivate plants) (Arnade 1951:
dating to the middle to late 71).
1600s. Certain extraneous factors
may have played a role in this. In addition to importing foodOne, by this time a great deal of stuffs from within Florida, food
Indian movement, especially from was also being delivered to the
the north, had begun. As popu- Florida missions and garrisons
lations were decimated by disease, from Cuba and Mexico. In evidence
and balances of power and alli- of this, Scarry has identified
ances reorganized, people were exotic New World cultivars from
fleeing impact areas, migrating, sixteenth century St. Augustine
and reassociating into new groups (Reitz and Scarry 1985:64). This
(Milanich and Proctor 1978). If might serve as an explanation for
some of these displaced people the unusually long, narrow squash
moved up the St. Johns River, they seeds (Type Three) from Hontoon
would have carried ideas about Island if Blake (1986) is correct
subsistence, including techniques, in identifying seeds of such size
equipment, and even seed stock and shape with Mexican varieties.
from cultivated plants. This
could account for the introduction The presence of cultivated plants
of corn and new cucurbit types at at Hontoon Island has raised more
Hontoon. questions than provided answers.
At present we cannot assess how
Another possible influence may they might have impacted tradihave been through the Spanish tional foodways--whether they
missions. By the middle to late provided a minor supplement to a
subsistence system based on wild A minimum of eleven wild plants
plants and animals, or if they are believed to have furnished
might have altered or have been food during the occupation of the
increasingly incorporated into the site. One of the most significant
existing subsistence system. questions generated from the
investigations at Hontoon Island,
SUMMARY AND DISCUSSION is when and why did corn and the
obligate cultivated pumpkin/
The plant remains occurring in two squashes enter the middle St.
column samples from Hontoon Island Johns region. These plants are
have been identified and analyzed. only found in the upper levels at
The resulting information has been the site--the middle to late
supplemented with data from the historic occupation. Their
larger, coarse-screened exca- appearance is accompanied by the
vation units when appropriate first occurrence of some addi(e.g., corn remains). Plants were tional wild plants (overcup and
important to the river-oriented swamp chestnut oaks, broomweed,
inhabitants of Hontoon Island and and purslane), along with other
they undoubtedly supplied a major changes in the plant assemblage.
portion of the subsistence base. Overall, both wild and cultivated
plants appear to have been used
Marked differences in relative with greater intensity in the
frequencies and overall species historic period and a shift from
content exist between the two emphasis on aquatic/wetland
column samples. The 1982 column, species to terrestrial plants is
which primarily represents the evident. A comparable shift was
prehistoric period occupation, is observed in the faunal material-dominated by wild plants, parti- going from 94%-99% aquatics in the
cularly marsh/river swamp natives, lower levels to only 75% aquatic
Oak and hickory, cabbage palm, vertebrates (by weight of their
grape, peppervine, black gum, remains) in the upper levels (Wing
Polygonum spp., bulrush, nut 1986). A change in the size of
sedge, water shield, spatterdock, fishes caught was also docuAmaranthaceae, and pokeweed are mented, probably indicating the
abundant in this column. Two adoption or addition of new techdomesticated plants also occur in nologies or methods of capture.
the prehistoric levels: a small, The shift to terrestrial animals
thin-shelled gourd (Cucurbita pepo could at least partially be corcf. var. ovifera), and bottle related with the expansion of
gourd (Lagenaria siceraria). gardening/horticultural pursuits.
Gardens are known to draw various
Some of the same plants continued animals, resulting in an opporto be prevalent in historic period tunistic sort of hunting (see
deposits, but in general, a dif- Linares 1976 for a complete
ferent mix of species dominates, discussion). The point is that
In the 1985 column, successional changes in the subsistence base
or disturbed-habitat species are and technology were occurring
more common than the marsh/river simultaneously, and on several
swamp dominants of the first levels.
column. Three additional culti- The question remains why the
vated species appeared in these changes occurred at all, and what
levels--corn, and at least two provided the impetus for these
varieties of pumpkin/squash, changes. Human beings are gen-
erally most resistant to change St. Johns Culture-descendant
where subsistence is concerned, Freshwater Timucua were still
often relying on a number of living at the site, whether
alternate choices or feedback reduced in number or not,
loops which work to counteract primarily because ceramics
change (Cowen 1985:226-228; remained unchanged. There were
Flannery 1968). Therefore, change changes, though, in other cateoften results only under very gories of the artifact assemblage,
stressful conditions such as the most notably in the appearance of
failure of a primary resource or new art styles on bone carvings
human population expansion (Purdy, this volume)--perhaps an
stretching the limits and indication that disruptions occurpotentiality of the environment, red not only on the infrastrucand thus necessitating the tural level, but on other levels
adoption of new techniques to as well (i.e., ideological
boost production (Butzer 1982:281- systems).
294, 301-318; Cowen 1985:228-242;
Flannery 1968, 1973; Hassan 1978: If members of other native peoples
249-263). were being incorporated into the
group of inhabitants at Hontoon
That the environment had ample Island, they would have influenced
food resources appears to be in or have had the potential to
evidence at Hontoon Island, based encourage a redirection of subon the richness of the plant and sistence technologies and the
animal species in the archae- resource base. Also, people
ological record. The change in themselves do not necessarily have
shellfish could be an indication to move to influence change; the
of a failure of an important movement of ideas and commodities
resource, but enough alternative through down-the-line exchange and
providers existed that this should other forms of trade have also
not have stressed the exploitive been cited as primary directives
system to the point that a whole of culture change (cf. Binford
range of alterations were insti- 1972:195-197, 252-258). Another
tuted. A normally reproducing potential motivator of change was
human population would have under- the Spanish Mission system itself.
gone an increase over the fifteenhundred-plus years of occupation In conclusion, we as yet do not
at the site, but this must have know if Hontoon Island is typical
been checked during the historic or atypical of the rest of the
period due to the rapid spread of Middle River Basin, or even of the
European disease and other social St. Johns Culture and descendant
factors (cf. Crosby 1972). So, Timucua as a whole. That the
likewise, population expansion and people of Hontoon Island seem to
its concomitant stresses, does not stray from the norm for Timucuan
appear to be a viable hypothesis groups, which relied heavily on
to explain the cultural disrup- corn-beans-squash agriculture
tions or changes which occurred at (Deagan 1978; Milanich 1978), is
Hontoon beginning in the historic already apparent. The Hontoon
period. Islanders were, perhaps, more like
non-horticultural Indian groups
An hypothesis positing an influx (e.g., Ais) to the south, who
or immigration of people and ideas practiced fully hunting-gatheringis under consideration. It is fishing economies. The analysis
reasonable to assume that the of the plant remains from Hontoon
Island, limited as it is, is a Asch, D. L. and N. B. Asch
step toward the reconstruction of 1985 Prehistoric Plant Cultivation in
human-plant interactions in east- West-Central Illinois. In Precentral Florida and the under- America, R. I. Ford, ed., pp. 149standing of selective human 204. Anthropological Papers,
Museum of Anthropology, Univ.
behavior in such a resource-rich of Michigan No. 75. Ann Arbor.
Acknowledgments 1975 Farming in Prehistory: From
Hunter-Gatherer to Food Producer.
St. Martin's Press, New York.
