The Florida anthropologist

THE FLORIDA ANTHROPOLOGIST

Published by the FLORIDA ANTHROPOLOGICAL SOCIETY, INC.

VOLUME 63, NUMBER 3-4 September December 2010

r%

E
78
.F6
F58

THE FLORIDA ANTHROPOLOGIST is published by the Florida Anthropological Society, Inc., P.O. Box 357605, Gainesville, FL 32635.
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 may be initiated at any time during the year. and covers the ensuing twelve month period. Dues shall
be payable on the anniversary of the initial dues payment. Members shall receive copies of all publications distributed by the Society
during the 12 months of their membership xear. Annual dues are as follows: student S15. individual S30. family S35. institutional $30,
sustaining S100 or more. patron $1000 or more. and benefactor S2500. Foreign subscriptions are an additional S25 U.S. to cover added
postage and handling costs for indi idual. family, or institutional membership categories. Copies of the journal will only be sent to
members with current paid dues. Please contact the Editors for information on recent back issues.

Requests for information on the Society. membership application forms. and notifications of changes of address should be sent to the
Membership Secretary. Donations should be sent to the Treasurer or may be routed through the Editors to facilitate acknowledgment in
subsequent issues of the journal (unless anonymity is requested). Submissions of manuscripts should be sent to the Editors. Publications
for review should be submitted to the Book Review Editor. Authors please follow The Florida Anthropologist style guide (on-line at
wwwv.fasAweb.org) in preparing manuscripts for submission to the journal and contact the Editors with specific questions. Submit four (4)
copies for use in peer review. Only one set of original graphics need be submitted. The journal is formatted using Adobe In Design. All
manuscripts must be submitted in final form on CD in Microsoft format. Address changes should be made AT LEAST 30 DAYS prior to
the mailing of the next issue. The post office will not forward bulk mail nor retain such mail when "temporary hold" orders exist. Such
mail is returned to the Society postage due. The journal is published quarterly in March. June. September, and December of each year.

OFFICERS OF THE SOCIETY

President: Robert J. Austin. P.O. Box 2818. Rixerview. FL 33568-2818 (bob searchinc.com)
First Vice President: Steven Martin. 4642 St. Augustine Rd.. Monticello. FL 32344 smartinga tin-top.com)
Second Vice President: Theresa Schober. 15770 Lake Candle-wood Drive. Fort Myers. 33908 (theresa@ fortmyersbeachfl.gov)
Corresponding Secretary: Debra Wells. SEARCH. Inc.. 315 NW 138 Terrace. Jonesville. Florida 32669 (debra@searchinc.com)
Membership Secretary: Pat Balanzategui. P O Box 1434. Fort Walton Beach. FL 32549-1434 (wnpbal@ cox.net)
Treasurer and Registered Agent: Joanne Talley. P.O.Box 788. Hobe Sound. FL 33475 (joa xwhiticar.com)
Directors at Large: Chris Hardy. 1668 Nantucket Ct.. Palm Harbor 34683 (kasotagirl' y ahoo.com): Debra Wells. (debra@searchinc.
com): Jon Endonino. SEARCH. Inc.. 315 NWV 138 Terrace. Jonesville. Florida 32669. (jon@ searchinc.com)
Immediate Past President: Patty Flynn. P. O. Box 11052 Ft. Lauderdale Fl. 33339 (pflynn@a pbmnh.org)
.Vewsletter Editor: David Burns. 15128 Springviexw St.. Tampa. FL 33624 (daveburns prodigy.net)

JOURNAL EDITORIAL STAFF

Co-Editors: Deborah R. Mullins. P.O. Box 12563. Pensacola. FL 32591-2563 (dmullins.fl.anthropologista gmail.com)
Andrea P. White. Department of Anthropology. University of New Orleans. 2000 Lakeshore Drive. New Orleans. LA 70148
(awhite.fl.anthropologist@ gmail.com)
Book Review Editor: Jeffrey T. Moates. FPAN West Central Regional Center. 4202 E. Fow ler Ave NEC 116. Tampa FL 33620
(jmoates@ cas.usf.edu)
Editorial Assistant: George M. Luer, 3222 Old Oak Drive. Sarasota. FL 34239-5019 (gluer@grove.ufl.edu)
Technical Assistant: Beth Chambless. SEARCH. Inc.. 428 E. Government St.. Pensacola. FL 32502. bethd searchinc.com)
Printer: Durra-Print. 717 South Woodward Ave.. Tallahassee. FL 32304
Bulk Mail: Modern Mailers. 877 \W Orange Ave.. Tallahassee. FL 32310

EDITORIAL REVIEW BOARD

Albert C. Goodyear. Institute of Archaeology and Anthropology. University of South Carolina. Columbia. SC 29208
(goodyear@sc.edu)
Jerald T. Milanich. Florida Museum of Natural History. University of Florida. Gainesville. FL 32611 (jtm@ flmnh.ufl.edu)
Jeffrey M. Mitchem. Arkansas Archeological Survey. P.O. Box 241. Parkin. AR 72373 (jeffmitchem@juno.com)
Nancy Marie White. Department of Anthropology. University of South Florida. Tampa. FL 33620-8100
(nwhite@chuma .cas.usf.edu)
Robert J. Austin. P.O. Box 2818. River iexw. FL 33568-2818 (bob@ searchinc.com)

NOTE: In addition to the above Editorial Review Board members, the review comments of others knowledgeable in a manuscript's
subject matter are solicited as part of our peer review process.

VISIT FAS ON THE WEB: www.fasweb.org

U OFF LIBRARY

THE FLORIDA

ANTHROPOLOGIST

Volume 63, Number 3-4
September December 2010

TABLE OF CONTENTS

From the Editors

ARTICLES

The Location of the Paramount Town of the Ais Indians
and the General Location of the Indians of Santa Lucia.
Alan Brech and J. F. Lanham

Gotier Hammock Mound and Midden
on St. Joseph Bay, Northwest Florida.
Nancy Marie White

2010 FLORIDA FIELD SCHOOL SUMMARIES

ABOUT THE AUTHORS

OP1-

0V

IYCE i9A1

Cover: Photographs of Basin Bayou Incised jar from the Gotier Hammock site (cat nos. 08-39, -44, -49, -51,-99,
-107). See the White article beginning on page 149 for more information.

Published by the
FLORIDA ANTHROPOLOGICAL SOCIETY, INC.
ISSN 0015-3893

VI
0
t

r

UNIVERSITY OF FLORIDA LIBRARY

FROM THE EDITORS

In our first article, Alan Brech and J.F. Lanham reconsider
previously held ideas about the location of the paramount town
of the Ais along the east coast of Florida in the Indian River
Lagoon region during the seventeenth century and provide
a geographical reconstruction of the historic landscape.
Drawing mainly from primary accounts of Alvaro Mexia in
1605 and Jonathan Dickinson in 1699, the authors rigorously
examine available historical documents including a review
of the survey and distance measuring methods utilized by
various chroniclers. Using historic measuring systems (or the
range of historic measuring accuracies), Brech and Lanham
extrapolate a geographic range for the town of the Ais, examine
known archaeological sites within the range, and suggest
the paramount town of the Ais corresponds to the Kroegel
Homestead site, alternatively known as Baker's Bluff.
Additionally, by studying available geomorphological
information for the area, the authors highlight the dynamic
barrier island system in this part of Florida to explain
the constantly shifting inlets that potentially account for
discrepancies in the location of geographic and cultural
landmarks in various historical accounts spanning the
sixteenth, seventeenth, and eighteenth centuries. Finally, the
implications of the new proposed location of the town of
Ais also shift the interpretation of potential site locations for
the Indians of the Santa Lucia, several short-lived Spanish
garrisons, and other geographical features in the Indian River
Lagoon region. This article provides an excellent example of
how taking a critical and systematic look at past historic and
cartographic sources can often result in a reinterpretation of
our assumptions about site location and in the development of
fresh research agendas.
Our second article from Nancy White at the University
of South Florida (USF) focuses on the Gotier Hammock site,
a burial mound and midden site on St. Joseph Bay in Gulf
County, Florida. C.B. Moore's 1902 visit to the mound (along
with plenty of other looting activities through the years)
nearly destroyed the entirety of the site. However, USF field
investigations produced radiocarbon-dated deposits for the
mound at A.D. 650 and the associated midden yielded Middle
Woodland and later ceramics. These data confirm the link
between Swift Creek and early Weeden Island ceramics for
Middle Woodland sites across this part of Florida.
Data from Gotier Hammock contribute directly to White's
ongoing research on the natural and social landscapes of
past Floridians. In this informative and clearly written (and
humorous!) article, White adds to the deconstruction of once
typical "sacred vs. secular" explanations of ceramic wares
and other artifact types from non-domestic contexts and

demonstrates repeatedly that the idea of a separate and sacred
space may be only a reflection of modern esthetics and not an
accurate rendering of past realities. White notes that the relative
plainness of many of the ceramics from Gotier Hammock
mound might once have meant they were disregarded in
explanations of the site. According to White, what is sacred
"may not be beautiful"; it is the archaeological context of both
plain and fancier items, as well as the consideration of the
use (history) of that item that will allow for a more developed
story of the past inhabitants of Florida's panhandle.
Finally, we have an update on recent college and
university fieldwork spanning all corners (on land and sea)
of our great state. The 2010 Field School Summaries keep
The Florida Anthropologist readers informed of current
research projects and research questions about Florida's past
in addition to highlighting the variety of Florida's unique
cultural heritage. Thank you to our busy contributors, who
included undergraduate and graduate students and Principal
Investigators, who took the time to share snippets of their
fascinating research with the rest of us. Special kudos go
to the hard-working students and volunteers who are the
backbone of Florida archaeology. Field school provides one
of the cornerstones of training for the next generation of
Florida archaeologists. Without the students' toil and grind
out in the heat (and occasionally the cold), new avenues of
archaeological discovery could not be reported on in the pages
of this journal or elsewhere. Keep up the good work everyone!

In this issue we hope there is a little something for everyone.
Enjoy!

Deborah Mullins and Andrea White

THE FLORIDA ANTHROPOLOGIST FUND

An Endowment to

Support production of
The Florida Anthropologist,
the scholarly journal published quarterly by
the Florida Anthropological Society since 1948.

Donations are now being accepted from
individuals, corporations, and foundations.

Inquiries and gifts can be directed to:

The Editors
The Florida Anthropologist
PO Box 12563
Pensacola, FL 32591-2563

The Florida Anthropological Society is a non-profit organization under section
501(c)(3) of the Internal Revenue Code.
Contributions are tax-deductible as provided by section 170 of the code.

THE LOCATION OF THE PARAMOUNT TOWN OF THE Ais INDIANS
AND THE GENERAL LOCATION OF THE INDIANS OF SANTA LUCIA

ALAN BRECH' AND J. F. LANHAM2

' 1142 Wild Rose Drive NE, Palm Bay, FL 32905
E-mail: aebrech@aol.com

S1482 Anglers Drive NE, Palm Bay, FL 32905
E-mail: j_f lanham@vahoo. con

Ever since the first Anglo-American surveys of Florida
in the 1760s by Gerard de Brahm and Bernard Romans, the
barrier island inlet known as the River (or Bar) of the Ais
Indians was thought to have been located at the Old Indian
River Inlet, a natural passage to the ocean which finally
closed in the early twentieth century, located 4 km north of
the artificial Fort Pierce Inlet which replaced it (De Brahm
1772, in De Vorsey 1971:206; Romans 1962:281-282 [1775])
(see Figure 1). For over one hundred years, the paramount
town of the Ais Indians was thought to have been located
within two nautical leagues (about 10 12 km) of the Old
Indian River Inlet (Lowery 1905:434). Seventy years ago,
Charles Higgs (1942:35) identified the winter satellite to Ais
as the Beachland site (81R16), 13 km north of the former Old
Indian River Inlet on the barrier island in the modern town
of Riomar, corresponding to Alvaro Mexia'sl605 (in Rouse
1951) description that Ais towns existed as summer-winter
pairs straddling the Indian River Lagoon, the summer towns
along the mainland, and the winter towns along the barrier
island. The summer location of the paramount town of the
Ais on the mainland was presumed to have been destroyed
by the development of Vero Beach. With few exceptions
(Cato n.d.; Lyon 1976:129-130; Van Landingham 1988:4),
Woodbury Lowery's and Higgs' interpretation of Mexia has
been accepted by most archaeologists and historians (see
Table 1 for the derivation and perpetuation of the orthodox
view).
Against two-hundred and fifty years of received wisdom,
we intend to show that the once-enormous archaeological site
alternately known as Barker's Bluff or the Kroegel Homestead
(81R84), about 22 km north-northwest of the Beachland site
(Figure 2), corresponds to the location of the paramount
town of the Ais Indians of east-central Florida as mapped and
described by Alvaro Mexia in the summer of 1605. We propose
that the Inlet of the Ais (referred to as the Rio or Barra of the
Ais in Spanish accounts) existed at the Indian River Narrows
(opposite present-day Winter Beach), and that this inlet closed
sometime between the Plate Fleet Wreck of 1715 and De
Brahm's surveys in 1765. All traces of this inlet at the Narrows
were gone by the time of Romans' survey in 1770-1771.
Jonathan Dickinson's shipwreck journal from 1699 and the
chronicles of Pedro Menendez' Florida adventures corroborate
this interpretation of Mexia. In fact, the standard interpretation
of the locations visited by Dickinson during his 1696 shipwreck
(Andrews 1943: 40-41; Andrews and Andrews 1985; Bushnell

2007; Davidsson 2004:119-127; Rouse 1951:220) contradicts
or ignores the locational information provided by Dickinson.
Locating the town of Ais 22 km further up the Indian
River Lagoon (IRL) than previously thought means that
new locations must be derived for the Spanish fort and the
native group referred to as Santa Lucia, since most of the
locational references to the fort and the people of that name
are made relative to the town of Ais. A close reading of the
Dickinson narrative shows that the "Indians of Sta. Lucea"
(1787[1699]:31) resided in the general area just north of the
Old Indian River Inlet. We make no claim as to the precise
location of the Spanish garrison of Santa Lucia (1565 -1566),
but to the extent that there is a relationship between the Spanish
fort and the territory of the native group by that same name (cf
Milanich 1995:56), the implication is that the ill-fated garrison
was located further north than previously thought, certainly not
as far south as Jupiter Inlet (contra Hann 2003:78; Hoffman
2002:53; Lyon 1976:140, etc.). We also propose Bethel Creek
as the general location for the short-lived Spanish fort at Ais
called Puerto de Socorro (Port of Succor, late 1565), from
which they soon fled and established the Santa Lucia fort
further south. We also propose that 8IR831, the "Pregnant
Turtle" site, is a good exploratory candidate for the site of the
winter town ofAis on the barrier island. We also hope to clarify
some of the imprecision in the use of the place-names "River
of the Ais," "Bar of the Ais," and "River of Santa Lucia."
Our method can be summarized as re-mapping Mexia and
Dickinson-the two best sources of locational information
according to Rouse (1951:55)-and correlating those results
with the archaeological record and the barrier island landscape
signatures left behind by former inlets. Additionally, we also
correlate Mexia's landmarks with those of De Brahm's and
Vignoles' early Anglo-American surveys of the IRL area, as
well as some early Anglo-American maps and recollections
from Menendez' expedition to Ais. The unreliability of most
early geographies of Florida and their inapplicability to the
smaller-scale issues of town and inlet locations is also assessed.

Background on the 1605 Mexia Report
and the Early Geographies of Florida

Mexia (1605, in Rouse 1951:266-274) is the best source
of historical information available for Indian town locations
along the Mosquito, Banana, and Indian River Lagoons during
the seventeenth century (Dickel 1992:42; Rouse 1951:55).

Voi.. 63(3-4) THE FLORIDA ANTHROPOLOGIST SEPTEMBER DECEMBER 2010

VoL. 63(3-4)

THE FLORIDA ANTHROPOLOGIST

SEPTEMBER DECEMBER 2010

THE LORDA ATHRPOLGIST201 VOL 633-4

Figure 1. Overview of prominent locations mentioned in text. Base map courtesy of the Florida
Geographic Data Library (www.fgdl.org), University of Florida.

Unlike the earlier encounters of Menendez, or the later ones
by Dickinson, or Bishop Diaz Vara Calderon (Wenhold
1936), only Mexia predicated his journey on mapping and
documenting geographic and political information. Unlike
Menendez or Dickinson, Mexia made his voyage under no
duress during a period of peace (Rouse 1951:257), whereas
Dickinson and his party were captives in fear for their lives, and
Menendez was both escorting French prisoners and also trying
to pacify a presumably hostile area, all the while desperate to
find provisions for his men and their charges. Though lacking
latitude readings, the Mexia account is richer in geographic
details than either Menendez or Dickinson, especially since
the second (and final) latitude reading in Dickinson's account
is erroneous under any interpretation (less erroneous under our
model than the standard model, however). Many of Mexia's
landmarks along the Indian River Lagoon just north of Ais
are the same landmarks noted by Charles Vignoles during

his 1821-1822 survey of Florida. Their measurements of the
distances between these landmarks, it will be shown, are
similar.
The pandemic inaccuracy of early geographies of
Florida, including maps, nautical logs, latitude readings,
and documented recollections of distances traveled, is well
established. Examples are numerous: Alonso de Chaves
geography from 1537 has latitude readings for the bays of
Florida's west coast that are inaccurate by as much as 1.5
degrees, or 166 kilometers (Milanich and Hudson 1993:42).
Velasco's geography of Florida from 1575 (in Hann 1991:308-
315) places the Tocobaga at 29.50 latitude, or 1.5 degrees
north of the accepted location along the northern reaches of
Tampa Bay. Velasco also places "the River Asis" (Ais?) at 270
latitude, or 19.3 km south of the modern, artificial St. Lucie
Inlet (formerly a natural inlet in the 1830s and 1840s named
Gilbert's Bar), a location which is impossible to reconcile with

ATLANTIC
OCEAN

Blue Goose Midden (lr15)
IRL Narrows
-Bethel Creek (relict inlet)

-Beachland Site (Ir16)

S -Old Indian River Inlet (relict)
Fort Pierce Inlet

artificial) %
Blind Creek 1g
(relict inlet)

St. Lucie Inlet
(Gilbert's Bar)
Jupiter Inlet
/ (not shown,
see Fig. 13)

THE FLORIDA ANTHROPOLOGIST

2010 VOL. 63(3-4)

& (

Table 1. The genealogy of the standard view of the location of Ais.
Author and Role
Publication Date

De Brahm, Gerard Lacking official Spanish geographies, he incorrectly assumed that the
1772 and 1775 Old Indian River (OIR) Inlet was the Rio d'Ayz because it was the only
open inlet between Cape Canaveral and Jupiter Inlet
Romans, Bernard Followed De Brahm, his supervisor on the Florida survey, in assuming
1775 that OIR Inlet was Ais Inlet

Vignoles, Charles First American surveyor of eastern Florida, followed De Brahm and
1823 Romans' identification of the OIR Inlet as the Inlet of the Ais