I would like to thank David W.
Hall, Brenda J. Sigler-Eisenberg, Binford, L. R.
1972 An Archaeological Perspective.
Barbara A. Purdy, and Elizabeth S. Seminar Press, New York.
Wing, each of whom read preliminary drafts of this paper and Blake, L. W.
1986 Cultivated Plant Remains from
offered suggestions and comments. Historic Missouri and Osage
Several people assisted in some of Indian Sites. Paper presented at
the more difficult identifica- the Fifty-First Annual Meeting of
the Society for American Archaeotions, including David W. Hall of logy, New Orleans, Louisiana.
the Florida State Museum Herbarium, Regis B. Miller of the Wood Brinton, D. C.
Products Laboratory, Madison, Wis- 1859 The Florida Peninsula, its Literary History, Indian Tribes,
consin, Margaret Scarry of the and Antiquities. J. Sabin Co.,
State of Florida Bureau of Archae- Philadelphia.
ological Research, Gayle Fritz, Butzer, K. W.
Smithsonian Institution, Hugh 1982 Archaeology as Human Ecology:
Cutler, formerly of the Missouri Method and Theory for a ConBotanical Garden and Washington textual Approach. Cambridge UniBotnicl Grde an Wahintonversity Press, Mew York. University, and Deena S. Decker,
Marie Selby Botanical Gardens, Cabeza de Vaca, A. N.
1871 Relation of Alvar Nunez Cabeza
Sarasota, Florida. To these de Vaca. Trans. by Buckingham
people and others not mentioned, Smith. J. Munsell, New York.
and to the University of Florida,
Department of Anthropology, I am Callen, E. 0.
1965 Food Habits of Some Pre-Columbian
grateful. Mexican Indians. Economic
REFERENCES CITED Coon, N.
1974 The Dictionary of Useful Plants.
Alexander, M. M. Rodale Press, Rodale, Penn.
1984 Analysis of Plant Materials from
Three Fort Walton Sites: High Cordell, L. S.
Ridge, Velda, and Lake Jackson. 1977 Late Anasazi Farming and Hunting
Paper presented at the Forty-First Strategies: One Example of a ProbAnnual Meeting of the Southeastern lem in Congruence. American AntiArchaeologicl Conference, Pensa- quity 42(3):449-461.
Cowen, C. W.
Angler, B. 1978 The Prehistoric Use and Distri1980 A Field Guide to Edible Wild bution of Maygrass in Eastern
Plants. Stackpole Books, Harris- North America: Cultural and Phytoburg, Pennsylvania. geographical Implications. In The
Nature and Status of Ethnobotany,
Arnade, C. W. R. I. Ford, ed., pp. 263-288.
1959 Florida on Trial: 1593-1602. An Anthropological Papers, Museum of
Eyewitness Account of Life in Flor- Anthropology, Univ. of Michigan
ida Datelined at St. Augustine in No. 67. Ann Arbor.
1602. Univ. of Miami Hispanic
American Studies No. 16. Univ. of 1985 Understanding the Evolution of
Miami Press, Coral Gables. Plant Husbandry in Eastern North
America: Lessons from Botany, Flannery, K. V.
Ethnography, and Archaeology. In 1968 Archaeological Systems Theory and 81
Prehistoric Food Production in Early Mesoamerica. In AnthropoNorth America, R. 1. Ford, ed., logical Archaeology in the Ameripp. 205-244. Anthropological cas, B. J. Meggers, ed., pp. 67Papers, Museum of Anthropology, 87. Anthropological Society of
Univ. of Michigan No. 75. Washington, Washington, D. C.
Crosby, A.W. 1973 The Origins of Agriculture.
1972by, Te C a EAnnual Review of Anthropology
1972 The Columbian Exchange: Biolo- 2:271-310.
gical and Cultural Consequences
of 1492. Greenwood Press, West- Fontaneda, Do d'Escalente
port. 1854 Memoir of Do d'Escalente Fontaneda Respecting Florida. WritCushing, F. H. ten in Spain about the Year 1575.
1896 Exploration of Ancient Key Dwel- Trans. by Buckingham Smith. Washlet's Remains on the Gulf Coast ington, D. C.
of Florida. Proceedings of the
American Philosophical Society Gilmore, M. R.
of Philadelphia Volume 35, No. 153. 1931 The Vegetal Remains of the Ozark
Bluff-Dweller Culture. Papers of
Cutler, H.C. the Michigan Academy of Science,
1975 Two Kinds of Gourds from Key Marco. Arts, and Letters Vol. 24. Univ.
In The Material Culture of Key of Michigan Press, Ann Arbor.
Marco, Florida by M. S. Gilliland,
Appendix D, pp. 255-256. University Could, R. A.
Presses of Florida, Gainesville. 1978 Beyond Analogy in Ethnoarchaeology.
In Explorations in EthnoarchaeoCutler, H. C. and L. W. Blake loy R. 1. Could, ed., pp. 2491976 Plants from Archaeological Sites 293. Univ. of New Mexico Press,
East of the Rockies. American Albuquerque.
Archaeology Reports No. 1. University microfiche, L. C. #49422. Gould, R. A. and P. J. Watson
Deagan, K. A, 1982 A Dialogue on the Meaning and Use
1978n, Cult s iof Analogy in Ethnoarchaeological 1978 Cultures in Transition: Fusion and Reasoning. Journal of AnthropoAssimilation among the Eastern Tim- logical Archaeology 1:355-381.
ucua. In Tacachale, Essays on the Indians of Florida and Southeastern
Georgia During the Historic Period, 1909 Ta
J. T. Milanich and S. Proctor, eds.
Florida State Museum Ripley P. ages, Traffiques, and Discoveries
Bullen Monographs in Anthropology of the English Nation. Made by
and History No. 1, University Sea or Over-land to the Remote
Presses of Florida, Gainesville. and Farthest Distant Quarters of
the Earth volumes 8 and 9. Trans.
Decker, D. S. and L. A. Newsom by Richard Hakluyt of the (1587)
1987 Numerical Analysis of Archaeo- Narrative Notable Historie Conlogical Cucurbit Seeds from Hon- taining Foure Voyages Made by
toon Island, Florida. Paper pre- Certaine French Captaynes into
sented at the 10th annual confer- Florida. James MacLehose and
ence of the Society for E[hno- Sons, Glasgow, Scotland.
biology. Gainesville, Florida.
Decker, D. S. and H. D. Wilson 1951 A Brief and True Report of the
1986 Numerical Analysis of Seed Mor- Newfound Land of Virginia. Facphology in Cucurbita simile edition of the 1588 Quarto Systematic Botany 11:595-607. with an introduction by Randolph
G. Adams. The Clements Library
Dickinson, J. Association, Ann Arbor.
1945 Jonathan Dickinson's Journal or,
[16991 God's ProtectingProvidence, Harlan, J. R. and J. M. J. deWet
1965 Some Thoughts about Weeds.
E. W. Andrews and C. M. Andrews, Economic Botany 18(l):16-24.
eds., Yale Univ. Press, New Haven.
Hassan, F. A.
Ehrmann, W. W. 1978 Demographic Archaeology. In
1940 The Timucua Indians of Sixteenth Advances in Archaeological Method
Century Florida. Florida Hist. and Theory. Selections from Vol.