Lowery, Woodbury Translated Mexia, which showed that the town of Ais was two leagues
1905 north of Ais Inlet

Swanton, John Reed Followed Lowery, placed the main town of the Ais at or near the Old
1922:331 Indian River Inlet, and the Santa Lucia at the St. Lucie River

Higgs, Charles A. Identified a specific site (8IR16), located on the barrier island about
1942 two leagues north of the relict OIR Inlet, as corresponding to Winter
Ais
Andrews, Charles M. Plotted out Dickinson's itinerary in the Indian River Lagoon so as to
1943:41 align with Lowery's interpretation

Rouse, Irving Followed Higgs, Andrews, and Lowery, but with some misgivings,
1951:34, 68, 170-171, 219-220, 258, especially regarding the purported absence of the Sebastian River from
273 note 33 Mexia's map and itinerary
Lyon, Eugene and Homer Cato Discarded the Lowery Higgs interpretation, but proposed locations for
n.d. Ais town and inlet which are further north than those specified by
Mexia and Dickinson

Rudolph, Teresa P. Espoused the Lyon Cato view of the location of Ais town and inlet;
1981 rejected Higgs' notion that the Sebastian River was absent from
Mexia's map
Van Landingham, Kyle S. Stated in passing that Ais was located in the modern town of
1988 Sebastian, but did not specify a particular site

Dickel, David A. Followed Lowery, Higgs, Andrews and Rouse but with strong
1992:42, 122, 220 misgivings about the purported absence of the Sebastian River from
Mexia's map

Anonymous (Encyclopedia of Florida
Indians vol. I) 1998:73
Bushnell 2007:33,35,40
Connor 1964:126
Davidsson 2004:30,148
Hann 2003:86
Hoffman 2002:38 Followed the Lowery Andrews Higgs Rouse view of the
Hutchinson 1987:12,14 geography of the Ais and Santa Lucia Indians
McGoun 1993:10
Milanich 1995:58, 66
Pepe 2000:38
Sigler-Eisenberg and Russo 1986:22
Zimmerman 1996

BRECH AND LANHAM

LOCATION OF THE PARAMOUNT TOWN OF THE AIS

THE FI~oRIDA ANTHROPOLOGIST 2010 VOL. 63(3-4)

Figure 2. Three views of Barker's Bluff (81R84, now called
the Kroegel Homestead) prior to and during its destruc-
tion in 1904. Top: Looking south-southeast. Paul Kroegel's
thatched roof appears above the shell midden. Middle:
Looking west, probably from Pelican Island. Bottom: The
last vestiges of the shell midden, as it was being removed.
Dickel's informant Rodney Kroegel is shown here as a
young child. Photos courtesy of the Brevard County His-
torical Commission.

any other account or interpretation unless one takes Velasco
to mean the southern entrance to the IRL-Hobe Sound lagoon
system as a whole (i.e., Jupiter Inlet or a formerly open inlet in
the area of Gilbert's Bar), and not the Inlet of the Ais.
Early maps of Florida, even Spanish ones, are of little
utility for modern researchers trying to pinpoint precise
locations such as the Indian towns of Ais and Santa Lucia.
Most early maps are essentially armchair productions by
today's standards, culled from the impressions and stories
of explorers, rather than data from surveyors. Early maps
were neither "ground-truthed" nor peer-reviewed. Quite
the opposite-mapmaking was often a state secret so as to

prevent pirates and rival commercial powers from using such
information to their gain. Compounding the problem is the fact
that southeastern Florida was virtually terra incognita even
to the Spanish who controlled it. Well after the first Anglo-
American surveys of De Brahm and Romans (1765-1771),
the chronic inaccuracies and contradictions of early Florida
maps and geographies persisted into the nineteenth century,
especially for the southern interior and southeastern coasts.
As late as 1823, American maps of Florida (Fielding 1823)
showed connections between Lake Okeechobee (called "Lake
Mayaco," "Macoya," "Mayax," etc.) and the St. Johns or the
St. Lucie Rivers, just as often with a connection to the Gulf of
Mexico via the Caloosahatchee or some other west-flowing
river. This false idea of a cross-peninsular water route from the
east coast to the west coast of Florida was a chimera inherited
from Menendez' times, who probably derived the idea from
Indian travel routes (Milanich 1998:156).
Regarding the east-central coast of Florida, as late as
1835 and possibly 1840, some British and American maps
(Bradford 1835; Burr 1835; possibly Greenleaf 1840) showed
two inlets north of Gilbert's Bar, which was usually depicted
as closed, or not shown at all. Several other maps generated
after the publication dates of De Brahm and Romans (1772
and 1775, respectively) also show an inlet north of the Old
Indian River Inlet and south of Cape Canaveral (Arrowsmith
1804; Blunt 1827; Brue 1827; Carey and Lea 1822; Cruchley
& Laurie 1833; Fielding 1823; Grigg 1830; Hinton 1832;
Lewis 1817; Melish 1815; Melish 1816; Morse 1794; Tanner
1823; Von Humboldt 1811). A map from 1890 (Mitchell) even
shows some kind of break in the barrier island at the IRL
Narrows, which might be the very spot proposed here for the
location of the Inlet of the Ais. As convenient as this inaccurate
cartography is for our hypothesized location of Ais Inlet at the
IRL Narrows, nevertheless, we treat these as inherited errors,
and not as direct evidence of our hypothesis. Instead, we defer
to the accuracy of the observations of De Brahm, Romans,
Vignoles and others, who found only one open inlet north of
Jupiter Inlet, that being the Old Indian River Inlet (OIR Inlet).I
We disagree only with De Brahm's and Romans'
assumption that the Inlet of the Ais was the same inlet as the
OIR Inlet (De Brahm 1772, in De Vorsey 1971:206 and De
Vorsey 1972; Romans 1962:281-282 [1775]), which they
referred to as the Hillsborough Inlet. De Brahm named the
IRL "the Hillsborough River," not to be confused with the
modern Hillsborough River on the western side of the Florida
peninsula. We maintain that the Inlet of the Ais at the IRL
Narrows closed sometime after Dickinson (1696), or after the
Plate Fleet Wreck of 1715, and sometime prior to their surveys
(1765-1771).' From the non-Spanish maps available to them,
De Brahm and Romans knew about the existence of the Rio's
Ays, Santa Lucia, and Jobe, but not their precise locations,
thus causing them to assume that the first open inlet south of
Cape Canaveral must be the Inlet of the Ais.
Certainly, the departed Spanish did not go out of their
way to assist the new Anglo-American surveys, instead
maintaining their long tradition of geographic secrecy (Parry
1963:105; Weber 1992:55-56). Neither De Brahm nor Romans
mentions any Spanish map, document, or official assistance

THE FLORIDA ANTHROPOLOGIST

2010 VOL. 63(3-4)

BRECH AND LANHAM LOCATION OF THE PARAMOUNT TOWN OF THE Ms

Figure 3. Mexia's map of the coastal lagoons of east-central Florida. Top: Last two of three panels traced by Charles Higgs
and included in Rouse (1951:266). Bottom: Re-labeled version of Higgs' tracings, encompassing most of the third panel. In
both images, the left margin represents north-northwest while the right margin represents south-southeast. Maps courtesy
of Yale University Press.

from the former administrative center at St. Augustine. Three
factors thus combined to cause De Brahm and Romans to miss
the recently-closed Inlet of the Ais and to assign its location
elsewhere to the south, at the OIR Inlet: 1) Spanish geographic
secrecy; 2) the disjuncture of the British takeover of Florida in
1763; and 3) the dynamic nature of barrier island inlets on
wave-dominated coasts (see below). It was this single error
from the otherwise credible De Brahm and Romans which
led Lowery (1905:433) and most other researchers to assume
that Mexia's "Barra d'Avs" meant the OIR Inlet. Later maps
showing a fully open inlet somewhere north of the OIR Inlet,
however, should not be seen as evidence of an inlet somehow
overlooked by De Brahm, Romans and Vignoles, but rather
as merely symptomatic of the general inaccuracy of the early

cartography of Florida, even during the first decades of the
nineteenth century.
As will be seen below, Mexia is not necessarily better
than other early geographers in terms of the mathematical
accuracy of his distance measurements and certainly not in
terms of the proportions of the map he drew (Figure 3), but
nevertheless his account is much more valuable than any
other for archaeologists and historians in east-central Florida
because of the small scale of its measurements. That is, though
Mexia's margin of error may be as great as his contemporaries,
his field of analysis is much smaller and his data points are
much closer to each other than the large-scale geographies3 of
Florida otherwise available. Mexia's text records 34 explicit
distance estimates over approximately 250 kilometers of

BRECH AND LANHAM

LOCATION OF THE PARAMOUNT TOWN OF THE AIS

THE LORDA ATHRPOLGIST201 VOL 633-4

narrow, coast-parallel lagoons, averaging one distance reading
every 7.36 km. Additionally, Mexia's text and map provide a
great deal of other locational information not included in the
above calculation. Given the small scale, and given the ease
of relocating most of Mexia's landmarks on modern maps of
the IRL, his margin of error is easily compensated. In fact,
Mexia's margin of error is a problem only if you cumulate
all of his distance measurements into one or two large-scale
measurements, a method unprofitably employed by Charles
Higgs (1942:34-35) and criticized by Dickel (1992:42).
Mexia's narrative account, The Derrotero, covers the
entire distance from St. Augustine to the town of Ais and a
barrier island inlet two leagues south of the town, and purports
to be "[a] handy and useful guide which describes truthfully
in every detail the rivers, channels, lagoons, woodlands,
settlements, harbors, shoals, and camps from St. Augustine to
the Bar ofAis" (in Rouse 1951:269). Mexia was commissioned
by Pedro de Ybarra, the Spanish Governor at St. Augustine
(1603-1610), to reconnoiter and map the IRL to facilitate
future navigations and also to assist that year's treasure
fleet on its way back to Spain through the Bahama Channel.
Ybarra was working toward a genuine peace agreement with
the Ais, the most powerful chiefdom along the east coast of
Florida south of Cape Canaveral (Hann 1996:171; 2003:167).
The Ais were also among the most resistant native peoples
to Spanish culture and political influence, and earlier peace
agreements with them had repeatedly proved empty in terms
of their treatment of shipwreck victims and distressed seamen.
Governor Canzo, Ybarra's predecessor at St. Augustine (1597-
1603), had begun to improve Spanish relations with the Ais and
their allies (Hann 2003: 80), but had also failed to consummate
a genuine peace or prevent bloodshed.
The rapprochement forged by Ybarra and assisted by
Mexia's expedition, however, would soon prove genuine.
By December of 1605, the Ais and their allies had fulfilled
their promises made in September to succor and transport
shipwrecked Spaniards back to St. Augustine. A frigate
transporting nine friars headed for the missions of Florida ran
aground near Matecumbe Key during a storm in September
1605 (Geiger 1937:184-185; Hann 1996:172). The Matecumbe
Indians supplied the Spanish with food, water and safe harbor
for five days, after which they laboriously helped unload the
ship, extricate it, and re-load its contents without any pilfering
(Hann 2003:87). This suggests that the Ais exerted a level of
political influence or control far to the south of their paramount
town. Ninety-one years later, in 1696, the Ais exhibited a
similar level of control at least as far south as the Jobe Indians
located at Jupiter Inlet (Dickinson 1699, in Andrews and
Andrews 1985:34).
Less than two months after Mexia's return, the elusive
paramount chief of the Ais finally visited St. Augustine,
bringing with him the chiefs of Surruque and Uribia from the
Mosquito Lagoon, as well as twenty other persons of renown
(Rouse 1951:55). The Ais agreed to refrain from trading with
the English, Dutch, or French and to detain them for capture
by the Spanish. Several male children of both the governor and
the paramount chief were exchanged so as to learn each other's
language (Hann 2003:87). Two years later, Ybarra reported

that he entertained five hundred invited Indian guests from
south Florida during Holy Week (Hann 2003:87). A "Period
of Friendship" with the Ais (1603-1703) was under way,
according to Rouse (1951:53-57, 257), and some historians
(though not all) have traced the beginning of the "Golden Age
of Spanish Missions" back to 1606 (Matter 1990), one year
after Mexia's mission of diplomacy and cartography.
Regarding the interpretation of Mexia's map and report
by twentieth century historians and archaeologists, neither
Lowery nor Higgs nor Rouse nor any published authority has
ever actually plotted out Mexia's itinerary in the IRL. Instead,
attention has been deflected by Mexia's reference to a barrier
island inlet two leagues south of Ais and the lack of any such
inlet during historic times between Ponce de Leon Inlet and
the Old Indian River Inlet. Another deflector of attention has
been that Mexia's map of the IRL is disproportionately larger
south of the Sebastian River than it is north of it, making the
identification of the OIR Inlet as being the Bar of the Ais seem
tenable. The well-deserved credibility of De Brahm, Romans,
and Vignoles helped solidify the idea that the OIR Inlet was
the Inlet of the Ais, despite the irreconcilability of this location
with Mexia's observations and measurements in the IRL (see
below).
Higgs (1942) virtually ignored Mexia's specific
landmarks in the IRL and instead cumulated all of Mexia's
individual measurements into two large measurements: from
St. Augustine to Surruque in the Mosquito Lagoon, and from
Surruque to Ais in the IRL. Had Higgs instead confined
his analysis to the IRL, and cumulated the distance to Ais
from Pentoaya, or from the nearby tip of Merritt Island, the
impossibility of 8IR16 being the paramount town of the Ais
would have been obvious, even using Higgs' improbably large
league unit of 6.08 km per league: 7 /2 leagues from Pentoaya
to Ais means that 8IR16, which is 61 km away from Pentoaya
(62 km from Merritt Island), requires an impossibly large
league unit of 8.13 km per league.
Despite the references to a nearby inlet, and despite the
disproportions in Mexia's map, The Derrotero clearly places
the paramount town of Ais 1.5 leagues south of the Sebastian
River, as shown in the next section. Even some subscribers to
Lowery's and Higgs' interpretation qualify their acceptance of
the town's location at 81R16 with this caveat (Dickel 1992:42;
Rouse 1951:170-171, Rudolph 1981:12), although 8IR16 is 28
km south-southeast of the Sebastian River. As Dickel noted,
it is inconceivable that Mexia somehow missed the Sebastian
River, as maintained by Higgs (in Rouse 1951:273 note 33).4
As long ago as 1775, it has been known that this river is the
second largest freshwater input into the IRL, exceeded only by
the St. Lucie River in terms of volume: "No rivers of any note
fall into its [the IRL's] northern branch except St. Sebastian's,"
observed Romans (1962:273 [1775]).
Dickel (1992:42) comes within a hair's breadth of
stating that 81R84 (Barker's Bluff, now called the Kroegel
Homestead) was the likely location of the paramount town
of the Ais as plotted by Mexia, but nevertheless relents and
defers to the prevailing orthodoxy of Lowery-Higgs-Rouse
due to the apparent absence of Ais Inlet in the area just south
of Barker's Bluff. All of Dickel's other references to the

THE FLORIDA ANTHROPOLOGIST

2010 VOL. 63(3-4)

BRECH AND LANHAM LOcATION OF THE PARAMOIJN[ TOV~N OF THE Ais

town or Inlet of the Ais in his survey of Indian River County
(1992:40,41,47,122,170) are predicated upon the assumption
that 81R16 was the winter location of the town of Ais and that
the OIR Inlet was the Inlet of the Ais. A modem understanding
of the dynamic nature of barrier island geomorphology and tidal
inlet cuts, however, disabuses us of this apparent discrepancy
and any doubts one should have regarding Mexia's descriptions
or approximate measurements of the Indian River Lagoon
in 1605. Barrier island geomorphology also illuminates the
distinctive landscape signatures left behind in the IRL and its
barrier islands by recently-closed inlets, all of which apply to
the area proposed here for the former Inlet of the Ais, and all of
which should be generally useful to archaeologists working on
wave-dominated barrier island systems such those that enclose
the IRL and the Mosquito Lagoon.

Geomorphology of the Barrier Islands
of the Indian River Lagoon

Barrier islands are often called the most dynamic
landforms in the world by the people who study them (Davis
1997:4; Pilkey 1998:xxi; Pilkey and Dixon 1998:15; Pilkey
and Fraser 2003:1,10,12,35-38), and there is still no consensus
on their origins after more than one hundred and fifty years
of research (Davis 1994:168-169; Davis 1997:4-5; Pilkey and
Fraser 2003:2). It was only in the late 1960s and early 1970s
that scientists discovered that barrier islands migrate toward
the mainland during times of rising sea levels (Pilkey and
Dixon 1998:3; Pilkey and Fraser 2003:48). This migration can
be rather quick, even in human terms (Davis 1997:4; Pilkey
1998:4), as evidenced by the Cape Hatteras Lighthouse,
originally built 1500 feet (457.2 meters) from the shoreline
in 1870 (Pilkey 1998:2), but which had to be relocated inland
in 1999 due to shore erosion. The Morris Island Lighthouse in
South Carolina, built in 1876, now stands about 366 meters
offshore.
The very same "destructive" processes which tear barrier
islands apart during storms-overwash and inlet gouging
(Figure 4)-also deposit enormous amounts of sediment into
the lagoons behind the barrier island (Davis 1997:20,29; Fisher
and Simpson 1979:127), creating new land with the help of
sediment-trapping organisms such as grasses and mangroves
(Pilkey and Fraser 2003:48,72) to which the barrier island
retreats if ocean levels rise further (Davis 1997:20; Pilkey and
Fraser 2003:12). During storms and even calm weather, normal
long-shore currents both deposit and remove sediments from
the ocean shoreline, something especially noticeable along the
mouths of barrier island inlets. Romans (1962:287 [1775])
and Vignoles (1977:84 [1823]) both noted that Jupiter Inlet
had opened and closed several times in their recent memory.
Finally, the on-shore/off-shore diurnal alternation of winds
along the coast means that, with help from vegetation, larger
and larger sand dunes are continuously built up, especially
along the ocean face of the barrier island (Davis 1994:161-
164).
Fortunately, in the east-central Florida barrier island
system (BIS)5 the three "destructive" forces which cause
barrier islands to retreat toward the mainland-overwash, inlet

S8- Overwash lobe

~t4

Don ul e LI on
ows l of longshon
5 -Inlet current

ta i drift)
Ita
,-- Relict inlet

--Overwash lobe

S---Overwash apron
.agoon --Thin section

Ocean

Cuspate shoreline
Lagoon (both thin sections
widens and cuspate shorelines
S indicate no recent
inlets or overwash)

Figure 4. Diagram of barrier island processes and the re-
sulting cartographic signatures left behind.

cuts and shore face erosion-all leave distinctive landscape
signatures which can be easily spotted in the IRL using only
a county-scale map or smaller. Preservation of such relict
features seems to be higher in the IRL than in other lagoons
(Almasi 1983:205). Long-term ocean-face erosion is evident
along thin sections of the BIS, a process which primes them
for destruction/rebirth through future overwash and inlet cuts
(Pilkey and Fraser 2003:18,23). More recent ocean face retreat
is also evident whenever a maritime forest incongruously
borders the ocean beach (Pilkey and Fraser 2003:48-49), a
windy, salt-sprayed micro-environment to which they are not
adapted.
Overwash events can be seen as lower-energy versions
of inlet cuts (Almasi 1985; Mayhew 2000:4; Stauble 1988),
although the resistance of the antecedent topography is
equally important. Overwash events (also called "washover")
produce round, fan-shaped lobes of sediment along the back-
barrier shoreline (i.e., along the east shore of the IRL), while
inlet cuts leave cauliflower-shaped plumes of islands and
creeks called flood tidal deltas in the lagoon. Based on aerial
photographs of the artificial Sebastian Inlet from 1945, Almasi
(1985) noted that "relatively large tidal deltas are formed in a
relatively short time [in the IRL]." Inlets which migrate down
the coastline-generally in the direction of the prevailing
long-shore current (Davis 1997:10,29)-leave shore-parallel
tidal creeks and tidal islands in the lagoon behind the locations
where they once existed (Almasi 1983:183; 1985). A recent
example of inlet migration from North Carolina over the course
of just 6 years is shown in Figure 5, which also illustrates the
elongation of the tidal creeks caused by inlet migration. Both
overwash events and especially inlet cuts result in a thickening
of the barrier island and a consequent thinning of the lagoon

BRECH AND LANHAM

LOCATION OF THE PARAMOUNT TOWN OF THE AIS

I
.;

e

._ -,

Figure 5. Recent migration of Mason's Inlet located near Wrightsville Beach, North Carolina over the course of six years.
The white square and circle mark the fixed location of the Shell Island Resort. The total distance migrated is 393.3 meters.
Note how the inlet channel, originally semi-perpendicular to the shore in 1989, is "pulled" into an elongate, shore-parallel
tidal stream as the inlet migrates south. This is the same process that created Spratt Creek and Johns Island Creek in the
IRL. The inlet shown above was later artificially filled in so as to preserve the Shell Island Resort. Photos courtesy of Profes-
sor Alec M. Bodzin, Lehigh University.

BREd-I AND LANHAM LocATION 01 I iii PAR-~~loL N I Uo~~ N OF TIlE Ais

Ridge and Swale
Topography

.a of photo--I

Ridge and Swale
Topography

4 Kiometers

Figure 6. Cartographic and topographic landscape signatures of barrier island sections which have not been overwashed or
cut by inlets in the last 3000(+) years. Top left: the barrier island from Patrick's Air Force Base to Indiatlantic in 5 foot (1.52
m) contours. Top right: aerial photograph of cuspate shorelines at Satellite Beach and Indian Harbor Beach. Bottom left:
Semi-cuspate lagoonal shorelines of Cape Canaveral in the Banana River Lagoon, plus well-developed ridge-and-swale dune
system in 5 foot (1.52 m) contours. Bottom right: aerial photo of semi-cuspate shoreline along the Banana River Lagoon.
Base maps courtesy of the Florida Geographic Data Library.

BRECH AND LANHAM

LOCATION OF TIE PARAMOL NTTO\\ N OF THE AI.

figuree 7. The section of barrier island extending from
iebastian to Vero Beach, an area which shows numerous
igns of geologically recent inlets. The shore-parallel for-
nations of the tidal creeks and their surrounding islands
strongly suggests an inlet which migrated from north-to-
outh. Base map courtesy of the Florida Geographic Data
library University of Florida (www.fgdl.org).

behind it (Almasi 1983:96,185,205-206; 1985; Pilkey and
raser 2003:51-52).
Conversely, BIS landforms which have not been cut
,r over-washed (referred to here as "recently stable" BIS
indforms) show their own, easily identified cartographic and
apographic signatures (Figure 6). A well-developed dune
system, i.e., a series of sand ridges and swales paralleling
ie coast, is a sure sign of progradation (Davis 1997:19-20),
nd thus geomorphic stability (> 3000 years in the IRL). By
contrast, recently unstable sections of the IRL BIS will have
nly one dune line along the oceanic shoreline (Mayhew
000:15).
Most importantly, lagoonal shorelines along recently
table sections of the BIS will become regularly indented
Jith crescent-shaped "bays" of varying lengths, punctuated
y cuspate spits (Almasi 1983; 1985; Mayhew 2000:9-11;
.tauble 1988). It is thought that cuspate shorelines are caused
y naturally occurring circulation cells in the waters of the
igoon, and that these circulation cells require long periods
f geomorphic stability in order to shape the shoreline. In
addition to infusing new sediment into the lagoon in a non-

cuspate, deltaic shape (Davis 1997:29), a nearby inlet would
create strong, shore-parallel tidal flows that would change
direction every six hours or so, disrupting nearby circulation
cells. The IRL BIS has sections with fully-developed cuspate
shorelines (such as from Patrick Air Force Base south beyond
Melbourne Beach), and other areas, such as along the St. Lucie
Sound (the southern end of the IRL) and along the Banana
River shoreline of Cape Canaveral, which are semi-cuspate, or
cuspate-in-the-making (Stauble 1988).
This seeming digression into barrier island geomorphology
is germane to our hypothesis concerning the location of the
town and inlet of the Ais because any proposed location for
the paramount town of the Ais Indians must be 2 nautical
leagues (10-12 km) north of an inlet which survived at least
131 years (1565-1696, being from Menendez to Dickinson),
and probably longer, since the survivors of the Plate Fleet
Wreck of 1715 regrouped at the Bar of the Ais. Such an inlet
location, if now closed, should show distinct cartographic
and topographic signatures of prior inlet activity along the
barrier island and lagoon where it once existed. Conversely,
the geomorphic stability indicated by cuspate spits along the
lagoonal shore of the barrier island north of Sebastian6 is
strong negative evidence against two other proposed locations
for the Inlet of the Ais: 1) near Melbourne Beach (Wheeler
and Pepe 2002:232), some 42 km north-northwest of our
proposed location, and about 64 km north-northwest of the
OIR Inlet; and 2) near News Cut (Lyon and Cato n.d.; Cato
n.d., in Rudolph 1981:12), about 12 km north-northwest of our
proposed location.
The area of the barrier island system that stretches from
Sebastian to Vero (Figure 7), however, is shown by its lack
of pre-Malabar II sites (Brech 2004) to have been heavily
impacted by inlets from A.D. 800 to 1760, since inlets are
much more likely to destroy archaeological sites than overwash
events, which usually just truncate and bury them with oceanic
deposits. The net effect of repeated overwash events causes
many archaeological sites to "migrate" away from the lagoon
toward the ocean shoreline of the barrier island. By contrast,
all of the archaeological sites in this area of the BIS except for
81R56 (the Zaremba site) are still located along the backbarrier
shoreline.
The geomorphic landscape signatures in this area from
Sebastian to Vero also show the distinct signs of recently-
closed inlet activity. The lagoonal shorelines along this
section of the BIS are some of the most irregular in the IRL,
with numerous tidal islands, tidal streams, and other deltaic
configurations (Almasi 1983:103,182-183,205-206). Other
than the recently-closed OIR Inlet, this area is the thickest
section of the BIS between Cape Canaveral and St. Lucie Inlet
that does not show any signs of progradation, and the lagoon
behind it is more narrow here than anywhere in the IRL (even
at the former OIR Inlet) which Almasi (1983:96,185, 205-206;
1985) interpreted as being "due to an increase in the number of
former inlets in this area which caused part of the lagoon to be
filled by tidal flood delta (sic) sediments." Almasi's analysis of
the sediment profiles and faunal remains from his core samples
in the IRL also revealed this area to be the most impacted by
recent inlet cuts (1983: 94-111;1985).7

BRECH AND LANHAM LOCATION OF THE PARAMOUNT To~x N OF I HE Ais

Within this 20 km stretch from Sebastian to Vero, there
are four particular features which stand out: 1) Spratt Creek,
from Sebastian to Wabasso; 2) the large tidal creek used for
the Intracoastal Waterway from the Wabasso Bridge south
to Pine Island; 3) a similarly large tidal creek in the IRL
Narrows known as Johns Island Creek just to the south of,
and probably related to the above two creeks; and 4) Bethel
Creek to the south. As noted, elongated, shore-parallel tidal
creek formations signify that the now-closed inlet(s) migrated
parallel to the shore, "pulling" the normally deltaic (and thus
semi-perpendicular viz. the shoreline) tidal creek formations
into an elongated stream among the islands of the lagoon
(Davis 1997:29). The Bethel Creek relict inlet appears to have
been a separate inlet event unrelated to the above three creeks.
The large tidal creeks used for the Intracoastal Waterway
and the ones known as Spratt Creek and Johns Island Creek
represent the north-to-south migration of one or more inlets,
as per the prevailing north-south longshore current. The
existence of the large Blue Goose Midden site (81R15) near
Wabasso Beach, dated between A.D. 890 and 1300 (Handley
2001:120; Johnson et al. 2000), indicates perhaps that this
inlet was no further north than that site during that time period.
The landlocked nature of Johns Island Creek, the greater
relative depth of these tidal creeks, and the sharp indentation
in the barrier island at the IRL Narrows (modern Indian River
Shores, opposite the town of Winter Beach on the mainland)
is a strong indication that this was the terminal location of this
inlet. By inference, it would seem that Ais Inlet migrated to
this location sometime between A.D. 1300 (i.e., the later range
of the Blue Goose dates) and 1565 (the Mcnendez expedition
to Ais).
Mexia placed the inlet of the Ais 3.5 nautical leagues south
of the Sebastian River (see below). The indentation in the
barrier island at the IRL Narrows is almost exactly 3.5 nautical
leagues from the Sebastian River (17.6 km, see below). Mexia
also recorded that the lagoon "veered east" at the Inlet of the
Ais. An 1887 geodetic survey map (Figure 8) shows that this
section of the IRL, "The Narrows," corresponds to Mexia's
description to a greater degree than anywhere else in the IRL.
This, we propose, was the Rio (or Barra) of the Ais. Our
proposed site of Ais (81R84) is 10.5 kilometers north of the
IRL Narrows, a good match for Mexia's distance estimate of
two nautical leagues, as shown below.

Mexia's Landmarks and Unit of Measurement Along the
Mosquito and Indian River Lagoons

The near-linearity of the coastal lagoons and barrier
island system of east-central Florida, running north-northwest
to south-southeast except for a slight veer at Cape Canaveral,
conveniently allowed Mexia to make measurements of the
coastal lagoons in terms of linear distances along his route
of travel, albeit at the expense of accurate measurements
perpendicular to his route of travel, as revealed in the obvious
inaccuracies of his map. Thus, Cape Canaveral appears
improbably small in his map, whereas the southern extent
of Merritt Island is improbably wide, as are many segments
of the barrier island. There are no latitude readings given in

4-4-
"" .-. \- '. .
: .* > 1 1 */ * . ,

., -

'.' .... .

.17

Two "t .. ... '" ... *'
4 4..

"':; i ';

.:. l n
*^-;. f ', I' .
--

.II

". -'4-.,- -.

:., -. .- -

IRL Narrow ""' .. .fc' ;

(Ais Inlet) ':;'. | : '' "

Figure 8. Coastal geodetic map published in 1887, based
on 1881 geodetic survey. Note how the lagoon could be
interpreted to "veer east" (as per Mexia) at the IRL
Narrows, especially given the greater depth of the water
towards the east compared to the west shoreline behind
Hole-in-the-Wall Island (not labeled). Map detail courtesy
of Alabama Maps (http://alabamamaps.ua.edu/index.
html), University of Alabama.

Mexia's log or map, but then again most latitude readings do
not seem reliable prior to the mid-eighteenth century at the
earliest, and even then both De Brahm and Romans incorrectly
located the OIR Inlet at 270 20 min. instead of 270 30 min., an
error of 18.5 kilometers.'
Mexia's unit of measurement-the Spanish league-is
problematic for a couple of reasons. First, there is the issue
of land leagues versus nautical leagues. Griffin and Miller
(1978:71) suggest that Mexia was using "a Spanish league of
2.63 miles" (4.23 kin), i.e., an East Florida Spanish land league
(Chardon 1980:150). But Mexia's use of small boats to travel
the coastal lagoons, the anticipation of future navigations by
water, plus the one specific reference to a land league in the
area of the Banana River Lagoon (lucnay Lagoon in Mexia's

BRECH AND LANHAM

LOCATION OF THE PARAMOUNT TOWN OF THE AIS

THE LORDA AIHRPOLGIST201 VOL 633-4

distance = 25.2 kilometers
(5.5 "leagues" to Mexia)

SSouthern extent of
Mosquito Lagoon

distance = 29.2 kilometers

southern extent
of Mosquito Lagool

Figure 9. Top: Minimum length of the Mosquito Lagoon,
i.e., beginning from where it widens out and ceases to
resemble a "bayou." Bottom: Maximum length of the
Mosquito lagoon. Base map courtesy of the Florida
Department of Environmental Protection (www.labins.
org).
time) seems to imply that his numerous other uses of the term
"league" (legua) meant nautical leagues, not land leagues.
The second potential problem is that various sources give
different values for the Spanish nautical league during this
time period. Higgs assumed that Mexia and Menendez were
using the league of Las Casas, which he calculated as equaling
4 Italian nautical miles, or 6.08 km (3.78 miles) (1942:34).
Burgess and Clausen (1982:12) define the nautical league
as "three nautical miles," which in English reckonings of
nautical miles would mean 5.5 km. A dictionary of Spanish
nautical terms posted on the internet by a private salvaging
company defines the Spanish nautical league as equaling 3
miles (4.82 km), as does Lyon (1976:141, note 17). Chardon
(1980:144) suggests a different interpretation of the Spanish
nautical league in North America-between 5.56 and 5.93
km. His analysis, however, is based upon cosmographers'
attempts to measure the circumference of the earth so as to

recalibrate official units of measure, such as leagues, so that
they could then be easily divisible into degrees of arc. As such,
most European nautical leagues lengthened over time between
1500 and 1624, as the true circumference of the Earth became
better known (Chardon 1980). It is not at all clear, however,
how quickly or effectively these cosmographic recalibrations
filtered down to the level of junior officers such as Alvaro
Mexia.
Chardon also points out that modern expectations of
precision and standardization of measurement were simply
not part of the mindset of those times, and that "itinerary"
reckonings of distance, at least over land, differed from
legalistic divisions of property in terms of size and expectations
of precision. This is certainly the case with Mexia's nautical
league, which can be shown by taking known points along his
itinerary and correlating these distances on modern maps with
the distances in leagues specified by Mexia. Doing so indicates
that Mexia's league was an imprecise unit that varied between
4.58 and 5.25 km per league, and thus we can assume (as do
Griffin and Miller [1978:71]) that he probably "eyeballed" his
measurements.
We derive, then, not Mexia's league unit, per se, but rather
his margin of error, or rather a margin of tolerance that we
must apply to his measurements between his landmarks in the
immediate vicinity of the town ofAis. All landscape features or
archaeological sites within this range of tolerance would have
to be considered equally against the evidence. Conveniently,
most of Mexia's landmarks in the IRL are quite singular and
easily discerned. As for archaeological sites, there are only
two recorded sites within the range of tolerance we must
grant Mexia's measurements: the massive 8IR84, Barker's
Bluff (now called the Kroegel Homestead), and 8IR849, the
diminutive Duck Point site.
The best places for correlating the distances between
landmarks as specified by Mexia and thus deriving Mexia's
unit of measurement and margin of error are in the IRL south
of the tip of Merritt Island, but we can also use the Mosquito
Lagoon (Surruque Lagoon in Mexia's era) as a kind of control
sample, since it is not directly adjacent to Ais, as are the IRL
measurements below. Dividing the length of this part of the
Mosquito Lagoon (25.2 km minimum, 29.2 km maximum)9 by
the distance given by Mexia, 5.5 league, results in minimum
league unit distance of 4.58 km per league and a maximum
of 5.25 km per league (Figure 9), a range which is less than
the Spanish nautical leagues of 5.56 to 5.93 km indicated by
Chardon, but consistent with the results obtained below from
Mexia's measurements along the IRL. These calculations also
indicate that Higgs' league unit of 6.08 km per league is too
large to apply to Mexia's measurements, and that the East
Florida Spanish land league of 4.23 km is too small.
The near-linearity of the coastal lagoons breaks down at
the headland of Cape Canaveral and the Merritt Island area,
and thus the modern locations of Mexia's landmarks are
correspondingly less clear-cut than in the Mosquito and Indian
River lagoons. Fortunately, we can bypass the more confusing
and debatable points of reference mentioned by Mexia in
the area of Cape Canaveral and the Banana River Lagoon,
since Mexia clearly describes his exit from the Banana River

THE FLORIDA ANTHROPOLOGIST

2010 VOL. 63(3-4)

BRECH AND LANHAM LOCATION OF THE PARAMOUNT TOWN OF THE Ais

Lagoon into the northern IRL: "...there is a narrow pass which
stretches for one-half league to flow out into the Great Bay of
Ais. This narrow pass runs north and south, and at the mouth
of its outlet in the winter-time is the town of Pentoaya" (trans.
by Higgs, in Rouse 1951:273).
Mexia's descriptions in the area just north of the town of
Ais, i.e., coming south from Pentoaya (or the nearby southern
tip of Merritt Island) at the confluence of the Banana River
Lagoon and the IRL, are short enough to warrant quoting with
only a few minor deletions:

This bay [i.e., the IRL south of Pentoava or the
nearby tip of Merritt Island] extends for a distance
of five leagues....There are two mangrove islets
[sic] in the middle of the bay, one large and the other
small, beyond which the bay narrows and is shallow.
At a distance of a league a little fresh water channel,
which discharges from the hinterland, enters on the
west shore. The pine woods continue farther, and by
a mangrove swamp on the east side at a distance of
one and a half leagues is the principal town of this
bay (Ais). The bay proceeds on to the south as I have
said for a distance of two leagues and then veers to
the east and flows out to sea over the Bar of Ais"
(translation by Higgs, in Rouse 1951:273).