Quarterly 18:168-174. 1-4, M. B. Shiffer, ed., pp. 225281. Academic Press, New York.
Findlay, W4. P. K.
1982 Fungi: Folklore, Fiction, and Heiser, C. B.
Fact, Richmond Publishing Co., 1985 Some Botanical Considerations of
Ltd., Surrey. the Early Domesticated Plants
North of Mexico. In Prehistoric Laessle, A. M.
Food Production in North America, 1942 The Plant Communities of the
R. I. Ford, ed., pp. 57-72. Welaka Area, With Special RefAnthropological Papers, Museum of erence to Correlations Between
Anthropology, University of Mich- Soils and Vegetational SuccesNo. 75. Ann Ab.,. sion, The University Presses of
Hitchcock, A. S.
1971 Manual of the Grasses of the Landers, J. L. and A. S. Johnson
United States. Two volumes. 1976 Bobwhite and Quail Food Habits
Dover Pub. Inc., New York. in the Southeastern United States
with a Seed Key to Important
Hudson, C. F. Foods. Miscellaneous Pubs.,
1976 The Southeastern Indians. Univ. Tall Timbers Research Station,
of Tennessee Press, Knoxville. Tallahassee, Florida.
Ison, C. R. Larson, L. H.
1985 The Role of Fungi in Prehistoric 1980 Aboriginal Subsistence TechSubsistence Systems. Paper pre- nology on the Southeastern
sented at the Forty-Second Annual Coastal Plain During the Late
Meeting of the Southeastern Prehistoric Period. Univ.
Archaeological Conference, Bir- Presses of Florida, Gainesville.
Jones, C. 1975 Three Voyages. Originally
1986 Personal Communication, Archaeo- published (1586) as L'Histoire
logist, State of Florida Bureau Notable de la Florida, Situes
of Archaeological Research, Div. es Indes Occidentales, Charles
of Archives, History, and Rec- E. Bennett, translator. Univ.
ords Management, Tallahassee. Presses of Florida, Gainesville.
Jordan, D. F., E. S. Wing, & A. K. Bullen Linares, 0. F.
1963 Papers on the Jungerman and 1976 Garden Hunting in the American
Goodman Sites, Florida. Contri- Tropics. Human Ecology 4(4):
butions of the Florida State 331-339.
Museum, Social Sciences No. i0.
Univ. of Florida, Gainesville. Martin, A. C. and W. D. Barkley
1961 Seed Identification Manual.
University of California Press,
Kay, M., F. B. King, and C. K. Robinson Berkeley.
1980 Cucurbits from Phillips Sping:
New Evidence and Interprets- Milanich, J. T.
tions. American Antiquity 1974 Life in a Nineth Century Indian
45(4):806-822. Household. A Weeden Island FallWinter Site on the Upper AppalaKimbrough, J. W. chicola River, Florida. State of
1986 Personal communication, Botany Florida, Bureau of Historic
Department, Univ. of Florida. Sites and Properties Bulletin No.
Letter to author dated 10 October 4. Tallahassee.
1978 The Western Timucua: Patterns of
King, F. B. Acculturation and Change. In
1985 Early Cultivated Cucurbits in Tacachale: Essays on the Indians
Eastern North America. In Pre- of Florida and Southeastern Georhistoric Food Production in North gia During the Historic Period,
America, R. I. Ford, ed., pp. 73- J. T. Milanich and S. Proctor,
98. Anthropological Papers, eds., pp. 59-88. Florida State
Museum of Anthropology, Univ. of Museum Ripley P. Bullen MonoMichigan No. 75. Ann Arbor. graphs in Anthropology and History No. 1, Univ. Presses of
Knight of Elvas Florida, Gainesville.
1922 Narrative of the Hernando de Soto
Mission of 1539. In Narratives Milanich, J. T. and C. H. Fairbanks
of the Career of Hernando de 1980 Florida Archaeology. Academic
Soto in the Conquest of Florida, Press, New York.
as told by a Knight of Elvas and
in a Relation by Luys Hernandez Munson, P. J.
de Biedma, Factor of the Expedi- 1984 Comments on Some Addiional
tion. Together with an Account Species and Summary of Seasonof de Soto's Expedition Based on ality. In Experiments and
the Diary of Rodrigo Ranjel, his Observations on Aboriginal Wild
Private Secretary. Tran. by Plant Food Utilization in EastBuckingham Smith (1905) from ern North America, P. J. Munson,
Oviedo's Historia General y ed., pp. 459-473. Prehistoric
Natural de las Indias. American Research Series 6(2). Indiana
Explorer Series, New York. Historical Society.
Murray, P. M. and M. C. Sheehan 1943-1948 Keys to American Woods.
1984 Prehistoric Polygonum Use in In Tropical Woods various volumes.
the Midwestern United States.
In Experiments and Observations Reitz, E. J. and C. M. Scarry
on Aboriginal Wild Plant Food 1985 Reconstructing Historic SubsisUtilization in Eastern North tence with an Example from SixAmerica, P. J. Munson, ed., pp. teentb Century Spanish Florida.
282-298. Prehistoric Research Society for Historical ArchaeoSeries 6(2). Indiana Histori- logy, Special Publication Series
cal Society. No. 3.
Newsom, L. A. Ribaut, J.
1986 Plants, Human Subsistence, and 1964 The Whole and True Discoverye
Environment: A Case Study From of Terra Florida. A facsimile
Hontoon Island (8-Vo-202), Flor- reprint of the Lonldon edition of
ida. Masters thesis, Dept. of 1563. Univ. Presses of Florida,
Anthropology, Univ. of Florida, Gainesville.
Newsom, L. A. and D. S. Decker 1984 The Origins of Agriculture.
1986 Archaeological Cucurbitaceae An Evolutionary Perspective.
from Peninsular Florida. Paper Academic Press, New York.
presented at the 1986 Southeastern-Archaeological Conference, Sargent, C. S.
Nashville, Tennessee. 1965 Manual of the Trees of North
[1905) America. Dover Publications,
Panshin, A. J. an& C. deZeeuw Inc., New York.
1980 Textbook of Wood Technology,
Fourth Edition. McGraw Hill, Sigler-Lavelle, B. J. and M. Russo
New York. 1983 Production Strategies within
Florida's Coastal Lowland: EviPeterson, F. A. and P. J. Munson dence from an Early Formative
1984 Amaranth as a Food Resource in Period Household. Paper prethe Prehistoric Midwestern sented at the Third Annual MeetUnited States. In Experiments ing of the Society for Economic
and Observations on Aboriginal Anthropology, Iowa City, Iowa.
Wild Plant Food Utilization in
Eastern North America, P. J. Snedaker, S. C.
Munson, ed., pp. 317-337. Pre- 1971 Soils-Vegetation Complex of the
historic Research Series 6(2). Oklawaha Regional Ecosystem. In
Indiana Historical Society. Environmental Impact of the CrossFlorida Barge Canal with Special
Peterson, L. A. Emphasis on the Oklawaha Regional
1977 A Field Guide to Edible Wild Ecosystem. Report prepared by
Plants. Houghton Mifflin, Florida Defenders of the EnvironBoston. meant, Inc., Gainesville.
Puckett, C. Solis de Meras, G.
1982 Natural Resources, Facilities, 1964 Pedro Menendez de Aviles, Memand User Policy and Procedures orial by Gonzalo Solis de Meras.
at the Welaka Research and Edu- Translated by Jeannette Thurber
cation Center. Univ. of Florida, Connor. The University Presses
Institute of Food and Agricul- of Florida, Gainesville.
tural Sciences, Gainesville.
Speck, F. G.
Ranjel, R. 1909 Ethnology of the Yuchi Indians.
1922 Narrative of the Hernando de Soto Anthropological Publications,
Mission of 1539. In Narratives University Museum, Univ. of Pennof the Career of Hernando de sylvania Vol. 1(0). Philadelphia.
Soto in the Conquest of Florida,
as told by a Knight of Elvas and 1941 Gourds of the Southeastern
in a Relation by Luys Hernandez Indians. The New England Gourd
de Biedma, Factor of the Expedi- Society, Boston.
tion. Together with an Account
of de Soto's Expedition Based on Spencer, R. C.
the Diary of Rodrigo Ranjel, his 1985 The Economics of Passiflora Use
Private Secretary. Trans. by in the United States. Paper
Buckingham Smith (1905) from presented at the Thirty-Sixth
Oviedo's Historia General y Annual Meeting of the American
Natural de las Indias. American Institute of Biological Sciences,
Explorer Series, New York. Gainesville, Florida.
Record, S. J. and R. W. Hess Stephens, J. M.
1943 Timbers of the New World. Yale 1985 Personal Communication. Univ.
Univ. Press, New Haven. of Florida, Institute of Food
and Agricultural Sciences, Vege- Wing, E. S.
table Crops Division. 1986 Integration of Floral and
Faunal Data of Hontoon Island. Sturtevant, W. C. Paper presented at the Ninth
1962 Spanish-Indian Relations in Annual Meeting of the Society
Southeastern North America. of Ethnobiology, Albiquerque,
Ethnohistory 9(l):41-94. New Mexico.
Swanton, J. R. Wunderlin, R. P.
1946 The Indians of the Southeastern 1982 Guide to the Vascular Plants
United States. Bureau of Amer- of Central Florida. Univ.
ican Ethnology Bulletin No. 137. Presses of Florida, Tampa.
Yarnell, R. A. and M. J. Black Tutin, T. G., V. H. Heywood, N. A. Burges, 1985 Temporal Trends Indicated by
D. M. Moore, D. H. Valentine, S. M. Waters, a Survey of Archaic and Woodland
D. A. Webb, eds. Plant Remains from Southeastern
1968 Flora Europa. Volume Two. Rosa- North America. Southeastern
ceae to Umbielliferae. Cambridge Archaeology 4(2):93-106.
University Press, Cambridge.
United States Department of Agriculture,
Soil Conservation Service
1980 Soil Survey of Volusia County,
Florida. In cooperation with Lee Ann Newsom
the University of Florida Insti- Department of Anthropology
tute of Food and Agricultural
Sciences, Agricultural Experiment University of Florida
Stations, and Soil Science Dept. Gainesville, Fla. 32611
1985 Twenty-Six Ecological Communities of Florida. Gainesville.
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. Manuscript on file with the Dept. of
Anthropology, Univ. of Florida,
Watson, P. J.
1982 The Beginnings of Plant Cultivation in the Eastern Woodlands.
Paper presented at the Florida
State Museum, University of
1985 The Impact of Early Horticulture
in the Upland Drainages of the
Midwest and Midsouth. In Prehistoric Food Production in North
America, R. I. Ford, ed., pp. 99148. 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,
THE CONSERVATION OF WOODEN REMAINS FROM ARCHAEOLOGICAL WETSITES 85
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 artifacts, were stored in bags which contained a small continuously dry or continuously wet, oxygen-free amount of water obtained from the square or lagoon. conditions. The latter condition exists at many These bags were then kept in a large vat with several prehistoric sites in Florida. Although waterlogged inches of water in the bottom of it. This prevented wood looks sound when it is recovered, it is actually accidental drying if the bags leaked. highly degraded and, therefore, requires some form of
preservation to maintain its integrity. Excavated wood cannot be kept in water indefinitely.
Conservators working in England found that they could
It has been found that the inner structure of store wood in water for only a year because it absorbed waterlogged wood deteriorates and is replaced by the additional water to the point where it began to surrounding water. Degradation is caused by chemical deteriorate due to the solvent action of the water
poessand biological effects (de Jong 1979:33). (Murray 1982:14). In addition, this procedure is
prod cespiarl o eluoe space-consuming and the artifacts cannot be put on
hemi-cellulose, and lignin. The cellulose within the display in any satisfactory manner. wood is decomposed by exposure to oxygen, heat, and
ultra-violet light (Jespersen 1979:69). Both fungi and Many conservative techniques have been used on wooden bacteria attack and destroy these components. In artifacts around Florida including alum, polyvinyl addition, water dissolves and hydrolizes other acetate (PVA), polyethylene glycol (PEG), and structural components such as pectines and pentosans freeze-drying. The idea behind using alum, (de Jong 1979:33). Because of this extensive internal PVA, or PEG is to replace the water with a substance decomposition, wood becomes spongy, soft, and very that will support the cell walls and fill cell voids fragile. The presence of water, however, maintains the within the degraded wood, and not evaporate as water outer structure of the wood by bonding with the broken does. Alum, however, is not much more successful than cellulose chains. As waterlogged wood dries out, these air-drying. In this procedure, objects are soaked in a chains begin to bond to each other, thereby drawing the bath of hot, saturated potassium aluminum sulphate piece together; this is how shrinkage occurs solution (Jespersen 1979:71). The alum replaces the
(Rosenqvist 1975:12). In addition, the movement of water within the wood and also penetrates the cell water molecules out of the drying wood creates walls. Upon drying, it crystallizes, a process at first surface tension along the remaining cell walls. The thought to bulk the remaining intact cell structure, but stress arising from this tension eventually ruptures later found to damage it further (van der Heide the cell walls, resulting in cracking and checking. The 1979:18). Jespersen (1979:71) actually suggests amount of deformation can be so great as to make the retreating any artifacts that have been preserved with piece unrecognizable. Resoaking the wood will not alum. restore it to its original size and volume because of Asuycridota h nvriyo lrd the linkage of the broken cellulose chains and the Asuycridota h nvriyo lrd
additional damage caused by the liquid surface tension indicates that the use of PVA solutions is also an as the water leaves the sample. undesirable technique (Dochniak 1982b). This
experiment involved soaking pieces of wood
Before fieldwork begins at a wetsite, the approximately 3 cm in diameter and I cm thick (all cut
archaeologist, in cooperation with a trained from one large branch piece) in 10%, 25%, and 50%
conservator, should plan carefully to ensure the solutions of PEG 540 Blend and PVA over a month's survivability of the fragile and perishable material time. Several specimens remained in water as that will be recovered. Organic material must never controls. Upon drying, it was found that the pieces be allowed to dry out or else irreparable damage will soaked in PVA all lost about 85% of their original occur. At Hontoon Island, we found the easiest way to weight and 65% of their original volume. However, the remove the wood without damaging it was to use gentle loss in weight and volume of the pieces soaked in PEG water pressure from a hose to loosen and to dislodge decreased with increasing concentrations of PEG. A material which was then scooped up into a wire tray. specimen preserved in a 10% PEG solution lost 85% of Larger pieces were removed by hand and placed its wet weight and 42% of its volume, whereas the directly into plastic bags. Smaller pieces were sample from a 50% solution decreased only 38% in separated from other material at the screens and weight and 7% in volume. One of the control samples placed in bags. All perishable materials, including which was removed from the water and allowed to air