The first landscape feature in the IRL described by Mexia
which can be pinpointed on a modern map (Figure 10) are the
Straits of Pentoaya and the associated tip of Merritt Island.
Immediately nearby lie the summer and winter towns of
Pentoaya, which have been located at 8BR1978 (Lanham and
Brech 2007) and 8BR98 (Rouse 1951:199-201) in the towns
of Melbourne and Indian Harbor Beach, respectively. Five
leagues to the south, according to Mexia, are two islands in the
IRL, one larger than the other. A current map of the IRL shows
numerous small islands along the entire length of the lagoon,
but a study of spoil islands in the IRL by the Florida Fish and
Wildlife Commission (2002) shows that there are only two
natural islands in the IRL between 8BR1978 in Melbourne and
the Sebastian River, namely Grant Farm Island and the Mullet
Creek island group, the latter consisting of several closely-
packed islands with small channels throughout. This is also
confirmed by the 1887 map from which Figure 8 is a detail
further south in the lagoon. The distance between Summer
Pentoaya and these islands is 24.1 kilometers, if we measure
from 8BR1978 to the northern tip of Grant Farm Island,
resulting in a minimum league unit of 4.82 kilometers (exactly
3 miles) per league.
Charles Vignoles' survey of the Florida coast in 1821-
1822 provides strong confirmation that Mexia was referring to
Grant Farm Island, although Vignoles refers to it as "Pelican
Island' (1977:44 [1823]), as do certain early American maps
(Burr 1839; Tanner 1825; Tanner 1845, etc.):

Pelican Island is a small mangrove key eight miles
[12.87 km] south of Turkey Creek, nearest the west
shore; the beach from Meritt's [sic] island is bare of
tall growth with only one tuft of tall pines, and one

Figure 10. Mexia's landmarks in the Indian River Lagoon:
the southern tip of Merritt Island and nearby Summer
and Winter Pentoaya; two islands located 5 leagues from
Pentoaya; freshwater river 1 league from islands; town of
Ais 1 /2 leagues south of the river. Base map courtesy of the
Florida Department of Environmental Protection (www.
labins.org).

of cabbage trees, and no mangrove trees or bushes
until thus far when they recommence. Five miles
[8.05 kmn] below is the mouth of St. Sebastian river...
(Vignoles 1977: 44-45 [1823])

The actual distance from Turkey Creek to the northern
tip of Grant Farm Island is 13.4 km, or 8.33 miles. The actual
distance from the southern tip of Grant Farm Island to the
Sebastian River is 4.6 miles (7.4 km); it is 5.18 miles (8.3 km)
from the northern tip of the island. Note also how Mexia's
description that beyond the two islands "the bay narrows and
is shallow" corresponds with Vignoles' description of the IRL
shore as being free of mangroves "until thus far when they
recommence."
Despite the tight correspondence between Mexia's first
island and Grant Farm Island, the identity of the second,

BRECH AND LANHAM

LOCATION OF THE PARAMOUNT TOWN OF THE AIS

THE FLORIDA ANTHROPOLOGISI 2010 VOL. 63(3-4',

Figure 11. Natural islands and spoil islands in the IRLjust
north of the Sebastian River. Grant Farm Island and the
Mullet Creek group of islands are the only natural islands
in this area according to the Florida Fish and Wildlife
Commission. Base map courtesy of the Florida Department
of Environmental Protection (www.labins.org).

smaller island mentioned in Mexia's report is not as clear-cut
(Figure 11). One league south of these two islands, according
to Mexia, was a freshwater river which he included in his map,
unlike the islands, which were not drawn. The size and shape
of this river as rendered on the map closely corresponds to the
Sebastian River (Dickel 1992:42; Rouse 1951:170-171,273
note 33; Rudolph 1981:12; all contra Higgs 1951:273). The
fact that the Sebastian River is the largest freshwater input into
the IRL other than the St. Lucie River also suggests that Mexia
was referring to this drainage. The Sebastian River is 6.3
kilometers south of the southern tip of the Mullet Creek island
group. The resulting league unit distance (6.3 km per league)
is admittedly too large. Another problem is that Mexia's
text implies through its word order that the large island was
encountered first and the smaller one second, whereas Grant
Farm Island is smaller than the combined area of the Mullet
Creek island group, which extends further southeast along the
lagoon.
There are two possibilities: 1) Mexia made an error and
should have written "a little more than a league" or perhaps
"a league and a quarter," phrases he uses elsewhere in his
report; 2) Mexia was referring to a smaller island further south
than the Mullet Creek area, an island which has since eroded
away (as have some of the artificial spoil islands constructed
in the mid-twentieth century, and perhaps the very large island
depicted in the southernmost part of the Mosquito Lagoon in
Mexia's map). Absent further evidence, we subscribe to the
first possibility as a default position, but note there are some
shallow areas of the IRL about five kilometers north of the
Sebastian River which may have been above water in 1605,
near the height of the Little Ice Age.
As mentioned, Mexia's league unit as he applied it to
the Mosquito Lagoon was between 4.58 km and 5.25 kmn,
although in that case the range of difference is not due to
Mexia's inaccuracies per se but rather to the ambiguity of

interpreting Mexia's conception of where the Mosquito Lagoon
begins and ends. It is debatable whether any of his various
distance measurements can be cumulated into larger distance
measurements due to the accumulation of error (Dickel 1992:
42, contra Higgs 1942) and also due to what might be called
his "itinerant" method of data collection and presentation.
That is, Mexia seems to be recording the distances between
landmarks, and not mid-point to mid-point distances that can
be segmented together.
In the case of the Mosquito Lagoon, Mexia's landmarks
were the various Indian towns he encountered, and the
combined lengths of each town probably do not add up to
much distance relative to the overall length of the Mosquito
Lagoon. In the IRL, however, Mexia's two measurements
north of his final measurement to Ais (i.e., from Pentoaya to
the islands and from the islands to the Sebastian River) are
problematic for cumulative analysis due to the relatively large
distance between the northern tip of Grant Farm Island and the
southern tip of the Mullet Creek island group: 2.0 km, or 6.2
percent of the total distance in question (32.4 km). Obviously,
the distance is even greater if we suppose that Mexia was
referring to a now-eroded island south of Grant Farm instead
of the Mullet Creek island group.
In other words, if the first measurement (Pentoaya to the
islands) ends at the northern tip of Grant Farm Island, and the
second measurement begins at the southern tip of the Mullet
Creek islands, as per an "itinerant" form of distance calculation
and presentation, then the distance between the northern tip of
Grant Farm and the southern tip of the Mullet Creek group
was not included in Mexia's measurements. Therefore, this
2 km distance cannot be included in any cumulative average
of Mexia's league unit in the IRL. Thus, the distance from
8BR1978 to the Sebastian River (32.4 km) minus the distance
between the northern tip of Grant Farm and the southern tip
of the Mullet Creek islands (2.0 km) equals 30.4 km, which,
divided by 6 leagues (Mexia), results in an average league unit
in the IRL north of the Sebastian River of 5.07 km per league. "
Adding in the two measurements from the Sebastian River to
the town of Ais and thence to the Inlet of the Ais results in a
league unit of 5.03 kilometers.
Mexia's league unit values are recapitulated in Table 2.

The Location of the Town of Ais in 1605

One and a half leagues south of a freshwater river which
can only be the Sebastian River, according to Mexia, sat the
Pueblo Grande of the Ais Indians.
Using the minimum and maximum league unit distances
calculated above from the known locations provided by Mexia
(4.58 to 6.3 km per league), and multiplying by 1.5 leagues,
the town of Ais should exist some 6.9 to 9.5 kilometers south
of the Sebastian River along the mainland shore of the IRL.
At a distance of 6.9 kilometers south of the Sebastian River,
just inside the minimum range given above, once stood an
aboriginal shell mound alternately known as Barker's Bluff
or the Kroegel Homestead (81R84), which was by far the
most voluminous shell mound in the entire Ais area (from the
Banana River Lagoon to the modern St. Lucie Inlet).

THE FLORIDA ANTHROPOLOGIST

2010 VOL. 63(3-4)

BRECH AND LANHANI Loc ATION OF ~HE PARAMOUNT TOWN OF THE Ais

Table 2. Mexia's league units in the Mosquito and Indian River Lagoons.

Mosquito Lagoon

Mosquito Lagoon, minimum

Mosquito Lagoon, maximum

Average league unit in Mosquito Lagoon

4.58 km per league

5.24 km per league

4.92 km per league

Indian River Lagoon

Pentoaya (Brl978) to island pair (5 leagues)

Island pair to Sebastian River (1 league)

Sebastian River to Ais

(1 /2 leagues)

4.82 km per league

6.3 km per league

4.6 km per league

Ais to Ais Inlet (2 leagues)

Average league unit in IRL

Total Range of Variation

Unfortunately, Paul Kroegel sold this mound to St.
Lucie County in 1909 so as to provide road bed material for
Highway US 1 from Micco in Brevard County to Stuart in
modem Martin County (Westfahl and Keyes 2003), a straight-
line distance of over 78 kilometers, and an indication of the
mound's immense size. The site of the Kroegel Homestead
has been purchased by Indian River County and made into
a preservation area as a complement to the Pelican Island
Preserve just across the lagoon (not the Pelican Island referred
to by Vignoles, which is now called Grant Farm Island). Little
or nothing remains of the Native American components of
this site, although Dickel (1992:126, 217) reported a residual
"shelf of midden" measuring some 30 to 50 cm, and his
informant Rodney Kroegel alleged that a sand burial mound
still lies on the property. Kroegel's collection from the mound,
taken as it was being destroyed when he was still a child,
contains check stamped Malabar 11 pottery (Dickel 1992:217),
an indication that it was inhabited sometime between A.D.
800 and 1763 (though this does not exclude the possibility of
earlier occupations).
Rouse, who referred to this site as 8IR5, obviously did
not visit the remnants of this site, examine the old photos,
or talk to the Kroegel family, and seems to have conflated
Barker's Bluff with Two Dollar Bluff to the south: "William
Scott [harbormaster] of Melbourne informed us (personal
communication) that there is a sand or shell mound on the
Cragler [sic] place, at Two-Mile Bluff [sic], some 2 miles
south of the city of Sebastian [sic]" (1951:170-171).
The next large site south of Barker's Bluff is 8IR854,
the Wabasso Causeway site, surveyed by Dickel (1992:205-
6). According to Ruth Stanbridge, local historian and former

5.25 km per league

5.03 km per league

4.58 to 6.3 km per league

commissioner of Indian River County, this site is the remnant
of a once larger shell mound originally called Two Dollar
Bluff, destroyed for development during the construction of
the Wabasso dock c. 1890 and later the bridge itself c. 1926
(personal communication 2007). Both Barker's Bluff and
Two Dollar Bluff served as navigation landmarks prior to
their destruction, according to Stanbridge, and both were used
as triangulation points in the 1881 geodetic survey. Dickel's
description of the site (1992:206) is based on its condition
in March 1992, and does not reference its pre-development
enormity: a lagoon-fringing linear ridge of shell midden
measuring 30 to 50 meters wide by about 500 meters long and
70 centimeters deep, "[t]he site is difficult to interpret without
more data. It is not concentrated enough to consider [it] a
village site, but the abundance of ceramic and other midden
material suggests at least temporary, periodic occupations
over an extended period of time." Dickel found significant
quantities of check-stamped Malabar II pottery at the remnants
of this site, again indicating occupations later than A.D. 800
but not necessarily excluding earlier ones.
Stanbridge believes that 8IR854 might be a worthy
candidate for the main town ofAis, with the winter town ofAis
located at 8IR 15, the vast Blue Goose Midden site on the barrier
island across the IRL and several nearby middens (especially
8IR848, the Fishing Flat site, estimated to be 12 hectares in
extent according to Dickel [1992:193] and considered to be
part of 8IR15 by Handley and Chance [2000:5]). However,
the distance to the Sebastian River from 81R854 is 12.7 km,
a full three kilometers more than the maximum range of 9.7
km derived above, and thus 2.3 km per league greater than
Mexia's one aberrant distance unit of 6.3 km per league used

BRECH AND LANHAM

LOCATION OF THE PARAMOUNT TO1W OF THE AIS

THE FLORIDA ANTHROPOLOGIST 2010 VOL. 63(3-4)

in the IRL. That is, Mexia would have to have used a league
unit of 8.5 km per league in order to align his location ofAis
with 8IR854. Using Mexia's average league unit calculated
above (5.03 km per league, see Table 2), it seems likely that
Mexia would have said that Ais was 2 /2 leagues south of the
Sebastian River, not 1 V2, if this had been its location in 1605.
Nevertheless, the Wabasso Causeway site (8IR854,
formerly Two Dollar Bluff) has some characteristics that make
it worth considering as a paramount town location. First, of
course, is its presumably enormous size prior to destruction.
Second, as Stanbridge notes, both the Blue Goose Midden
site (8IR15) and the Fishing Flat site (8IR848) are directly
across the IRL on the barrier island, corresponding to Mexia's
description about winter and summer town locations. Third,
11.3 kilometers south of 8IR854, or about two nautical leagues,
are the unmistakable landscape signatures of a former barrier
island inlet cut, now called Bethel Creek (see Figure 7). In
fact, the landscape signatures of this geologically recent inlet
are perhaps more obvious here than at the Ais Inlet location
proposed by us, although our location shows a much larger
and more elongated tidal creek system (the Intracoastal and
John's Island Creek etc.) than at Bethel Creek.
Environmental Services' reports of its excavations
at the Blue Goose Midden (Handley 2001; Handley and
Chance 2000) directly across from 8IR854 does not mention
the possibility of this being Winter Ais, nor does Florida
Archeological Services' report (Johnson, Basinet and Fradkin
2000), but neither was this part of their research designs, nor
was the winter-summer relation between Blue Goose and the
remnants of 81R854. Both authors of this paper served as
excavators on the Blue Goose project for ESI, and witnessed
first-hand the density and vastness of this site, as well as the
numerous well-crafted items found in the excavation units
which were wet-sifted, including a very large shell drinking
cup and many large, drilled shark teeth.
However, none of the excavations at Blue Goose, both by
ESI and FAS Inc. produced any Spanish artifacts, unlike some
much more limited excavations further north on the barrier
island at the Pregnant Turtle site (8IR831), also excavated by
the authors of this paper for FAS Inc. (Johnson and Basinet
2002). Two radiocarbon dates from shell were obtained
from Blue Goose (81R15) by ESI (Handley 2001:113), both
from basal levels, one from a pit feature beneath the midden
(A.D. 890-1180), and one from just below the leaching zone
underneath the midden (A.D. 1060-1300). FAS Inc. obtained
a single carbon date from the Blue Goose Midden of A.D. 890
(Johnson, Basinet and Fradkin 2000:2,48).
Given the lack of any Spanish artifacts out of numerous
excavation tests (42 square meters, combining FAS' and ESI's
test units, plus about 150 shovel tests)," and given the pre-
Spanish Period carbon dates, and given the incorrect distance
from the Sebastian River as specified by Mexia (even using
the most liberal range of tolerance), the Wabasso Causeway
site (8IR854, formerly Two Dollar Bluff) and the Blue Goose
Midden site (81R 15) can be confidently excluded as candidates
for the summer and winter towns of Ais (respectively) at
the time of Mexia's 1605 expedition. However, given the
relocation of "Old Surruque" and "Old Ais" noted in Mexia's

map, it remains logical to suppose that 8IR854 and the large
barrier island sites of 8IR15 and 8IR848 could have been the
summer-winter paramount towns of the Ais/Malabar II peoples
at some point between AD 800 and prior to Mexia's visit.
The only other archaeological site along the mainland
shore of the IRL between Barker's Bluff (8IR84) and the
Wabasso Causeway site (8IR854) is the Duck Point site
(81R849), located 2.41 km south of 8IR84, and surveyed by
Dickel (1992:195-196), FAS Inc. (Johnson and Basinet 2000)
and New South Associates (Cantley et al. 2002). All three
reports describe a relatively small site. Johnson and Basinet
(2000:1,18,33-34) thought that the entire site was dredge
material, since Duck Point does not seem to appear on an
1859 plat map. A follow-up report on a nearby site by New
South Associates, however, notes that this landscape feature
appears on a soil survey map from 1915 and thus they question
Johnson and Basinet's conclusions (Cantley et al. 2002:11,13).
The 1887 geodetic coastal survey map from which Figure 8 is
a detail shows that Duck Point appears even earlier-indeed, it
was used as a triangulation point on the 1881 survey map from
which this 1887 map was derived. Given the unlikelihood of
any dredging of the IRL between 1859 and 1881, Johnson and
Basinet's diagnosis should be reexamined. Their conclusion
that the 226 pottery sherds found along the shore of the site
were somehow "pushed up" from the IRL (2000:32-34) also
seems curious.
In any case, none of the three archaeological investigations
at Duck Point revealed anything that could be construed as the
remnants of a large Indian village, much less the paramount
town of the Ais. Thus, there is no archaeological site other than
81R84 that lies within the possible range of distances Mexia
specified between the town of Ais and the Sebastian River.
Dickel's reconnaissance of 8IR84 involved an interview
with Rodney Kroegel, the grandson of its original owner:
"Family lore suggests there is at least one burial [on the
property]... Kroeger [sic] family legend has it to be 'an Ais
chief,' although this of course is unverified as to existence,
let alone alleged status!" (Dickel 1992:217). Given that 8IR84
is the best and only viable candidate for the paramount town
of the once-powerful Ais Indians, Kroegel's "family legend"
seems accurate on this account, even if by accident, and it
is interesting to speculate whether they derived this family
lore from the native Seminoles with whom they occasionally
interacted (Lockwood 1975:17). Even if they did, of course,
the Seminoles themselves may have derived their assessment
of 8IR84's chiefly status simply based on its enormous size,
and not from any information handed down from its original
Ais inhabitants.
Pinpointing the summer location of the paramount town
of the Ais on modern maps at a location 22 km north-northwest
of Higgs' and Rouse's Winter Ais location at 8IR16 means
that other areas must be examined for the location of Winter
Ais on the barrier island, as well as the Spanish garrison of
Santa Lucia, since most locational references to this garrison
are based upon its distance from Ais. Regarding Winter Ais,
there are several small to medium-sized sites on the barrier
island opposite 81R84: 81R49-51, 81R828, 81R829, 81R986
and 8IR831 ("Pregnant Turtle"), the last one having revealed

THE FLORIDA ANTHROPOLOGIST

2010 VOL. 63(3-4)

BRECH AND LANHAM LOCATION OF THE PARAMOUNT TOWN OF THE Ais

a Spanish olive jar fragment and several Spanish-origin nails
and metallic objects out of 14 square meters of excavation
units (Johnson and Basinet 2002:47,65), all from the southern
fringe of the main part of the site, which remains unexcavated,
and hopefully protected.
"Pregnant Turtle" (8IR831) might not be an impressive-
looking site but it is important to note, however, that
Dickinson's description of "Jece" (Winter Ais) indicates that
it did not boast the largest shell mounds in its vicinity. When
the chief's house was flooded to its rafters during a violent
storm, the people in the house sought refuge elsewhere, "a
considerable distance [away]....where a place was made for
the Caseeky [cacique] or king...[where there was] another
oyster hill that the water was not got over yet" (Dickinson
1699, in Andrews and Andrews 1985:32). The point is that
Winter Ais, unlike Summer Ais, need not be characterized by
vertically large shell middens, or at least not the largest in its
area.
It should also be noted that in these passages Dickinson is
clearly describing an overwash event that flooded over Winter
Ais. Given the virtual disappearance of the Ais soon after
Dickinson (Brech and Lanham 2009), this overwash event that
he recorded means that the site of Winter Ais, if undisturbed
to the present, might be covered over with a layer of overwash
deposits.
As mentioned, 10.5 km (or two of Mexia's leagues)
southeast of 8IR84 are the IRL Narrows, our proposed location
for the Inlet of the Ais, referred to as both the "Rio d'Ayz" and
"Barra d'Ayz" in Spanish accounts. Immediately south of the
Inlet of the Ais, Mexia's map refers to the southern half of the
lagoon as "the River of Santa Lucia." It should be noted that
under both the orthodox geography, whereby the inlet shown
is the OIR Inlet, and our proposed correction, this cannot refer
to the mainland St. Lucie River, but must instead refer to the
southern portion of the IRL. Both Dickinson and Romans,
independently of themselves and Mexia, referred to a northern
and southern half of the IRL drawn at the IRL Narrows. The
fact that Mexia would refer to this part of the IRL, so close to
the Inlet of the Ais, as the River of Santa Lucia suggests that
the abandoned Spanish fort or the indigenous people referred
to by that name were located further north than either Jupiter
Inlet or the St. Lucie River.
Once De Brahm decided that the OIR Inlet must be the
Inlet of the Ais, he solved the "problem" of where to locate
the "Rio" of Santa Lucia (which was frequently depicted on
non-Spanish maps) by assigning it to the mainland river still
known as the St. Lucie.