Volume 40 Number 1 THE FLORIDA ANTHROPOLOGIST March, 1987
FLORIDA ANTHROPOLOGICAL SOCIETY PUBLICATIONS NUMBER 13
dry lost 91% of its original wet weight and 71% of its hardwoods is contained in tough fibers with small volume. Figures I and 2 illustrate graphically the diameters, narrow lumina, and thick walls (Wilcox increased efficiency of a 25% and 50% solution of PEG 1973:110). The efficiency of a preservation process, 540 Blend in maintaining the sample's original weight such as soaking in PEG, depends on the permeability of over these same concentrations of PVA, and also the the wood to liquids. The complex structure in the improvement when a higher percentage of PEG is used. hardwoods makes them partially impenetrable to PEG and, therefore, when the specimen is removed from the
As illustrated, the wood in 50% PEG loses much less solution, the residual water begins to evaporate and weight than the piece in 25% PEG. The probable reason the wood begins to split, crack, and check. for the failure of the PVA is that the PVA molecules
are too large to penetrate the wood completely Tetcnqeo reedyn rvdsamto hc
(Huffman, personal communication 1980). In summary, reoesalmstqa of thee wtrywtin rvdsamtheod whi preservative solutions of PEG 540 Blend were more rmintainingt oinall haer aind size of the l effective than solutions of PVA or air drying, and mitiigteoiia hp n ieo h higher concentrations of PEG worked better than lower specimen. Freeze drying is a process whereby liquid is concentrations. removed from a substance by first freezing the
specimen and then subliming the ice rapidly under
Based on these preliminary results, a second vacuum. If the pressure and temperature are kept low experiment was performed in which one of the control enough, the liquid will go directly from a solid to a specimens from the first experiment was submitted to vapor without passing through the liquid phase. (This increasingly concentrated solutions from 10-80% PEG is what is meant by sublimation.) The vapor is then 540 Blend over a I-month period (Dochniak 1982a). carried away from the material and condensed outside Upon drying, this piece retained its original wet weight of the freeze drier. This process can be very efficient, and volume, thus showing 100% efficiency for leading to 90-100% removal of the liquid (Flosdorf preservation using a final solution of 80% PEG. The 1949:29). original samples were reweighed and remeasured on
November 1, 1985, three years after the termination of Several schedules for freeze drying have been the PEG/P VA experiment. Table I clearly shows that developed by conservators at other institutions (for the specimens preserved in PEG have remained stable, example, R.J. Barbour at the University of Washington, thus providing confidence in PEG as a long-term and David Grattan and Cliff McCawley at the Canadian preservation method. Conservation Institute). Following their method, we
begin by soaking the wood in a 20% solution of PEG
Several conservators have found that superior results 1000 for three months (Ambrose 1972; Barbour 1983:1, can be obtained by using PEG 540 Blend as a bulking 1984:1; Smith and Barbour 1983:23). This concentration agent instead of chemicals such as acetone rosin, of PEG 1000 is employed because it can be frozen more tertiary butanol, or other molecular weights of PEG effectively than can other concentrations or other (Barbour 1983:1; Grattan 1982:132-34). We have found molecular weights of PEG. A concentration of 20% puts the following procedure most successful in treating enough PEG into the degraded wood to improve its woods from Hontoon Island: immerse the waterlogged mechanical properties without making it waxy. There material in increasingly concentrated solutions of PEG are several reasons for soaking the wood in PEG prior 540 Blend from 10-80%, raising the concentration by to freeze drying. Water expands as it freezes and this 10% every month (duration may be increased if damages the wood. The contraction of PEG during specimens are large). This stepped procedure is freezing counteracts the expansion of the water necessary since high initial concentrations draw the (Rosenqvist 1975:13). Frozen PEG is soft, not rigid like water out faster than it can be replaced by the PEG, ice crystals (Grattan and McCawley 1978:158). In causing excessive shrinkage and cracking. We do not addition, upon sublimation of the water, the PEG is left find it necessary to heat the solution or use biocides; behind as a supportive medium (Hartley and Grosso instead we simply stir it occasionally and skim off 1979 :7).
orgaism if heyare resnt.Preliminary results from this part of our experiment We are presently conducting a long-range study show that some of the wood has a tendency to break monitoring the stability of various types and species of apart while soaking in the PEG 1000 and, when removed wood conserved with PEG 540 Blend in 1980. from the solution, must be frozen immediately or else
Preliminary results show that the majority of severe cracking results. This suggests that further softwood species, such as pine, cypress, and cedar are experimentation with solution concentrations and preserved excellently using the method described soaking times is needed. After soaking, the material is above, while almost all of the hardwoods such as oak, placed in plastic bags or foil and frozen at -201C. This hickory, and persimmon are not. The cellular structure process, as opposed to rapid freezing in dry ice, helps of hardwood is considerably more complex than that of to prevent cracking (Grattan 1982:127). The frozen softwoods. Although the basic functions of tissues in wood can then be stored for long periods without hard- and softwoods are the same, they are carried out decay, or freeze dried immediately (Murray 1982:18). by a large number of specialized cell types in the After freezing completely, the wood is placed in the hardwoods (Wilcox 1973:110), but by only a few kinds of vacuum chamber of the freeze drier while the cells in the softwoods (Lee Newsom 1984: personal temperature is raised incrementally until sublimation communication). Most of the cell wall material in is complete (Grattan 1982:127; Barbour 1984:1). The
Solution Sample. O yIng SoIu tion S namplea Drylng
i IfI I I I
0.0- 2 4 6 8 10 12 14 16 18 20 22 24 26 28 30 1438
0 2 4 6 8 1012 14 16 18 20 22 24 26 28 30 1438
TIME (Days) TIME (Days)
Treatment With 25% PEG Treatment With 25% PVA Treatment With 50% PEG Treatment With 50% PVA
Sample A-2 Sample B-2 Sample A-3 Sample B-3
Figure 1. Changes in Sample Mass During and After Treatment in Figure 2. Changes in Sample Mass During and After Treatment in a 25% Solution a 50% Solution
88PEG Group Weight Diameter Thickness Volume
(g) (cm) (cm) (cm3)
19811985 1981 1985 1981 1985 1981 1985
A-I 10% 0.7 0.6 1.9-2.2 1.9-2.0 0.7 0.75 2.3 2.2
A-2 25% 1.5 1.4 1.8-2.5 1.8-2.5 0.8 0.8 2.9 2.9
A-3 50% 2.8 2.8 2.3-2.7 2.3-2.7 0.8 0.9 3.9 4.4
C-I 80% 4.5 4.4 2.4-2.8 2.3-2.8 0.8 0.9 4.2 4.5
C-3 control 0.4 0.3 1.2-1.7 1.3-1.5 0.7 0.7 1.2 1.1
B-I 10% 0.6 0.5 1.7-1.8 1.5-1.7 0.6 0.8 1.4 1.0
B-2 25% 0.7 0.55 1.6-1.9 1.3-1.8 0.7 0.8 1.7 1.5
B-3 50% 0.8 0.6 1.4-1.6 1.6-1.4 0.8 0.8 1.4 1.4
C-2 Control 4.5 4.3 2.4-2.8 2.4-2.8 0.8 0.8 4.2 4.2
0-4 Control 0.4 0.3 1.2-1.7 1.1-1.7 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 = rrh = (Hl)(d/2) (thickness).