The 1565 Menendez Expedition to Ais and the
Establishment of Puerto de Socorro and Santa Lucia

Ais: Forty years before Mexia reconnoitered the IRL,
Pedro Menendez de Aviles had cajoled his way through Ais
territory while escorting 75 French prisoners from Cape
Canaveral, leaving behind gifts in the empty Ais villages as
he proceeded south along the ocean beach (Lyon 1976:129)
toward the winter town ofAis, where he was warmly received.
But Spanish relations with the Ais soon soured after Menendez

left most of his poorly provisioned men and captives behind
at an ill-fated garrison three leagues south of the winter town
of Ais while he and a few others went to Havana to obtain
supplies. During his absence, this temporary garrison, called
Puerto de Socorro in some accounts (Lyon 1976:130) and
Rio de Socorro in others (Menendez de Aviles 1565, in Lyon
n.d.:l) was abandoned, its defenders retreating 12, 15, 20 or
23 leagues to the south (depending on the account) where they
established a fort called Santa Lucia.
The reports from the Menendez expedition are consistent
regarding the distance between Cape Canaveral and Ais being
15 leagues (Barcia, in Kerrigan 1951: 98; Menendez de Aviles
1574, in Connor 1925:33; Solis de Meras 1964: 126), and the
context of these statements indicate that the distance given is
that of Ais Inlet and not the town of Ais. The ocean beach
opposite the Kroegel Homestead (8IR84), i.e., the general
location of the winter town of Ais, is 73.6 km south of Cape
Canaveral measured point-to-point, whereas our proposed Ais
Inlet is 81.8 km (Figure 12). Dividing the first distance by 15
leagues yields a league unit of 4.91 km per league; dividing
the distance to the IRL Narrows by 15 leagues yields a league
unit of 5.45 km per league.
Higgs claimed to employ an improbably large league
unit of 6.08 km per league in order to make the Beachland
site (81R16) correspond to this description in the Menendez
documents (1942:34). In fact, 6.08 km per league is too short
to reach the Beachland site (8IR16), which is 92.9 kilometers
from Cape Canaveral, requiring an actual league unit of 6.19
km per league. Obviously, an even larger league unit is needed
to make it 15 leagues from Cape Canaveral to the OIR Inlet
(106.3 km, thus 7.1 km per league), the location of Ais Inlet
according to De Brahm, Romans, Lowery, Higgs, Rouse,
etc. Chardon's analysis of the Spanish league unit (1980)
concludes that the Spanish nautical league equaled only 5.56
-5.93 kilometers. As we have seen, Mexia seems to have used
a nautical league unit averaging between 4.8 to 5.3 km per
league. An Anglo-American sailing guide from 1796 gives
the distance between Cape Canaveral and the OIR Inlet "Ayes
Inlet, now called Hillsborough Inlet" as "about 16 leagues"
(Furlong 1796:80), not 15, although it is perhaps unfair to
compare Spanish and American nautical league units hundreds
of years apart.
Lyon and Cato's view that Ais Inlet was located at News
Cut (see Figure 7), just north of Spratt Point (n.d., and n.d. in
Rudolph 1981:12) requires a nautical league unit of just 4.73
km per league (71 km divided by 15 leagues). Wheeler and
Pepe's casual reference to an Ais Inlet near Melbourne Beach
(2002:232; also Johnston 2000:12, who places De Brahm's
"Hillsborough Inlet" at this location), just 42.7 km south of
Cape Canaveral, is obviously disqualified by this 15-league
estimate. Stanbridge's Bethel Creek suggestion (n.d.) is 88.6
km south of Cape Canaveral, resulting in a league unit of 5.91
km per league, at the larger end of Chardon's 5.56 5.93 range
of values, but still tenable.
Santa Lucia: These same accounts and interpretations
from the Menendez expedition to Ais and its aftermath
frequently contradict each other such that, taken together,
they almost do not present a coherent sequence of events

BRECH AND LANHAM

LOCATION OF THE PARAMOUNT TOWN OF THE AIS

Figure 12. The 81.8 km distance between Cape Canaveral and the IRL
Narrows corresponds to the 15 league distance between the Cape and the
Inlet of the Ais as recorded by Menendez and others. The resultant league
unit is 5.45 km per league. It requires an impossibly large league unit of 7.1
km per league in order to make the Old Indian River Inlet correspond to
this 15 league distance estimate. Satellite photomontage by Kathleen Hill,
provided courtesy of the Smithsonian Marine Station, Fort Pierce, Florida.

when it comes to the establishment of the Spanish garrison
named Santa Lucia, making any detennination of its location
extremely problematic. Hutchinson (1987:12-15) considers
it equally likely that Santa Lucia was located at Jupiter Inlet
or the St. Lucie River, 26 km away. It is not even known for
certain how long Santa Lucia persisted as a fort-until 1566
or 1568. Its precise location "remains unknown" according to
Lyon (1989:154; reiterated by Davidsson 2002:74), and we do
not propose a specific site location in this paper.
Nevertheless, a necessary corollary of a more northerly
location for the town of Ais should involve a more northerly
location for the Spanish fort of Santa Lucia, since most
locational references to this garrison are phrased in terms of
its distance from Ais. Also, acknowledging and itemizing the

ambiguities and confusion that exists regarding the location
of Santa Lucia is a necessary first step towards resolving it,
whereas many historical accounts tend to minimize or conceal
the ambiguities in order to present a digestible account to the
reader. We offer the reader no such accommodations.
There are currently four general areas thought to be
possible locations for the ill-fated fort (Figure 13): the St. Lucie
Inlet or Gilbert's Bar (Dickel 1992:40; Lowery 1905:434), the
north shore of the St. Lucie River (Held 1949; Hutchinson
1987:8,13; Milanich 1995:56; Sahl 1986:108), Jupiter Inlet
(Davidsson 2002:69; Hann 2003:78; Hoffman 2002:53;
Hutchinson 1987:14-15; Lyon 1976:140), or, very much the
minority, the town of St. Lucie located just opposite the site
of the now-closed OIR Inlet (Chatelain 1941: Map 21; Cline

BREd AND LANI-IAM LOCATION OF 1 I-fE PARAMOUNT To~ N OF THE Ais

Old Indian
River Inlet

North shore
of St. Lucie
River
St. Lucie
Inlet

Lake
Okeechobee Jupiter
Inlet
scale

Figure 13. The four general areas proposed for the ill-
fated Spanish garrison of Santa Lucia (1565-1566). From
south to north: Jupiter Inlet; the north shore of the St.
Lucie Inlet; the north shore of the St. Lucie River on the
mainland; and the mainland town of St. Lucie, opposite the
OIR Inlet. Base map courtesy of the Florida Department
of Environmental Protection (www.labins.org).

1974: Map 3; Matter 1990:26). Our contribution is that the
evidence from Mexia and Dickinson clearly indicate against a
location as far south as Jupiter Inlet, contra Lyon (and others),
and in favor of a location at the OIR Inlet, i.e., the minority
position mentioned above.
After reaching the town of Ais on November 4, 1565,
Menendez explored the lagoon by boat and noted a small
harbour (Solis de MerBs 1964:127) located 15 leagues south
(15 nautical leagues south of 8R84 is the modern St. Lucie
Inlet). Another account states that Menendez searched for a
harbor further south in the IRL and failed to find one (Lowery
1905:217). On his return to Ais, Menendez and the Chief of
Ais decided to relocate most of the soldiers and their French
prisoners three leagues south of the paramount town of the
Ais so as to avoid conflicts (Lowery 1905:217; Lyon 1976:
140; Solis de Meras 1964:128). This location is referred to
as Puerto (or Rio) de Socorro, the Port (or River) of Succor,
and is described by Barrientos (1965:74) as being "on a river
bank...where cocoa-plums (an Indian fruit) are available,
as well as palmetto buds, prickly pears, and fish." Connor's
translation of Solis de Meras does not use the term "river
bank," but rather "it was on the river" (1964:128).
Menendez left his men at this initial garrison while he
went to Havana to obtain much-needed supplies. During his
absence, which he prolonged by a side-trip to the Calusa in
southwest Florida, food shortages and attacks from the Ais
induced a large portion of this garrison to flee south, possibly in
mutiny. The distance traveled varies depending on the account
but Lyon believes that the actual distance was between 12 and
15 leagues (1976:144, note 17). It is never explicitly stated
whether these recollected distances begin at the town of Ais,
the Inlet of Ais, or Puerto de Socorro, although it seems logical
to assume the latter. Upon coming to either a river or an inlet

which impeded their flight (Lyon n.d.), normally considered
to be either the St. Lucie Inlet/Gilbert's Bar (Dickel 1992:40-
41; Lowery 1905:434) or the north shore of the St. Lucie
River (Lyon 1976:140), the fleeing (or mutinying) garrison
established a second redoubt, or at least a rallying point. Prior
to Lyon, most historians considered this to be the location of the
garrison of Santa Lucia, but Lyon interpreted the evidence to
indicate that a third redoubt was established six leagues south
of the second redoubt (or rallying point), and that this third
redoubt, founded on Saint Lucia Day (December 13, 1565), is
the true location of Santa Lucia. Lyon thus places Santa Lucia
just north of Jupiter Inlet, an interpretation which has been
accepted by some historians and archaeologists (Davidsson
2004:68-73,161; Hann 2003:78; Hoffman 2002:53).
The Mexia map, crafted 39 years after the abandonment
of the fort, notes that just south of the Inlet of the Ais was the
River of Santa Lucia (Figure 3), something hard to reconcile
with a Jupiter Inlet location for the fort of Santa Lucia if we
assume that there was a relationship between Mexia's place
name and the eponymous fort. The Mexia place name also
implies that the southern section of the IRL was called the
River of Santa Lucia, as was its oceanic inlet, while "the River
of the Ais," to the extent that it was used for anything other than
the Inlet of the Ais, referred only to the northern section of the
IRL. This usage of place names corresponds with Dickinson's
description that "the Indians of Sta. Lucca" commanded the
southern part of the IRL, while the "Jece" (Ais) commanded
the northern part.' As we have seen, Romans also divided the
IRL into a northern and southern branch, as did other early
Anglo-American observers.
Also, it is impossible to reconcile Lyon's location of Santa
Lucia at Jupiter Inlet with Dickinson's account of the amount
of distance between the various inlets they crossed, a point
noted by Wheeler and Pepe (2002:233), and elaborated on
here. In an October 1992 lecture at the McClarty Museum,
Lyon (1992) admitted the difficulty of locating Dickinson's
landmarks based upon his understanding of the political
geography of the east coast of Spanish Florida. In fact, the only
way to reconcile Lyon and the orthodox view of Dickinson
(setting aside our revised interpretation, for the moment) is
to suppose that Dickinson's identification of the native group
and the ruins was either mistaken or merely unrelated to the
Spanish fort.
To be fair, this is not as unlikely as it might sound."
None of the coastal Indians south of Cape Canaveral were
economically tied to any particular place along the incredibly
productive IRL other than shifting inlet cuts, to the extent
that inlets in the IRL have been shown to increase biological
diversity (Buzas 1995; Brown-Peterson and Eames 1990;
Gilmore 1995; Mikkelsen et al. 1995; Schmalzer 1995)
and perhaps draw human habitation as well (Brech 2004;
McMichael 1977:188-192). The "Indians of Sta. Lucea" in
1696 need not have resided in the same general vicinity of
where they were in 1565, especially if we assume there was a
sharp demographic contraction over those 131 years.5
One hundred and thirty-one years is also more than
enough time for natural inlets in the IRL to form, migrate,
or close. Early maps and descriptions of the OIR Inlet, with

LOCATION OF THE PARAMOUNT TOWN OF THE AIS

BRECH AND LANHAM

THE FLORIDA ANniRoPoLoGis I 2010 Voi.. 63(3-4)

Figure 14. Proposed general location for the short-lived
Spanish encampment at Ais called Puerto de Socorro, said
to be located three leagues south of Ais. Base map courtesy
of the Florida Department of Environmental Protection
(www.labins.org).

its large deltaic islands, its three channels, and its "1500
link" breadth (301.8 meters) (De Brahm 1772, in De Vorsey
1971:207), all indicate a very large inlet, and thus probably,
but not necessarily, a long-lived one as well. The OIR Inlet
is reliably documented from the late 1760s by De Brahm and
Romans. It is an assumption on our part (and everyone else's)
that it also existed during Dickinson's 1696 shipwreck, almost
74 years earlier. But it may not have existed at all in 1565,
and thus the native population later referred to as Santa Lucia
might have been clustered elsewhere to the south.
The best evidence for Lyon's more southerly location
for Fort Santa Lucia comes from the eyewitness account of
Juan de Soto (in Connor 1925:67; Lyon 1976:140-141), which
mentions that the fort of Santa Lucia was at Jeaga/Xega, a
group normally considered to be either the same as the Jobe
at Jupiter Inlet (Davidsson 2004:5,13,33,41,121,137; Rouse
1951:38-39,57; Swanton 1922:389-390), or located south
of Jobe, perhaps at Lake Worth (Milanich 1995:56; Wheeler
1992; Wheeler and Pepe 2002:233). Nevertheless, a special
pleading is required to explain the discrepancy between the
northward location of Dickinson and Mexia's "Santa Lucia"
vs. the southerly location of Santa Lucia as posited by Lyon.
Absent further data, the preponderance of evidence would
seem to require that we accept the more northerly location
indicated by Mexia and Dickinson, at least for the native group
in the years 1605-1696.
Puerto de Soccoro: Unlike the ambiguity regarding
the distance from Ais to Santa Lucia, the distance from Ais
to Puerto de Socorro, when given, is always recounted as 3
leagues. The only ambiguity is deciding whether to begin the
measurement at the town of Ais (Hann 2003:78) or the Inlet
of the Ais, although the former seems more likely. Sixteen
kilometers (or three nautical leagues) south of the town of
Ais (8IR84) is Bethel Creek (Figure 14), a relict inlet which
corresponds to the description of a river bank or port. The fact
that there are no known Indian sites in the immediate vicinity

of Bethel Creek is not negative evidence, since the garrison
was deliberately located away from the indigenous population.
Lyon's unpublished manuscript on the Santa Lucia
garrison (n.d.:l) allows for the possibility that Puerto de
Socorro and the second Spanish redoubt (or rallying point)
could have been located on either the mainland or the barrier
island, whereas his book on Pedro Mcnendez (1976:140-141)
states unequivocally that the defenders (or mutineers) fled
as far south as the St. Lucie River on the mainland, which
they could not cross. The testimonies of Captain Juan Velez
de Medrano and Miguel Rodriguez, however, clearly imply
that the Spanish fled down the barrier island beach since they
were unable to cross "inlets" according to Velez, or "a bar,"
according to Rodriguez (Lyon n.d.:2-3). Yet many historians
and archaeologists seem to assume a mainland location for
both Puerto de Socorro and the second redoubt (Davidsson
2004:65,159; Milanich 1995:56).
In addition to the Velez and Rodriquez testimonies cited
above, and the coastal location implied by the name "Puerto,"
we believe the logic of the situation dictated that all of the
Spanish redoubts would have been located on the barrier
island and not the mainland. A mainland location is much
harder to defend for several reasons: on the mainland, there are
between 180 and 360 degrees in the fort's field of fire which
have to be cleared of underbrush and trees so as to maximize
the advantage of firearms. On a thin section of barrier island,
the field of fire that needs to be cleared is much smaller and
probably easier to cut through (mangroves from the lagoon,
and beach scrub from the thin stretches of barrier island to the
north and south). A mainland location also has a much bigger
demographic radius, meaning that a fort on the mainland
would be more surrounded by native groups than one on a thin
section of barrier island.
More importantly, broad avenues of retreat, and freedom
from ambush thereon, are far superior along the barrier island
ocean beach, especially at low tide, than on a mainland trail
through the woods or along the western shoreline of the IRL.'
Potable water is universally present on the barrier island
just by digging below sea level, a point noted as early as
Dickinson (1699, in Andrews and Andrews 1985:12), Romans
(1962:283 [1775]), and Vignoles (1977:43 [1823]). Based on
the remains of "barrel wells" found at the Survivor's Camp of
the 1715 Plate Fleet Wreck (Burgess and Clausen 1982), the
Spanish certainly knew of this water source at least as early
as the eighteenth century. Barrel wells dating to the sixteenth
century have been discovered at St. Augustine (Deagan 1987).
Most importantly, the absolute necessity of signaling and
rendezvousing with ocean-going ships would seem to require
a barrier island location for any ad hoc Spanish fort such as
Puerto de Socorro or Santa Lucia.
While we cannot propose a specific location for the
Spanish fort of Santa Lucia as we can for the main town of Ais
(81R84), its possible winter satellite (8IR83 1), and the nearby
Ais Inlet (IRL Narrows), the indications from Mexia's map
and from Dickinson's account are that the Inlet of Santa Lucia
was located at the OIR Inlet. Admittedly, the distance between
the OIR Inlet and our proposed location for the town of Ais
(33 km) is too small to accommodate the recollected distances

THE FLORIDA ANTHROPOLOGIST

2010 Vol.. 63(3-4)

BRECH AND LAN 11AM LOCATION OF THE PARAMOUNT TOWN OF THE Ais

IRL Narrows
Inlet of Me Ais
accerdho to our model

Old Indian River Inlet
Inlet of the Als
ac. to the standaidnmodel
Inlet of the Santa Lucia
ace. to our model

St. Lucie Inlet
Inlet of he Santa Lucia
ac. to standard model
Closed, not open In 169B
asc. to our mowd

RJupiter Inlet
Figure 15. Map of Dickinson's journey along the ocean beach from Jobe
(Jupiter Inlet) to the Inlet of the Ais. Dickinson's journal indicates that it
took him about three times as long to travel from Jobe to the first inlet (the
Inlet of the Santa Lucia) as it took him to travel from there to the Inlet of
the Ais. These distance intervals flatly contradict the standard geography
of the Ais and the Santa Lucia Indians, but fit rather nicely into our
geographic model. Background map courtesy of the Florida Department
of Environmental Protection (www.labins.org).

traveled by the survivors of the Puerto de Socorro garrison.
We believe it likely, however, that these infantrymen reckoned
their flight down the barrier island beach in terms of itinerant
land leagues, not nautical leagues, and that the extreme duress
they were under, especially malnutrition and the need to
forage, would have caused them to overestimate their progress
rather than underestimate it. Since itinerant land leagues are
"calculated" based upon the distance walked in one hours'time
(Chardon 1980:150-151), it is safe to assume that the general
recollection of 12 to 15 leagues is based upon a travel time of
12 to 15 hours. Admittedly, this travel time slightly exceeds
the estimated travel time of Jonathan Dickinson and his party
as they walked from St. Lucia to Ais (9.5 to 13.5 hours, see
below), especially if we add in the three leagues between Ais
to Puerto de Socorro, the likely starting point for the Spanish.
On the other hand, the immensity of the OIR Inlet as
described by De Brahm-1500 links (301.8 m) wide (1772 in
De Vorsey 1971:207)-corresponds to the Menendez accounts
of the garrison that fled from Puerto de Socorro being unable
to cross an inlet or a "rio" (Lyon n.d.). Dickinson was also
unable to cross the OIR Inlet on foot or by swimming. In
any case, the native people whom Dickinson refers to as "the
Indians of Sta. Lucea" were certainly located at the OIR Inlet
and not the St. Lucie River, the St. Lucie Inlet, or Jupiter inlet,
as shown below.

The Locations of Ais and Santa Lucia as Recorded by
Jonathan Dickinson in 1699

The orthodox view of Dickinson's journey from "Hoe-
bay" (Jobe) to "Jece" (Winter Ais) (Figure 15) is that his party
started at Jupiter Inlet and crossed two inlets (Gilbert's Bar and
the Old Indian River Inlet) before coming to the winter town
of Ais five to seven miles north of the OIR Inlet (Andrews
1943:40-41; Andrews and Andrews 1985; Bushnell 2007;
Davidsson 2004:33; Dickel 1992:43; Higgs 1942; Hutchinson
1987:16-18; McGoun 1993:10; Milanich 1995:56-59,66;
Rouse 1951:57, etc.). This interpretation ignores the fact
that Gilbert's Bar did not exist during the surveys of De
Brahm and Romans in the 1760s-1770s, nor during Vignoles'
reconnaissance of 1821-1822. It first appears during the 1830s
and 40s, when it was used as a shallow-draft escape route
by the pirate Don Pedro Gilbert, prior to which there was no
inlet between Jupiter and the OIR Inlet. Thus, any objection
to our geographic model based on a reluctance to assume the
existence of a now-closed inlet at the IRL Narrows encounters
the exact same problem embedded into the standard model,
except that the assumed inlet is situated at the future location
of Gilbert's Bar.
In fairness, it is likely that there were former inlets prior
to Gilbert's Bar in the region of the modern St. Lucie Inlet,

BRECH AND LANIHAM

LOCATION OF THE PARAMOUNT TOWN OF THE AIS

THE FLORIDA ANTHROPOLOGIsT 2010 VOL. 63(3-4~

given its general location across from the St. Lucie River. Nor
can we fault the standard model for assuming the existence of
an inlet in 1605-1696 which did not exist by 1765, since we
do the same in regard to the IRL Narrows. Rather, even when
we grant the possibility of an earlier inlet at Gilbert's Bar, the
standard model can shown to be wholly inconsistent with the
detailed time and distance estimates related by Dickinson, as
well as every other geographic reference in his narrative.
We have already shown how the orthodox view of the
location of Winter Ais at 81R16 does not conform to Mexia's
1605 reconnaissance report, nor with Menendez' estimate (and
others) that it was 15 leagues from Cape Canaveral to the Inlet
of the Ais. A close reading of the Dickinson account reveals it to
be also incompatible with the proposition that the town of Ais
was located some two leagues north of the OIR Inlet. In fact,
adherence to the traditional geographic model necessitates that
the adherent discount virtually all of Dickinson's locational
information: the second and final latitude reading of 27 45
min. for the Inlet of "Sta. Lucea" (Santa Lucia); his specific
recollections of the time and distance spent traveling between
Jobe, Santa Lucia and Winter Ais; his more general assessment
of the relative distances between these locations; the distance
estimate given by the Jobe chief; Dickinson's opinion that it
was "about a degree of latitude" between Jobe and Santa Lucia;
and the relative sizes of the two inlets Dickinson crossed.
Our interpretation of the location of Santa Lucia at the
OIR Inlet (latitude 27' 30.5 min.) and Winter Ais 10.5 km
north of the IRL Narrows imputes much less error to this
important primary source, essentially limited to a 14 minute
overestimation of latitude as they crossed the first inlet near
the Indian town of Santa Lucia (i.e., a 26 km error). The
traditional view-that the first inlet Dickinson crossed was
Gilbert's Bar located at or near the modern, artificial St. Lucie
Inlet (latitude 270 10 min.)-demands that Dickinson's second
latitude reading (27' 45 min.) was inaccurate by 35 minutes
of latitude, or 64.8 km, more than double the error imputed
by our geographic reconstruction. Even then, the traditional
view is inconsistent and selective regarding the accuracy of
Dickinson's latitude readings, since all interpretations accept
the first latitude reading at the site of their shipwreck (270
08 min.) as only slightly skewed to the north of actual (e.g.,
Milanich 1995:56-57).'7
Still, a margin of error of 14 minutes of latitude (26.8
km) is enough to invalidate any geographical reconstructions
based upon Dickinson's latitude readings. Fortunately, the
Dickinson text is surprisingly rich in time and mileage
estimates, especially as he travels on foot between the Indian
town of Santa Lucia and the town ofAis.'" These time/distance
estimates can be cumulated to derive the overall travel time
and also a range of possible distance covered, a method
already employed by Swanton (1922:390) for the distance
between Santa Lucia and Ais. Unfortunately, Dickinson often
switches from time estimates to mileage estimates at various
points along the journey. But even if one hesitates to attribute
sufficient precision to these specific details in Dickinson's
account so as to take such calculations credibly, at the very
least we can make a "greater than/less than" comparison of the

relative travel times or distances between the three locations of
Jobe, Santa Lucia, and Ais.
Under the orthodox view of Dickinson's itinery, the
distance/time Dickinson traveled from Jobe to the first inlet
(just below Santa Lucia) should be less than the distance from
that inlet to the second inlet, since Jupiter to St. Lucie Inlet
equals 25.6 km, while St. Lucie to the OIR Inlet equals 41.2
km, a 5:8 ratio. As shown in Table 3, however, the Dickinson
account reveals that the travel time from Jobe to the inlet at
Santa Lucia was about three times as long as the travel time
from Santa Lucia to the inlet below Ais. This ratio fits rather
well when we compare the distance from Jupiter Inlet to the
OIR Inlet (66.8 km) against the distance from the OIR Inlet to
the IRL Narrows (24.9 km), the site of our proposed Inlet of
the Ais. The result is a 2.7 to 1 ratio, very close to Dickinson's,
and certainly much more congruent than the 5:8 distance ratio
of the orthodox interpretation.
Dickinson also notes that the chief of Jobe considered the
distance between his village and Santa Lucia to be "two or
three days journey" (1699, in Andrews and Andrews 1985:10).
Clearly, the 25.6 km distance between Jupiter Inlet and the
modern St. Lucie Inlet (fonnerly Gilbert's Bar) does not
come close to matching this description, whereas the 66.8 km
distance from Jupiter to the OIR Inlet is a much better fit. The
fact that Dickinson's party made this journey in 25 to 37 hours
on foot, taking only three short breaks, corresponds well with
both the description provided by the Jobe chief and with the
actual point-to-point distances proposed by our interpretation
of the geographies of Santa Lucia and Jobe.
One indication ofthe accuracy of Dickinson's recollections
of distances traveled is that his 5-7 mile distance (8-11.3 km)
between the Inlet of the Ais and the town ofAis is very close to
Mexia's estimate of 2 nautical leagues made 91 years earlier.
This reflects well on Dickinson's overall accuracy especially
given that his levels of duress certainly increased after they
were seized by the Ais down at Ais Inlet (the Jobe and the
Santa Lucia had already warned them about the ferocity of the
Ais). This correspondence between Dickinson and Mexia also
indicates that the Inlet of the Ais probably did not migrate very
far during those 91 years. Dickinson also seems to indicate a
tighter proximity between the towns and the inlets for both
Jobe and Santa Lucia.
It should also be noted that whereas Dickinson's party
could not cross the first inlet without a boat since it was "a
furlong over" (201.2 meters), they were able to ford or swim the
second inlet with the help of their Indian guides. This indicates
that the first inlet (Santa Lucia) was much bigger and broader
than the second inlet (Ais). Such a configuration of inlets,
however, poses a problem for the standard interpretation of
Dickinson because it does not accord with historical accounts
of the relative sizes of Gilbert's Bar vs. the OIR Inlet. De
Brahm's Report (1772:207, in De Vorsey 1971) describes the
OIR Inlet as "1500 links wide," meaning 301.8 meters (each
link equals 7.92 inches, or 201.17 mm).
Unfortunately, Higgs'purported quotation from Dickinson
that the distance between the towns of Santa Lucia and Ais
was "about 20 miles" (32.2 km) (1942:34) does not appear in
Dickinson's text. Instead, Higgs was probably quoting Swanton

THE FLORIDA ANTHROPOLOGIST

2010 VOL. 63(3-4)

Blffcu AND LANHAM LOCATION OF THE PARAMOUNT TOWN OF THE Ais

Table 3. Time and distance intervals recorded by Dickson.

From Hoe-Bay (Jobe) to Santa Lucia:

BEGIN: morning of September 28, 1696
END: evening (before midnight) of September 29, 1696

Maximum total travel time:
Minimum total travel time:

6:00 AM to 10:00 PM the next day
9:00 AM to 6:00 PM the next day

MINUS three short breaks:
1) salvaged food off their shipwreck during late morning of first day; 2) "small respite" c. 6:00 PM on first day to
cook and eat fish; 3) stayed "a very short time" at unnamed Indian town on afternoon of second day.

Maximum estimated break time:
Minimum estimated break time:

8 hours
3 hours

Maximum estimated travel time:
Minimum estimated travel time:
Average:

37 hours
25 hours
31 hours

From St. Lucia to Inlet helow .lece Winter Ais):

1. Depart St. Lucia at 10:30 PM or 12:30 AM
2. Traveled 5 miles to next town
3. Stayed 2 hours at Indian town
4. Traveled "all night"
5. "After sunrise," came to 2nd shipwreck site
6. One mile later, came to Inlet of Ais
7. Took a siesta at inlet "as the day began to get warm"

Minimum travel time to Ais Inlet: 12:30 AM to 8:00 AM
Maximum travel time to Ais Inlet: 10:30 PM to 10:00 AM
Average travel time between St. Lucia and Ais Inlet:

From Ais Inlet to "Winter Town of Ais on Barrier Island:

7.5 hours
11.5 hours
9.5 hours

Captured by Ais, taken "4 or 5 miles" and meet Ais chief
Traveled with Ais chief for unknown duration, but probably not far
"One or two miles later" came to Ais town (c. afternoon)

Minimum distance from Ais Inlet to Ais Town:
Maximum distance from Ais Inlet to Ais Town:

5 miles (min. time = 2 hours)
7 miles (max. time = 3 hours)

(1922:390), who estimated that Dickinson traveled "about
20 miles" (32.2 km) between those two towns. And yet the
distance between the St. Lucie Inlet (Gilbert's Bar) and 8IR16
is 54.2 km, another example of the flagrant inconsistencies
within the orthodox geography. A more cautious reading
of Dickinson, however, allows for only a range of possible
values, but still excludes the orthodox geography: between 18
and 27 miles (30 to 43.5 km), with an estimated cumulative
travel time between 9.5 and 14.5 hours." Such a broad range
of values by it itself cannot be used to locate the town of Ais
with any precision, but it can bracket a general area in which

the town might be found. The point-to-point distance between
the OIR Inlet and the Pregnant Turtle site (8IR831) on the
barrier island opposite our proposed Ais location (8IR84) is
33 km (20.5 miles), well within this bracket, and very close to
Swanton's and Higg's estimate of 20 miles (32.2 km).
The point-to-point distance from the OIR Inlet to the
ocean beach across from the Blue Goose Midden (81R15,
Stanbridge's tentative candidate for Winter Ais) is 28.2 km,
while the distance to the area north of the modern, artificial
Sebastian Inlet (Lyon and Cato's Winter Ais) is 44.5 km. Both
are just slightly outside the distance bracket calculated above,

40 hours
33 hours

TOTAL:

BRECH AND LANHAM

LOCATION OF THE PARAMOUNT TOWN OF THE Als

THE FLORIDA ANTHROPOLOGIST 2010 VOL. 63(3-4)

but since the point-to-point, straight-line measurement of
distance is always smaller than the distance actually traveled,
this bracket does not exclude 8IR15 as much as it does Lyon
and Cato's proposed area for Winter Ais.
Dickinson notes that Winter Ais was located V2 mile (0.8
km) from the ocean beach, and that it was nearly surrounded
by mangroves, neatly concealed. Lyon's location of Winter
Ais along a thin section of barrier island (1976:129) is thus
contraindicated. Higgs (1942:38) described 8IR16 as easily
visible from the ocean, also ignoring or countermanding
this detail from Dickinson, although, in fairness, 8IR16 is
currently 0.44 km from the ocean beach, and was almost
certainly further from the shoreline back in 1696. Also, a large
island borders the lagoonal shoreline of 81R16. In fairness
to Stanbridge, it should be noted that modem maps of both
81R15 (the Blue Goose Midden) and our proposed tentative
candidate for Winter Ais (8IR831) opposite 8IR84 show
them to be surrounded by mangrove islands; both of these
proposed areas of the barrier island are equally thick measured
perpendicular to the shoreline-about 0.8 km (0.5 miles),
exactly as Dickinson wrote.
The orthodox interpretation of De-Brahm-Romans-
Lowery-Higgs-Rouse etc., wherein the Inlet of the Ais was
located at the OIR Inlet, can only maintain coherence with
Dickinson's account of his journey if one posits a different
unknown inlet in between Jupiter and the OIR Inlets other
than St. Lucie/Gilbert's Bar as being the Santa Lucia Inlet
that Dickinson crossed. Such a hypothetical inlet would have
to have been located closer to the OIR Inlet than to Jupiter
Inlet (unlike the modem St. Lucie Inlet), preferably by an
approximate 3-1 ratio in order to adhere to Dickinson's
account. Dividing the distance between OIR Inlet and Jupiter
Inlet (66.8 km) by 4 means that this hypothetical Santa Lucia
would have existed in the general area of a point roughly 16.7
km south of the OIR Inlet and 50.1 km north of Jupiter Inlet.2"
This would mean that the Jobe chief considered 50.1 km to be
"about two or three days' journey," which is unlikely, though
not impossible. It would also mean that Dickinson walked
only 16.7(+) km during his 7.5 11.5 hour journey from Santa
Lucia to the inlet below the town of Ais.
There are other serious problems with this hypothetical
scenario as well, including the large size of the OIR Inlet
recorded in early historic documents versus the diminutive
size of Ais Inlet back in 1696, Dickinson's estimate of "a
degree of latitude" (111.3 km) between Jobe and Santa Lucia,
and the fact that Dickinson traveled at least 32.5 (and as much
as 51.5) combined hours between Jobe and the Inlet of Ais.
Moreover, this hypothetical scenario of an unknown inlet
somewhere between the OIR Inlet and the modern St. Lucie
Inlet does nothing to reconcile the contradictory information
provided by Mexia and the accounts of Menendez' expedition
to Ais in 1565, as discussed in the preceding sections. Our
interpretation, on the other hand, integrates all the information
provided by Mexia, Dickinson, and Menendez regarding
the location of the paramount town of the Ais Indians in the
sixteenth and seventeenth centuries.

Conclusions

If the assumption of inaccuracy regarding culturally-
neutral distance estimates in two of the best eyewitness
sources of information concerning the location of the town of
Ais (Rouse 1951:55) is a greater assumption than assuming
that these primary sources are accurate, then our model of the
locations along Mexia's and Dickinson's journeys in 1605
and 1696 (respectively) is much more parsimonious than
the traditional conception. Both the traditional view and our
revision thereof presume the existence of an inlet that was not
present when Florida was surveyed in the 1760s nor again in
the 1820s. But the traditional view necessitates that not only
were Mexia's and Dickinson's specific distance estimates
completely inaccurate, even their general recollections of
distances were erroneous.
With Mexia, the standard model would have us believe
that he forgot or did not notice that it was 28 km (5 leagues)
from the Sebastian River to the town of Ais when he wrote that
it was only about 7.5 km (1.5 leagues). Our model imputes a
much smaller error to Mexia: the town ofAis was 6.9 km south
of the Sebastian River, or about 1.4 leagues. With Dickinson,
the standard model would have us believe that Dickinson
was completely mistaken in his recollection that the distance
between Jobe and the first inlet he crossed was three times
as great as the distance between the first inlet and the second
inlet. The standard model requires that the latter distance was
actually 60 percent larger than the first, whereas our model
proposes that the distance ratio was just under 3:1 (2.7 to 1).
In fact, the only major error we assume in all of the
first-hand sources cited so far is that De Brahm and Romans
incorrectly believed that the OIR Inlet was the Inlet of the Ais
because it was the only fully-open inlet that they encountered
in the IRL. Incorrect locations for the Indian towns of Ais
and Santa Lucia were then derived by Lowery and Higgs
and reaffirmed by Rouse and others based upon this incorrect
location of the Inlet of the Ais.
The salient points of this article can be summarized as
follows:
* Distance estimates from Menendez, Mexia, and Dickinson
all point to 8IR84 being the paramount town of the Ais
Indians in 1605 and 1696 to the exclusion of any other
mainland archaeological site. The Winter Town of Ais
was located at or near 8IR831 on the barrier island.
* Distance estimates and descriptions from Menendez,
Mexia, and Dickinson all point to the Indian River
Narrows being the location of the now-closed Inlet of the
Ais (Rio d'Ays and Barra d'Ays).
* The geomorphic landscape features of the barrier island
system enclosing the IRL also indicate that the Indian
River Narrows was the recent location of a former inlet.
* The first Spanish encampment of Puerto de Socorro
(1565) was located in the general area of the Bethel Creek
relict inlet.
* Both Mexia and Dickinson indicate a location for the
Spanish fort or the indigenous people known as Santa
Lucia that is much closer to Ais than either the St. Lucie
Inlet, the St. Lucie River, or Jupiter Inlet. The OIR Inlet

THE FLORIDA ANTHROPOLOGIST

2010 VOL. 63(3-4)

BRECH AND LANHAM LOCATION OF THE PARAMOUNT TOWN OF THE Ais

corresponds to Dickinson's recollection of the large
size of the inlet he crossed in the territory of the Santa
Lucia Indians, very close to their main village. It also
corresponds to his recollection of the distance from Jobe,
and to the distance to Ais Inlet if we posit an inlet at the
IRL Narrows as per #2 above.
S The source of much of the error and confusion in the
cartography and native political geography of the former
domain of the Ais derives from a single error made by De
Brahm and Romans during their survey in the 1760s and
early 1770s-the incorrect assumption that the OIR Inlet
was the Inlet of the Ais.
S "Rio d'Ays" (River of the Ais) refers to both a barrier
island inlet, now closed, which once existed at the Indian
River Narrows, and to the northern IRL as a whole.
"Barra d'Ayz" refers exclusively to the Inlet of the Ais,
not a thin section of barrier island. Similarly, "Rio Santa
Lucia" refers to both a barrier island inlet (the Old Indian
River Inlet) and to the southern half of the IRL. This same
conception of the IRL as being divided into roughly equal
halves at the Indian River Narrows was independently
re-invented by Romans and other Anglo-American
observers.
S Future research into the archaeology and Holocene
geology of the barrier islands of the IRL could profit from
shared information and overlapping research designs,
especially in the area that stretches from Sebastian to
Vero Beach. The north-to-south migration of the former
Inlet of the Ais from Sebastian to Indian River Shores
(opposite Winter Beach on the mainland) could be
dated and compared to the distribution of archaeological
remains in this area, none of which predate AD 800. The
area proposed here for the general vicinity of Winter Ais
could be tested for an overwash event dating from 1696.
Submerged landforms beneath the waters of the IRL and
Mosquito Lagoon could be tested for aerial exposure at
the time of Mexia's reconnaissance, specifically a small
island now eroded or covered over by spoil one league
(about 5 km) north of the Sebastian River, and perhaps a
larger island depicted by Mexia at the southeastern end of
the Mosquito Lagoon.