Table 1. Remeasurement of Samples Used in PEG/P VA Experiment
temperature is then slowly brought up to 2000 and held The prehistoric inhabitants of Florida could not build
there until drying is complete (Barbour 1984:1). monuments and statues of stone because suitable raw
Experimentation with freeze drying at the University of material was not available. Their achievements in wood
Florida is in progress. should be preserved for future generations to enjoy in
the same way as we enjoy the accomplishments of the The primary goal of any conservative technique with Egyptians, Greeks, and Romans (Purdy 1982:2).
waterlogged wood is to maintain the original dimensions of the wood by preventing deformation due to shrinking, cracking, and checking. Any method which References Cite
accomplishes this can be considered effective and viable. However, the cost and safety factors of the Ambrose, W.R.
various techniques must also be taken into account 1972 The Treatment of Swamp Degraded Wood by Freeze-drying.
when choosing a specific procedure. The final Report to the International Council of Museums Committee for
appearance and texture of the object after completion Conservation, Madrid.
of the treatment must also be considered. The most Barbour, R.J.
1983 Personal communication to Barbara A. Purdy, January 13. important challenge is to determine whether or not an (Letter on file Purdy'soffice, Department of Anthropology,
object will be preserved permanently with the method University of Florida.)
chosen. Only time will answer that question. 1984 Personal communication to Barbara A. Purdy, February 9.
(Letter on file Purdy's office, Department of Anthropology, Organic artifacts recovered from archaeological wet University'of Florida.)
sites constitute a unique and non-renewable heritage. de Jong, J.
1979 The Deterioration of Waterlogged Wood and its Protection in the Archaeologists and conservators, therefore, have a Soil. Conservation of Waterlogged Wood: international symposium
responsibility to preserve them for future study. on the Conservation of Large Objects of Waterlogged IWood.
Archaeologists must be prepared to implement the pp. 31-40. Government Printing and Publishing office, The Hague.
proper excavation and conservative techniques Dochniak, Craig C.
necssay wen rgaic ateialis oin tobe1982a The Maxium Efficiency of PEG in waterlogged Wood
necesar whn orani maeril isgoig t bePreservation. Report presented to Dr.Barbara A. Purdy, March 30, recovered. This requires establishing cooperative 1982. (On file Purdy's office, Department of Anthropology,
relationships with conservators before fieldwork University of Florida.)
begins and necessitates requesting enough money in a 1982b Polyethylene Glycot Compared to Polyvinyl Acetate in
pooabugtto cover the high costs of conservation. Waterlogged Wood Preservation. Report presented to Dr. Barbara A.
propoal bdgetPurdy, January 30,1982. (On file Purdy's office, Department of In addition, serious Anthropology, University of Florida.)
consideration should be given to finding adequate Flosdorf, Earl W.
storage facilities for the artifacts after they are 1949 Freeze-drying: Drying by Sublimation. Reinhold Publishing
preserved. Corp., New York.
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
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,
1980 Personal communication. School of Forestry, University of Florida, Gainesville.
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.
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 PersonaI 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
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.
Wiloex, Wayne W.
973 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
BOOK REVIEWS, CURRENT RESEARCH, COMMENTS AND EVENTS
FLORIDA ARCHAEOLOGY, NUMBER 2 "surrounding circumstances" as he proceeds
Florida Bureau of Archaeological Research, with his translations. He also lets you
Tallahasssee, 1986. vi + 225 pages, know when he encounters difficulties and
figures, tables, bibliography. $10 (paper) why. His most severe comment, however,
This document certainly deserves a better
title than Number 2. At least, Dr. John H. The elliptical features of this sentence
Hann, whose work is here presented, should make its full thrust difficult for the
be celebrated somewhere on the cover, translator to resolve. (Note 95)
FLORIDA ARCHAEOLOGY, NUMBER 2 presents in Although not as harsh as "elliptical," he
four sections some of Dr. Hann's work: sometimes slips in a "muddy" when the ink
has splashed or the copyist has changed
Translation of the Ecija Voyages of 1605 gears. Despite these difficulties, he has,
and 1609 and the Gonzalez Derrotero of 1609 within the limitations he describes, made
them very readable.
Translation of Governor Rebolledo's 1657
Visitation of Three Florida Provinces and In this first section, Dr. Hann also has
Related Documents supplemented the Ecija material with translations of several relevant documents that
Church Furnishings, Sacred Vessels and bear on the subject of the navigation of
Vestments Held by the Missions of Florida: the Atlantic coast from the Bahama Channel
Translation of Two Inventories to the bar of Jacan (mouth of Chesapeake
Bay), and the activities of the French and
Translation of Alonso de Leturiondo's English in this area during the period.
Memorial to the King of Spain
One need not be an avid student to appreEach section is of significant value to ciate these documents. A novelist could
anyone who may be interested in any portion get several good plots out of the Ecija of the Spanish contact period. Each section voyages: the Frenchman who was captured is considerably enhanced by the introduc- and taken as a slave to the interior of
tion, explanation and notes of Dr. Hann. the Carolinas just before Ecija arrived in
The voyages of Ecija and Gonzalez are very the area on his 1605 voyage and was later
much easier to follow through the excellent ransomed back from his Indian captors by
maps of Charles Poe. Likewise the study of Ecija during the 1609 trip. Or the conthe visitations of Governor Rebolledo are frontation between the Spanish and an
aided by Mr. Poe's illustration of the English ship at the mouth of the Chesageography involved, although the mission peake Bay as they jockeyed back and forth
locations will certainly be disputed. for position while Ecija and his committee
of officers tried to determine if a shootThe Ecija Voyages ing match would be worth the effort to
accomplish their mission. Discretion led
Dr. Hann provides introductory comment to Ecija and his men to return without having
give perspective before undertaking the forced the issue and without having actualsomewhat stilted language of Ecija's trip ly seen the new English settlement, reasonlogs. Ship's logs haven't changed much ing that all they really had to do was
over the years and remain terse, often in locate it, they really didn't have to see
a colloquial shorthand and require a it. This latter is a prime example of how
thorough knowledge of surrounding circum- Ecija survived his more than thirty years
stances to arrive at a correect interpre- of sailing up and down the southeast
tation. Dr. Hann carefully notes the Atlantic coast for the Spanish.
Volume 40 Number I THE FLORIDA ANTHROPOLOGIST March, 1987
FLORIDA ANTHROPOLOGICAL SOCIETY PUBLICATIONS NUMBER 13
Each reader will find something different Diego Pe? a in 1716 (Boyd translation) puts
to whet the appetite. I was struck by the Azile west of the Aucilla. So does B.