Notes

1. Romans' maps are not entirely free from major
inaccuracies, however, as shown in his errant depiction
of the southern tip of Merritt Island extending all the way
to the Sebastian River (Romans 1962:front piece [1775];
also in Burgess and Claussen 1982). As noted below,
many of the latitude readings given by Romans and De
Brahm are off by five to ten minutes.
2. Even then, De Brahm may have noted the vestiges of an
inlet in the general location proposed here for the Inlet of
the Ais. In fact, the persistent cartographic error of two
inlets north of the St. Lucie River almost certainly derive,
in part, from an ambiguous reference in De Brahm's The
Atlantic Pilot (1772:26, in De Vorsey 1974) to a "Spanish
Admiral Creek" at 27o. 40 min. This reference to a

"creek" was interpreted to be an inlet by Thomas Jefferys
(Royal Cartographer, Great Britain) or his publishers,
London mapmakers Robert Sayer and John Bennett,
who posthumously used Jefferys' name after the latter's
death in 1772. Thomas Jeffreys' 1775 and 1776 maps and
Samuel Lewis' 1817 map both show a Spanish Admiral
Inlet (or "Entree" in Lewis' map) in the general area
specified by De Brahm, from whose work they almost
certainly derived their information, with some distortion.
Other maps showing two inlets north of the area of the St.
Lucie Inlet usually leave the northernmost inlet unnamed.
Some place this unnamed inlet just south of Cape
Canaveral while others show it in the general vicinity of
the IRL Narrows and still others show it very close to the
OIR Inlet, separated by a very small island.
In any case, De Brahm's reference to "Spanish
Admiral Creek" cannot, unfortunately, be defined as a
semi-open, or vestigial inlet without some qualifications,
due to inconsistencies in De Brahm's body of work. Unlike
his Atlantic Pilot pamphlet which appeared in 1772, De
Brahm's map, dated 1769 and found in his Report of the
General Survey of the Southern District of North America
(1772, in De Vorsey 1972) places "Spanish Admiral
Rivulet" in the area of the Sebastian River on the mainland,
and not on the barrier island. No other cartographer or
explorer that we know of ever referred to the Sebastian
River by this name, but then again De Brahm was wont
to make up new names wherever he went in Florida. The
1769 map, however, also shows an anomalous indentation
in the oceanic coastline of the barrier island north of the
OIR Inlet, which De Brahm labels "Palmira Bay."
The name "Palmira" and the description provided by
De Brahm ("this bay has its name from the great number
of Palm [Cabbage] trees growing here") is very similar
to a landscape feature referred to by other geographers
and explorers (such as Romans, De Brahm's underling
on the East Florida survey project) as "El Palmar d'Ays"
or simply "El Palmar"-the Palm Groves of the Ais.
However, most references to El Palmar place it in the area
immediately south of the Sebastian River. In other words,
there seems to be some sort of switching or ambivalence
in De Brahm's choice of place names between "Spanish
Admiral" and "Palmar/Palmira" as they pertain to
the Sebastian River on the mainland and some sort of
landscape feature along the ocean shoreline in the area of
27. 40 min. It seems likely that De Brahm revised the data
from his 1772 pamphlet for his second, more extensive
report to the Crown (also in 1772 but presumably later),
just as he did for the latitude of the OIR Inlet-The
Atlantic Pilot has the OIR Inlet at 270 20 min. (as does
Romans 1962:274 [1775]), while De Brahm' report to the
Crown has it at the correct location of 27 30 min. 53
sec. It seems possible that De Brahm reassigned the name
Spanish Admiral to the Sebastian River on the mainland,
changing "creek" into "rivulet." In doing so, he removed
the name "Palmar" from the Sebastian River area (contra
Romans, etc) and placed it along the ocean shore of
the barrier island somewhere north of the OIR Inlet.

BRECH AND LANHAM

LOCATION OF THE PARAMOUNT TOWN OF THE AIS

THE FLORIDA ANTHROPOLOGIST 2010 VOL. 63(3-4~

There are two feasible resolutions of this ambiguity in De
Brahm's place names: 1) De Brahm's "Spanish Admiral
Creek" always meant the Sebastian River, and "Palmira
Bay" was just an added detail in De Brahm's Report that
he did not include as a coastal landmark in his pamphlet;
or 2) De Brahm noted the last vestiges of the Inlet of the
Ais during his 1765 survey without recognizing it as such,
calling it instead "Spanish Admiral Creek." Once closed,
sediment accumulation and colonization by vegetation
at the former Inlet of the Ais was sufficiently rapid for
the entire feature to go unnoticed by Romans a few years
later (1770-1771), perhaps leading De Brahm to revise his
Report.
A case could be made for either interpretation, but
obviously Thomas Jefferys or his posthumous publishers
thought it was the latter. The best evidence for the first
interpretation comes from the small scale map of the OIR
Inlet in De Brahms Report (in De Vorsey 1971:206-207)
which seems to indicate that Palmira Bay is just the broad
re-curved section of barrier island some 11 -14 km north
of the OIR Inlet.
Nevertheless, the latitude reading and the context of
De Brahm's initial use of the name "Spanish Admiral"
would seem to argue for a barrier island shoreline feature:
The Atlantic Pilot is a pamphlet describing how to sail
north from the Florida Keys into the Atlantic Ocean
proper. "Spanish Admiral Creek" is mentioned as the
last landmark south of Cape Canaveral to be spotted
while making this northward voyage. It was located, as
mentioned, at 270 40 min, the precise location of Bethel
Creek, but also just three minutes of latitude south of
the IRL Narrows, well within the margins of error of
De Brahm's other latitude readings in his pamphlet. The
Sebastian River, on the other hand, lies at latitude 270 51
m. 12 18 sec., a difference of at least 11 minutes. Romans
errs in his latitude of reading of the Sebastian River by
only 5 minutes of latitude (1962:xvii [1775]). Also, it
seems doubtful that the mainland Sebastian River would
make a good landmark for ocean voyagers, especially for
the northbound voyages anticipated by The Atlantic Pilot,
since it would require the observer to somehow look
over the barrier island in order to discern this mainland
river.
The point of the foregoing discussion is that although
the Inlet of the Ais was closed by the time of De Brahm's
and Roman's surveys, leading them to incorrectly assume
that it was located at the OIR Inlet (24.5 km to the
south-southeast), there are also indications that perhaps
a vestigial or even semi-open inlet might have existed
at the Indian River Narrows during De Brahm's earliest
reconnaissance in 1765.
3. We use "small scale" or "large scale" in the colloquial
sense, not in the strictly scientific sense of the terms, which
reverses the common understanding of their meaning. A
map in which 1 cm equals 1 meter is "larger scale" than a
map in which 1 cm equals 1 km, since .01 / 1 (i.e., 1/100)
is a larger fraction than .00001 / 1 (i.e., 1/100,000/0. Yet
it is natural to think of smaller landscape features, such

as the topography of a residential backyard (versus, say,
the topography of a entire county), as constituting smaller
scale features.
4. Dickel (1992:42) seems to attribute to Rouse some of
Higgs' opinions regarding Mexia and the Sebastian River
which a close reading suggests that Rouse (1951:170-171,
273 note 33) did not fully agree with.
5. The barrier island system (BIS) which encloses the IRL,
Banana River Lagoon and the Mosquito Lagoon is of a
different type than that found on the west coast of Florida,
or in the Georgia Bight further north, due to the relative
strength of waves vs. tide (Hayes 1979:1-24). Wave-
dominated BIS's produce a different configuration than
tide-dominated BIS's. The latter tend to be short and
stumpy, often resembling the "drumstick" part of chicken
legs (Hayes 1979:13-16). Barrier islands in wave-
dominated systems such as along the IRL (and coastal
North Carolina, for another example), however, produce
a long, thin, relatively continuous stretch of shore-parallel
island, broken here and there by inlet cuts, and rejoining
the mainland, often at headlands, hundreds of kilometers
apart from each other (Davis 1997:7).
6. The lagoonal shoreline of the IRL barrier island is mostly
cuspate from Sebastian (north of Spratt's Point) to Patrick
Air Force Base except for 1) a short, intriguing stretch
in the Floridana Beach area, from Ballard Cove north to
Snagg's Point; and 2) the Long Point relict inlet delta just
south of Ballard Cove.
7. We disagree with Sigler-Eisenberg and Russo's (1986:24-
25) interpretation of Almasi's geological study of the
IRL (1982:103) specifically as it pertains to the sequence
of inlets in the area of his Transect #4 (see Figure 7
for location). Almasi does not provide any date for the
prehistoric inlets that he detected in that area via sediment
analysis, only that these inlet-derived sediments were
deposited sometime after 2065 BP (+/- 195), and that it
represents[] the last 2,000 years." The ambivalence is
due to the fact that inlets scour away earlier Holocene
sediments, sometimes even scouring into Pleistocene
sub-strata (Niedoroda et al. 1985:363,365). Almasi's date
of 2065 BP is a terminus post quem, not an event date.
Furthermore, Almasi's Transect #4 is almost 3 km south
of the IRL Narrows.
8. After publishing the incorrect latitude reading for the OIR
Inlet in 1772 (The Atlantic Pilot), somehow De Brahm
published the correct latitude reading of 270 30 min.
53 sec. in his 1772 Report of the General Survey of the
Southern District of North America to the Crown (in De
Vorsey 1971:206). Romans, who excoriated De Brahm as
incompetent and insane, and claimed to have taught him
how to take astronomical latitude readings, nevertheless
published the wrong latitude reading for the OIR Inlet
(27 20 min.) three years later in 1775.
9. The difference here depends on where you begin and end
Mexia's measurement of the Mosquito Lagoon. Also,
Mexia's report and map indicate a large island at the
southern end of the Mosquito Lagoon, an island which no
longer exists. However, satellite photos of this part of the

THE FLORIDA ANTHROPOLOGIST

2010 VoL. 63(3-4)

BRECH AND LAMIANI LOCATION OF THE PARAMOL NT TO~~N OF THE Ais

Lagoon (via labins.org and Google Earth etc.) as well as
the 1887 coastal geodetic survey map seem to show some
submerged landfonns that somewhat resemble the shape
of the island as drawn by Mexia.
10. In fairness to Higgs (1942), it should be mentioned
that subtracting 2 km from the 61 km distance between
Pentoaya and 81R16 (Higgs' Winter Ais) still results in an
impossibly large league unit distance: 59 km divided by
7.5 leagues = 7.87 km per league.
11. Dickel (1992:209) mentions a Spanish olive jar sherd in
a local collection which may or may not have come from
81R15, the Blue Goose Midden.
12. Even then, the picture is not so simple. It should be
noted that Lyon's interpretation places Winter Ais at
least 30 kilometers further north than the orthodox view
of Lowery-Higgs-Rouse while at the same time Lyon's
interpretation of Santa Lucia places it 26 km south of
the orthodox view that Santa Lucia was at the St. Lucie
Inlet (or 28 kilometers when measured from the St. Lucie
River). That is, Lyon's view does not merely adjust the
existing locations and distance intervals further south or
further north; it expands the distance along both directions
between Ais and Santa Lucia, from 54 kilometers (the
distance between 81R16 and the St. Lucie Inlet/Gilbert's
Bar, as per the "orthodox" view) to 109 kilometers
measured point-to-point, more than doubling it. Our
interpretation, however, merely adjusts the locations of
Ais and Santa Lucia one inlet further north, while keeping
Jobe at Jupiter Inlet.
13. "Commanded" meaning direct control via occupation,
not indirect control via political arrangements or threat
of force. The Ais exercised political and economic
hegemony at least as far south as Jobe in 1696, as shown
in Dickinson's narrative.
14. As shown below, we question Dickinson's sighting of the
ruins across the IRL on the mainland, since, as mentioned,
we believe the logic of the situation demanded a barrier
island location for all ad hoc Spanish redoubts, including
Santa Lucia. Dickel (1992:43) also questions the identity
of the ruins sighted by Dickinson, but for different
reasons.
15. Fleshing out this hypothetical scenario (which we present
out of fairness to Lyon et al. to show how the Indians called
Santa Lucia could have been located further north in 1696
than the fort named Santa Lucia from 1565): the Mexia
map would indicate that this northward re-nucleation of
this native group took place sometime before 1605. We
might speculate further that it was sometime during this
40-year time span that the OIR Inlet first opened up. It is
also not farfetched to suppose that sustained initial contact
between the Spanish and the Indians from November 1565
until at least March 1566 (the Santa Lucia episode) led
to unrecorded outbreaks of disease and illness among the
native population of this area during the first few years of
this period. A 1569 letter from Menendez to the King (in
Kenny 1934:237) mentions the death of Brother Domingo
Baez from a "fever and ague" and an epidemic among
the Indians around St. Augustine. Hoffman refers to a

nine-year drought from 1562 to 1571 in northern Florida
(2002:59-62). If such conditions also applied to central
Florida, then it might make sense for the hegemonic Ais
Indians to tolerate or even welcome a relocation of the
Santa Lucia chiefdom much closer to their southernmost
habitations, and to a large newly-formed inlet (the OIR
Inlet) that the Ais could not fully exploit without relocating
themselves.
16. Vignoles' 1823 description of the western shore of the IRL
between the Narrows and opposite the bar (OIR Inlet) as
"covered by fine marshes half a mile wide beyond which
is a low hammock of rich growth" (1977:45 [1823]) would
seem to preclude any retreat along the western shoreline of
the IRL. Thus, mainland locations for Puerto de Socorro
and the other redoubts or rallying points would require
that the garrison fled south using either Indian trails or
overland through the brush, and thus quite vulnerable to
ambush.
17. Dickinson estimated that Jobe was 5 miles (8 km) south
of his shipwreck site, and walked it twice, yet the actual
distance from 270 8 min. to Jupiter Inlet is 13.9 miles
(22.3 km). Five miles (8 km) north of Jupiter Inlet is
latitude 270 0 min. 52 sec., meaning that Dickinson's first
latitude reading was off by 7 or 8 minutes according to
both the traditional interpretation and ours.
18. Dickinson accomplished the first leg of the journey, from
Jobe to St. Lucia, by boat out in the ocean, while the rest
of his party was on foot along the beach. Time estimates
can be derived based upon the land party. For instance, at
one point Dickinson and his boat mates row hard to avoid
an Indian sailing vessel, and end up outdistancing the
land party. They land the boat and wait two hours for the
remainder of their party catch up. Those two hours of rest
are thus not deducted from the total travel time estimated
for the journey to Santa Lucia.
19. Since the exact time of Dickinson's arrival at the town of
Ais is not as clearly indicated as his arrival at the Inlet of
the Ais, this is derived by converting the 5-7 mile distance
between Ais Inlet and the town of Ais into a two to three
hour time range, and adding this to the 7.5-9.5 hour time
estimate for Dickinson's travels from St. Lucia to the Inlet
of the Ais.
20. It just so happens that 17.5 km south of the OIR Inlet are
the distinct remains of a former inlet cut, now called Blind
Creek (latitude 27 21 min. 49 sec.) and Big Mud Creek,
the latter being the imprudent location of the St. Lucie
Nuclear Power Plant. The distance between Blind Creek
and 81R16 is 30 km (19 miles), very close to Swanton
and Higgs' estimate of 20 miles (32.2 km), and within our
bracket of 30 to 43.5 km. There are several small sites
just north of Blind Creek (Franklin and Mueller 2007;
Mueller 2008), but none seem especially likely to have
been the political center of the Santa Lucia Indians during
the First Spanish Period (1565-1763).
For what it's worth, Romans (1962:xvii) mentions a
possible inlet at 27 15 min., or 6 V2 minutes of latitude
(13.5 km) south of Blind Creek and Mud Creek. Vignoles
(1977:46) states that "the Gap" was located 16 miles (25.7

LOCATION OF THE PARAMOUNT TOWN OF THE AIS

BRECH AND LANHA.11

THE LORDA ATHRPOLGIST201 VOL 633-4

km) south of "the bar" (presumably the OIR Inlet), or
about 7.9 km south of Blind Creek, and this feature is also
shown in that general location on his 1823 map, albeit not
as an open inlet. Other early American maps also place
"the Gap" in this general area, but never depict it as an
open inlet. Williams (1962:43) also refers to a 16 mile
(25.7 km) distance viz. the location of the Gap, but his
language is ambiguous, and the ambiguity is complicated
by the fact that it is almost exactly 25.7 km from Jupiter
Inlet to the modern St. Lucie Inlet, fonnerly the location
of a natural inlet called Gilbert's Bar.
Romans' location is 35.8 km north of Jupiter Inlet,
while Vignoles' "the Gap" is 41.6 km north ofJupiter. Both
distances seem too short to correspond to Dickinson's and
the Jobe chief's estimations of the distance between Jobe
and Santa Lucia.

References Cited

Almasi, Mohammad M.
1985 The Importance of Inlets and Sea Level in the
Development of Hutchinson Island and the Indian
River Lagoon. In The Indian River Lagoon:
Proceedings of the Indian River Resources
Symposium, edited by Diane D. Barile. Marine
Resources Council, Palm Bay.
1983 Holocene Sediments and Evolution of the Indian Rivert
(Atlantic Coast of Florida). Doctoral dissertation,
University of Miami, Miami.

Anonymous
1998 Encyclopedia of Florida Indians Volume I: Tribes,
Nations and Peoples of the Woodland Areas A Z.
Somerset Publishers, Inc. St. Clair Shores, Michigan.