Spanish concern that the French might know Calvin Jones in 1987 (Jones and Shapiro:
something about the interior of what is now "Nine Mission Sites in Apalachee," Society the southeastern part of the United States of Historical Archaeology Annual Meeting,
that the Spanish didn't. Something that Savanna). But was Azile town in Apalachee?
perhaps Pardo forgot to mention. As Dr. Hann notes, the literature generally
treats Azile as Timucuan and therefore part
Finally, it should be noted that Dr. Hann of Ustega. Somewhere I read (Swanton, I
continues the struggle with the Spanish think) about the Creek use of "peacetowns"
word monte, which he notes "can mean woods along the Chattahoochee. It is just possior thicket or hill." Other translators ble that Azile town was in normally Apalahave had the same experience. Where we run chee territory though populated with Usteinto this in the interior we tend to sub- gans, that it was a "peacetown" established
stitute the word "hammock." That may not shortly after Prieto's 1608 visit, when
necessarily be correct in all instances. I Ivitachuco was the first town west of the
should think that Dr. Hann's notes would Aucilla. It is possible that Azile was esserve as a textbook on the problems experi- tablished with Ustegans (as sort of hosenced in attempting translations of these tages) just within the Apalachee territory
old documents. to cement the peace just then established
between the formerly warring Ustega and
Rebolledo's 1657 Visitation Apalachee. This would make the answer to
the third question: "Yes," and possibly
Turning next to the section on the Rebolledo explain why, by the time of Rebolledo, the visitation, practically all students have nephew of an Apalachee chief could be chief
covered some phase of this somewhere in some of Asile without it necessarily being an article, but not like it is here presented. "Apalachee town." As Dr. Hann, with his own pick and shovel
well in hand, noted in the April 1986 I have struggled with the place names in
Florida Historical Quarterly: the last paragraph because Dr. Hann emphasizes that the correct spelling should be
Although that (Rebolledo) visitation Azile, not Asile or Assile or Assily, etc.;
record has been mined extensively during Ustega not Yustega and Abalache not Apalathe last two decades for two disserta- chee. But none of us could break the habit
tions and for a number of journal arti- completely (See Note mentioned above).
cles, a valuable aspect of that document's contents has not yet received There is such a wealth of information conmuch attention, tained in the Rebolledo visitation records
that the "mining" is sure to continue for
Indeed. This presentation, supplemented as some time to come. No doubt those who have it is by extensive comment and authority, spent all that de Soto money looking for
should be made mandatory for any student of the mission road knowing that someday the Spanish-American history. auditors will discover that de Soto did not
travel up Route 41 and hang a left at U.S.
With my own little "pick in hand," I note 90, must have seized on Rebolledo's remark
that Dr. Hann and his map-maker are contin- that uing the controversy concerning the location of Azile (San Miguel de Azile mission, ... as the places of this said province
See Map page 82 and Note 3, page 138). Was of Timucua are far apart from one another
it east or west of the Aucilla (Azile or along crosswise paths and not along the
Asile) River? In Ustega (Yustega) or royal road, ... (p. 104).
Apalachee? And, was the Aucilla River
really the eastern boundary of Apalachee? Thus, they will know that finding the mis-
Sian road, "the royal road," will not help of the Rebolledo visita concerning the
much. The early missions (which no doubt priest's taking the Indian products and
were located at the oldest villages, the reselling them at considerable profit is
towns most likely subject to the earliest true, or God was truly working miracles
European visits) were on "crosswise paths" at the missions. Dr. Hann notes that
and "far apart" and "not along the royal just the pieces of art averaged twenty
road." So it appears that locating the per mission. If you take into account
camino, real where some of these missions the plunder taken by the British and
were relocated after 1657 will be an exer- their Indian allies from the various
cise in futility as far as the earliest missions, the amount stated to be buried
European (make that de Soto) contact is and/or burned by the Spanish themselves
concerned. Disease, rebellion, death and as they abandoned the missions and add
official edict dissipated the clues, it to the amount brought back to St.
Augustine and actually inventoried, you
The quoted remarks of Rebolledo both star- must conclude that the priests did much
tled and refreshed those in Utina who are better than heretofore thought and the
joined in the search; closing, perhaps, raids on the missions could have been as
one line of endeavor but opening up other much for the plunder as for being an expossibilities. To date Utina has suffered tension of English foreign policy.
from the provencialiasm of those employed
in this field. For the most part (happily, Bushnell and others in their studies also
there are exceptions) these "professionals" have been led to the conclusion that there
have become oriented to their habitant was a lot more wealth in La Florida than
province, either Potano or Apalachee. It can be accounted for officially. Take a
is refreshing and most exceedingly helpful look at this third section, let your mind
to have Dr. Hann turn his attention to the wander around a bit and see.
middle provinces. A few more diggers would
be of considerable help also. Hopefully Leturiondo's Memorial to the King
there will soon be more than just press
releases to show for these efforts. The last section in Number 2 is Dr. Hann' s
translation of Leturiondo's memo to the
Church Furnishings King. This one turned out to be not my
cup of tea, but three out of four is not
Hann observes, as have others, that the bad. For those not familiar with this
1657 visita of Rebolledo was probably set document it is a very long letter from a
up to secure testimony favorable to the Florida born, St. Augustine priest to his
governor, his troops and his prior policies European king addressing what the priest
and unfavorable to the priests with respect considers to be the major problems with
to the treatment of the Indians. History respect to Florida and offering solutions,
shows that this was an exercise in futility or at least suggesting changes, that might
on the part of Rebolledo, his fate having improve the situation as he sees it. Actualready been sealed by other powers. Here ally, the entire section including the
Dr. Hann, with his third contribution on introduction, notes and references by Dr.
Church Furnishings, adds balance to the Hann runs only 59 pages, but it seems to
equation. He points out that which has go on forever. Leturiondo's prose moved
been glossed over in most studies. That Dr. Hann to his most violet comment in the
is, it appears the priests did very well entire 225 pages: "turgid."' It certainly
economically. Far better than they were is. But so is quantum physics and this
in the practice of telling the king or piece contains a lot of information.
maybe even their church superiors for that
matter. Most intriguing is the part about how the
local farmers and ranchers with the assisEither a substantial part of the testimony tance of the local government officials
paid a very small percentage of the taxes or tithes properly due. Admittedly a matter of interpretation but of vital interest to those who depend on the tithes. Leturiondo mentions a number of ways that the situado (subsidy) is misdirected, misused 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 outside, 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 matters, 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 laurelsit. There must be more.
W.S. Eubanks, Jr.
P.O. Box 880
Live Oak, Florida 32060
EDITOR' S NOTE:
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 323990250. The cost is $10.00 plus 5% (50 ) 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 found nere on the use of palynology
94 Rapp, Jr. and John A,Gifford. 435 pages, (paleontology of spores ad pollen) and
hardbound. Published by Yale University archaeomagnetism. Archaeomagnetism,
Press, New Haven and London. ISBN 0-300- according to one article, may have a
03142-4. Retail price: $35. potential accuracy of 25 years. In
Geoarchaeology quarterly journal published addition, there are articles on tephroby John Wiley & Sons, 605 Third Avenue, chronology (the dating of volcanic ejecta)
New York, NY 10158. Subscription price: and a technique for the radiocarbon dating
$96/year. of lime mortar.
Archaeology and Geology have long been Determination of the source of material
held as distinct subjects in the public found at archaeological sites has become an
and scientific mind. The geologist and important part of recent archaeologic work.
archaeologist are separated by different Such information can be used to interpret
backgrounds and by interests that do not trading routes or cultural influence. A
appear to overlap geology in general variety of geochemical and geological
does not apply directly to man. These techniques that have archaeologic applidifferences have long obscured the many nations are discussed in this book,
similarities between these fields. Both including aerosol (soil) analysis, the
involve the retrieval and examination of mineralogic and petrologic analysis of
objects obtained within the earth, in ceramic and building materials. There is
order to interpret the past. In recent also an interesting article in which
years, many techniques and methodologies isotopic analysis is used to determine the
commonly used in geology have been refined provenance, and thus the authenticity, of
to the point that they can be used in very an ancient marble bust.
There are also articles on geomorphology,
"Archaeological geology" is defined as sedimentology, soil analysis, and
"geology pursued with an archaeological their use in interpreting many aspects of
bias or application." "Geoarchaeology" is ancient environment. Much more could have
archaeology practiced with the aid of been said about these applications, algeological techniques and methodology, though that would have required a more
The difference between these may be simply lengthy and expensive text. But of
a matter of emphasis, or the background of enormous value to the researcher is a 55the principle investigator. By whatever page selective bibliography, which is in
name it might be called, the field has addition to the references cited in the
very recently come into its own. At the articles themselves. Thus, Archeological
most recent national meeting of the Geo- Geology makes very large amount of the past
logical Society of America (November, work in this interdisciplinary field avail1986, San Antonio), there were about able to the reader. The book is a very
twenty talks (in two sessions) and a number useful reference work, and is highly reconof poster presentations relating to this mended for research archeologists.
An additional source of information on
The book Archaeological Geology does a good this interdisciplinary field is the new
job in showing the variety of geological journal Geoarchaeology. This journal
subjects that have applications to archaeo- seeks to publish articles from "all
logy. In general, geology can be used in aspects of geology, paleontology, climatodetermining age, provenance, or the envir- logy, oceanography, and geochronology
onment of archaeological sites or arti- which either include archaeological data
facts. Technologies commonly used in or can be applied to archaeological
geology, such as geophysical surveying studies." The issue reviewed here (vol.
methods, can also be used in finding or 1, no. 3) has articles on geophysical
determining the extent of an archaeological methods, sedimentation rates, sea level
site. In each of these cases, there are a change and their archaeological
number of geological specialties that can implications. There are also very useful
provide information to help solve book reviews and a calendar of upcoming
archaeological problems. meetings. The issue has 5 articles and
about 100 pages, although a number of the
Perhaps the most important similarity pages are blank. The subscription cost,
between geology and archaeology is the unfortunately, is more than many people
concern with time. In each science the can justify, but hopefully it will soon be
determination of relative and absolute available at many libraries.
time is essential in understanding the
history of the past. Geologists use a Submitted by:
variety of methods to determine age, in- Kevin McCartney
cluding paleontology, isotopic analysis Department of Geology
and geomagnetism. Of these, only isot pic Florida State University
analysis has seen extensive use, as C Tallahassee, Florida 32306-3026
dating, in archaeology. Articles can be
Volume 40 Number 1 THE FLORIDA ANTHROPOLOGIST March, 1987
FLORIDA ANTHROPOLOGICAL SOCIETY PUBLICATIONS NUMBER 13
THE NORTH AMERICAN INDIAN is a 21-volume Volume 14. A Blackfoot Sourcebook: Papers
set reproducing "in facsimile over 370 by Clark Wissler, edited with an introducof the most important articles on a num- tion by David Hurst Thomas.
ber of topics in Indian studies." Each
Volume Editor provides an introduction Contents:
to the articles in the volume and a bibli- 1. Material Culture of the Blackfoot Indians. AMNHAP (1910). ography for further reading. Many of the 2. The Social Life of the Blackfootlndians. AMNtAP (1911).
3. Societies and Dance Associations of the Blackfoot Indians. AMNHAP
reproduced articles are extremely rare. (1913).
4. The Sun Dance of the Blackfoot Indians. AMNIJAP (1918).
The volumes in this series are printed on 510pages LC83-47645 lSBN0-824058887 $65
acid-free, 250-year-life paper. David
Hurst Thomas is the General Editor for This volume reprints four of the six volthis series, which is published by Garland umes which Clark Wissler prepared for puPublishing, Inc., 136 Madison Avenue, New blication by the American Museum of Natural
York, NY 10016. (202) 686-7492. History between 1909 and 1918. The two
volumes omitted are the lengthy work on
This is an important series which will be Blackfoot mythology (Wissler and Duval
an asset to any reference library, and will 1909) and the extensive descriptive work be of interest to many of our readers. Re- dealing with Blackfoot ceremonial bundles
views of the individual volumes are being (Wissler 1912).
presented in clusters over a series of four issues of this journal. The first twelve The Volume Editor, David Hurst Thomas,
volumes were reviewed in the last two is- reports that:
sues. The review of this series continues in this issue, and will be completed in the It is important to note the degree to
next issue. which Wissler was assisted in his research by David C. Duvall (18779-1911).
Volume 13. A Plains Archaeology Sourcebook: Duvall's mother was a full-blood Piegan,
Selected Papers of the Nebraska State His- and his father was a French-Canadian
torical Society, edited with an introduction fur trader ...
by Waldo R. Wedel.
Wissler engaged Duvall as an interpreter
Contents: in 1903. ... He took an incresingly innovative role in Wissler's Blackfoot re1. Hill, AT., and Paul Cooper. The Archaeological Campaign of 1937. NHM
(1938). search, ultimately contributing several
2. Cooper, Paul L. The Archaeological Exploration of 1938. NHM (1940). hundred pages of manuscript to the effort 3. Hill, AT., and M.F. Kivett. Woodland-Like Manifestations in Nebraska.
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. NtIM (1936). Duvall's untimely death in 1911 abruptly
464 pages LC 83-47639 ISBN 0-8240-5887-9 S45 terminated Wissler' s research on Blackfoot medicine rituals ...
The Nebraska State Historical Society was
incorporated in 1878 as a private organi- As a result of his informant, Wissler's
zation, and later designated in 1883 as a work is dominated by Piegan Material, with
state agency. Dr. Wedel provides a brief less treatment on other Blackfoot groups.
overview of this society as a background Nevertheless, I found these works to be
setting for the five articles which he very interesting and informative, includselected for reprinting in this volume. ing the graphics which provide a greater
The five articles are important, relative- depth of understanding than one might obly little known, works originally publish- tain from the text alone. I recommend ed by the Nebraska State Historical Society. the acquisition of this volume. I found the text and illustrations in these reports to be interesting and informative. They provide essential background informa- Volume 15. The Dunbar-Allis Letters on the tion to researchers in Nebraska. I recom- Pawnee, edited with an introduction by Waldo
mend the acquisition of this volume. R. Wedel.
Volume 40 Number 1 THE FLORIDA ANTHROPOLOGIST March, 1987
FLORIDA ANTHROPOLOGICAL SOCIETY PUBLICATIONS NUMBER 13