Andrews, Charles M.
1943 The Florida Indians in the Seventeenth Century.
Tequesta, 1:36-48.

Andrews, Charles M., and Evangeline Walker Andrews
1985 Jonathan Dickinson's Journal, or Gods Protecting
Providence, Being the Narrative of a Journey from
Port Royal in Jamaica to Philadelphia, August 23,
1696 to April 1, 1697; a True Stoir of Shipwreck
and Torture on the Florida Coast in 1696. Florida
Classics Library, Port Salerno.

Amade, Charles W
1996 Raids, Sieges, and International War. In The New
History of Florida, edited by Michael Gannon.
University Press of Florida, Gainesville.
1959 Florida On Trial: 1593 1602. University of Miami
Press, Miami.

Arrowsmith, Aaron
1804 Chart Of The West Indies And Spanish Dominions In
North America [Map]. Jones Smith & Co., London.

Electronic document, www.davidrumsey.com, last
accessed January 5, 2009.

Barile, Diane D. (editor)
1985 The Indian River Lagoon: Proceedings of the Indian
River Resources Symposium. Marine Resources
Council, Palm Bay.

Barrientos, Bartolome
1965 Pedro Menendez de Avilds, Founder of Florida.
Translated by A. Kerrigan. University of Florida
Press, Gainesville.

Blunt, Edmund March
1827 The Bahama Banks and Gulf of Florida [Map]. E.
& G.W. Blunt Inc., New York. Electronic document,
www.davidrumsey.com, last accessed Janury 5,
2009.

Bradford, Thomas G.
1835 Florida [Map]. Unknown publisher. Electronic
document,http: //alabamamaps.ua.edu/
historicalmaps/us_states/florida/index.html, last
accessed January 5, 2009.

Brech, Alan
2004 Neither Ocean nor Continent: Correlating the
Archaeology and Geomorphology of the Barrier
Islands of East Central Florida. Masters thesis,
Department of Anthropology, University of Florida,
Gainesville.

Brech, Alan and J. F. Lanham
2009 The Destruction of the Ais Chiefdom and Other
Overlooked Ethnographic and Ethnological
Information from Jonathan Dickinson's 1699
Shipwreck Narrative. Paper presented to the 107th
meeting of the Florida Historical Society, Pensacola.

Brown, Richard F. (editor)
2007 Coastal Encounters: The Transfbrmation of the
Gulf South in the Eighteenth Century. University of
Nebraska Press, Lincoln.

Brown-Peterson, Nancy and Ross W. Eames
1990 Ichthyofauna Associated with Spoil Islands, Indian
River Lagoon, Florida. Florida Scientist, 53:226-
238.

Brue, Adrien Hubert
1827 Carte des lies-Antilles ou Indes Occidentales, du
Golfe du Mexique et d'une parties des pays adjacent
[Map.] A. Brue, Paris. Electronic document, www.
davidrumsey.com, last accessed January 5, 2009.

THE FLORIDA ANTHROPOLOGIST

2010 VOL. 63(3-4)

LOCATION OF THE PARAMOUNT TOWN OF THE AIS

Burgess, Robert F. and Carl J. Claussen
1982 Florida Golden Galleons: the Search for the 1715
Spanish Treasure Fleet. Florida Classics Library,
Port Salerno.

Burr, D. H.
1839 Map of Florida Exhibiting the Post Offices, Post
Roads, Canals, Rail Roads, etc. John Arrowsmith,
London. Electronic document, www.davidrumsey.
cor, last accessed January 5, 2009.
1835 Map of the Territory of Florida. Unknown publisher.
Electronic document, http://alabamamaps.ua.edu/
historicalmaps/us_states/florida/index.html, last
accessed January 5, 2009.

Bushnell, Amy Turner
2007 Escape of the Nickaleers: European-Indian Relations
on the Wild Coast of Florida in 1696. In Coastal
Encounters: The Transformation of the Gulf South
in the Eighteenth Century, edited by R. F. Brown.
University of Nebraska Press, Lincoln.
1996 Republic of Spaniards, Republic of Indians. In The
New History of Florida, edited by Michael Gannon.
University Press of Florida, Gainesville.

Buzas, Martin A.
1995 Diversity, Distribution and Density Monitoring of
Foraminifera of the Indian River Lagoon (abstract
only). Bulletin of Marine Science, 57:279.

Cantley, Charles E., Koski, S., Langdale, J.B., Littman, S.B.
2002 Phase I Cultural Resources Survey of the IR-2
Dredged Material Management Area, Indian River
County, Florida. New South Associates, Stone
Mountain.

Carey, H. C., and I. Lea
1822 Geographical, Historical, And Statistical Map Of
Florida, No. 37. H.C. Carey & I. Lea, Philadelphia.
Electronic document, www.davidrumsey.com, last
accessed January 5, 2009.

Chardon, Ronald
1980 The Linear League in North America. Annals of the
Association ofAmerican Geographers, 70:129 153.

Chatelain, Verne E.
1941 The Defenses of Spanish Florida: 1565 to 1763.
Carnegie Institution of Washington Publication No.
511, Washington D.C.

Cline, Howard F.
1974 Florida Indians II: Provisional Historical Gazeteer,
with Locational Notes on Florida Colonial
Communities. Garland Publishing, New York.

Connor, Jeannette Thurber
1964 Translation of, and accompanying notes to Pedro
Menendez de Avilds Memorial by Solis de Meras.
University of Florida Press, Gainesville.
1925 Colonial Records ofSpanish Florida Volume 1: 1570-
1577. Florida Historical Society, Deland.

Cruchley, G. and Richard Holmes Laurie
1833 Fredonia Or The United States of North America
[Map.] G. Cruchley, London. Electronic document,
www.davidrumsey.com, last accessed January 5,
2009.

Davidsson, Robert I.
2004 Indian River: A History of the Ais Indians in Spanish
Florida. Florida Heritage Series, West Palm Beach.

Davis, Richard A. Jr.
1997 The Geology of Holocene Barrier Island Systems.
Springer-Verlag, New York.
1994 The Evolving Coast. Scientific American Library,
New York.

De Vorsey, Louis Jr.
1975 De Brahm's East Florida on the Eve of Revolution:
the Materials for its Recreation, in Eighteenth
Century Florida and Its Borderlands edited by S.
Proctor. University Presses of Florida, Gainesville.
1972 The Atlantic Pilot: A Facsimile Reproduction of the
1772 Edition, edited by L. De Vorsey. University
Presses of Florida, 1974.
1971 De Brahm Report of the General Survey of the
Southern District of North America. Edited by L.
De Vorsey. University of South Carolina Press,
Columbia.

Degan, Kathleen A.
1987 Artifacts of the Spanish Colonies of Florida and the
Caribbean, 1500-1800: Ceramics, Glassware, and
Beads. Smithsonian Institution Press, Washington D.
C.

Dickel, David N.
1992 An Archaeological Survey of Indian River County,
Florida. The Archaeological and Historical
Conservancy, Technical Report #55, Miami.

Dickinson, Jonathan
1787[1699] Gods Protecting Providence, Man S Surest Help
and Defense in the Times of the Greatest Difficulty
and Most Imminent Dangeri Evidenced in the
Remarkable Deliverance of Divers Person from the
Devouring Waves of the Sea, Amongst Which They
Suffered Si,,/'In ..t and also fiom the More Cruelly
Devouring Jaws of the Inhumane Cannibals of
Florida, Faithfidly Related by One of the Persons
Concerned Therein, Jonathan Dickinson. Electronic

BRECH AND LANHAM

THE FLORIDA ANTHROPOLOGIST 2010 VOL. 63(3-4's

version of 1787 edition available at: http://palmnm.
fcla.edu/fh/, last accessed January 3, 2009.

Fielding, Lucas Jr.
1823 Florida [Map]. Lucas Fielding Jr., Baltimore.
Electronic document, www.davidrumsey.com, last
accessed January 5, 2009.

Fisher, John J. and Elizabeth J. Simpson
1979 Washover and Tidal Sedimentation Rates as
Environmental Factors in Development of a
Transgressive Barrier Shoreline. In Barrier Islands.
From the Gulf of St. Lawrence to the Gulf of Mexico,
edited by S. P. Leatherman, pp. 127-148. Academic
Press, New York.

Florida Fish and Wildlife Commission
2002 Spoil Islands of the Indian River Lagoon, Brevard
County. Florida Inland Navigation District, Jupiter.

Franklin, Ryan and Bradley M. Mueller
2007 A Phase I Cultural Resources Assessment Survey
of the Blind Creek East Parcel, St. Lucie County,
Florida. Report prepared for the St. Lucie County
Mosquito Control District, by the Archaeological and
Historical Conservancy, Miami. FLMSF Report No.
14038.

Furlong, Capt. Lawrence
1796 The American Coast Pilot. Blunt and March,
Newburyport.

Geiger, Maynard
1937 The Franciscan Conquest of Florida (1573-1618).
The Catholic University of America, Washington,
D.C.

Gilmore, R. Grant Jr.
1995 Environmental and Biogeographic Factors
Influencing the Ichthyofaunal Diversity of the Indian
River Lagoon, Florida. Bulletin of Marine Science,
57:28-36.

Greenleaf, Jeremiah
1840 Map of the Territory of Florida. G.R. French,
Battleboro. Electronic document, www.davidrumsey.
com, last accessed January 5, 2009.

Griffin, John W. and James J. Miller
1978 Merritt Island National Wildlife Refuge: Cultural
Resource Assessment. Report prepared for the U.S.
Fish and Wildlife Service by Cultural Resource
Management Inc., Tallahassee.

Grigg, John
1830 North & South Carolina, Georgia & Florida [Map].
J. H. Young, Philadelphia. Electronic document,

www.davidrumsey.com, last accessed January 5,
2009.

Handley, Brent M.
2001 The Blue Goose Midden (8IR15): A Malabar II
Occupation on the Indian River Lagoon. The Florida
Anthropologist, 54:103-121.

Handley, Brent M. and Marsha A. Chance
2000 Addendum to "An Intensive Cultural Resource
Assessment Survey of the Proposed K. Hovnanian
Companies Government Lot 3 Development
Project Located in Section 26, Township 31 South,
Range 39 East, Indian River County, Florida by
Storm L. Richards and Jeanne Fillman-Richards."
Environmental Services Inc., Jacksonville.

Hann, John H.
2003 Indians of Central and South Florida. Florida
Classics Library, Port Salerno.
1996 A History of the Tinmucua Indians and Missions.
University Press of Florida, Gainesville.
1995 Demise of the Pojoy and Bomto. Florida Historical
Quarterly v. 74:184-200.
1991 Missions to the Calusa. University Press of Florida,
Gainesville.

Hayes, Miles O.
1979 Barrier Island Morphology as a Function of Tidal and
Wave Regime. In Barrier Islands: From the Gulf of
St. Lawrence to the Gulfof'Mexico, edited by Stephen
P. Leatherman, pp. 1 27. Academic Press, New
York.

Held, Ray E.
1949 Hernando de Miranda. Florida Historical Quarterly,
28:111-130.

Higgs, Charles
1951 The Derrotero of Alvaro Mexia, 1605. In Rouse, A
Survey of Indian River Archeology, Yale University
Press, New Haven.
1942 Spanish Contacts with the Ais (Indian River) Country.
Florida Historical Quarterly, 21:25-39.

Hinton, Isaac Taylor
1832 Map of the state of Florida. Thomas Wardle,
Philadelphia. Electronic document, www.
davidrumsey.com, last accessed January 5, 2009.

Hoffman, Paul E.
2002 Florida s Frontier Indiana University Press,
Bloomington.

Hutchinson, Janet
1987 History of Martin County. Florida Classics Library,
Port Salerno.

THE FLORIDA ANTHROPOLOGIST

2010 VOL. 63(3-4)

BREd AND LANHAM LOCATION OF THE PARAMOUNT TOWN OF THE Ais

Jeffreys, Thomas
1775 The Peninsula and Gulf of Florida or Channel of
Bahama with the Bahama Islands [Map]. Mimento
Art, South Kent, Connecticut.
1776 The Coast Of West Florida and Louisiana [Map].
Sayer and Bennett, London. Electronic document,
www.davidrumsey.com, last accessed January 5,
2009.

Johnson, Robert E., and B. Alan Basinet
2002 Archeological Investigations of a Portion of the
Windsor Development Tract, with Limited Phase
II Testing of 81r831 (Partial), Indian River County,
Florida. Florida Archeological Services, Inc.,
Jacksonville.
2000 A Cultural Resources Assessment Survey of a Portion
of the Florida Inland Navigation District Duck
Point Tract, Indian River County, Florida. Florida
Archeological Services, Inc., Jacksonville.

Johnson, Robert E., B. Alan Basinet, and Arlene Fradkin
2000 A Phase II Investigation of a Portion oflrl5, the Blue
Goose Midden, Sea Oaks Plantation Development
Tract, Indian River County Florida. Florida
Archeological Services, Jacksonville.

Johnston, Sidney P.
2000 A History of Indian River County. Indian River
County Historical Society, Vero Beach.

Kerrigan, Anthony (translator)
1951 Barcias ( 1/;, -;., al. History of the Continent of0
Florida. University of Florida Press, Gainesville.

Lanham, J. F., and Alan Brech
2007 Summer Pentoaya: Locating a Prominent Ais Indian
Town Along the Indian River Lagoon, Florida. The
Florida Anthropologist, 60:21-38.

Leatherman, Stephen P. (editor)
1979 Barrier Islands: From the Gulf ofSt. Lawrence to the
Gulf of Mexico. Academic Press, New York.

Lewis, Samuel
1817 A Correct Map of the United States, With The West
Indies, fi-om the best Authorities. T.L. Plowman,
Philadelphia. Electronic document, www.
davidrumsey.com, last accessed January 5, 2009.

Lockwood, Charlotte
1975 Florida Historic Indian River County. Indian River
County Bicentennial Committee, Vero Beach.

Lowery, Woodbury
1905 The Spanish Settlements Within the Present Limits of
the United States: Florida 1562-1574. G. P. Putnam
and Sons, New York.

Lyon, Eugene
1996 Settlement and Survival. In The New History of
Florida, edited by Michael Gannon. University Press
of Florida, Gainesville.
1992 Lecture on the Ais Indians at the McClarty Museum.
Tape recording on file at the Indian River County
Library, Vero Beach.
1989 Pedro Menendez' Plan for Settling La Florida. In First
Encounters: Spanish Explorations in the Caribbean
and the United States, 1492 1570, edited by Jerald
T. Milanich and Susan Milbrath. University Presses
of Florida, Gainesville.
1976 The Enterprise of'Florida: Pedro Menendez de Aviles
and the Spanish Conquest of 1565-1568. University
Presses of Florida, Gainesville.
n.d. Additional Material on the Origin of the Name "Santa
Lucia." Unpublished manuscript on file at the Indian
River County Library, Vero Beach.
n.d. More Light on the Ays Indians. Unpublished
manuscript on file at the Indian River County Library,
Vero Beach.

McGoun, William E.
1993 Prehistoric Peoples of South Florida. University of
Alabama Press, Tuscaloosa.

McMichael, Alan E.
1977 A Model for Barrier Island Settlement Pattern. The
Florida Anthropologist, 30:179 195.

Matter, Robert Allen
1990 Pre-Seminole Florida: Spanish Soldiers, Friars, and
Indian Missions. 1513-1763. Garland Publishing,
New York.

Mayhew, Troy Alan
1996 Late Holocene Geologic Evolution of a Barrier
Island Complex, East-Central Florida. Masters
Thesis, Florida Institute of Technology, Melbourne.

Melish, John
1816 Map of the United States with the contiguous British
& Spanish Possessions Compiled fi-om the latest
& best Authorities. John Melish, Philadelphia.
Electronic document, www.davidrumsey.com, last
accessed January 5, 2009.
1815 Southern Section of the United States including
Florida etc. [Map]. John Melish, Philadelphia.
Electronic document, www.davidrumsey.com, last
accessed January 5, 2009.

Mexia, Alvaro
1605 Deroterro. In A Survey of Indian River Archeology,
edited by Irving Rouse. Yale University Press, New
Haven.

BRECH AND LANHAM

LOCATION OF THE PARAMOUNT TOWN OF THE AIS

THE LORDA ATHRPOLGIST201 VOL 633-4

Mikkelsen, Paula M., Paul S. Mikkelsen, and David J. Karlan
1995 Molluscan Biodiversity in the Indian River Lagoon,
Florida. Bulletin of Marine Science, 57:94-127.

Milanich, Jerald T.
1998 Florida Indians from Ancient Times to the Present.
University Press of Florida, Gainesville.
1995 Florida Indians and the Invasion from Europe.
University Press of Florida, Gainesville.

Milanich, Jerald T. and Charles Hudson
1993 Hernan de Soto and the Indians ofFlorida. University
Press of Florida, Gainesville.

Milanich, Jerald T. and Susan Milbrath (editors)
1989 First Encounters: Spanish Explorations in the
Caribbean and the United States, 1492 1570.
University Presses of Florida, Gainesville.

Mitchell, Samuel Augustus
1890 County map of Florida. John Y. Huber & Co.
Philadelphia. Electronic document, www.
davidrumsey.com, last accessed January 5, 2009.

Morse, Jedidiah
1794 A Map of East and West Florida. John Stockdale,
London. Electronic document, www.davidrumsey.
cor, last accessed January 5, 2009.

Mueller, Bradley M.
2008 A Reconnaissance Survey and Monitoring of the
Blind Creek West Parcel, St. Lucie County, Florida.
Report prepared for the St. Lucie County Mosquito
Control District by the Archaeological and Historical
Conservancy, Miami. FLMSF Report No. 15623.

Niedoroda, Alan William, with Donald J. P. Swift and Alberto
G. Figueiredo Jr. and George L. Freeland
1985 Barrier Island Evolution, Middle Atlantic Shelf,
U.S.A., Part II: Evidence from the Shelf Floor.
Marine Geology 63:363-396.

Parry, John H.
1963 The Age of Reconnaissance: Discovery, Exploration
and Settlement, 1450 1650. World Publishing Co.,
New York.

Pepe, James
2000 An Archaeological Survey of St. Lucie County,
Florida. Technical Report #280, The Archaeological
and Historical Conservancy, Miami.

Pilkey, Orrin H.
1998 The North Carolina Shore and Its Barrier Islands:
Restless Ribbons of Sand. Duke University Press,
Durham.

Pilkey, Orrin H. and Katharine L. Dixon
1998 The Corps and the Shore. Island Press, Washington
D.C.

Pilkey, Orrin H. and Mary Edna Fraser
2003 A Celebration of the Worlds Barrier Islands.
Columbia University Press, New York.

Proctor, Samuel (editor)
1975 Eighteenth Century Florida and its Borderlands.
University Presses of Florida, Gainesville.

Randazzo, Anthony F. and Douglas S. Jones (editors)
1997 The Geology of Florida. University Press of Florida,
Gainesville.

Romans, Bernard
1962 [1775] A Concise Natural History of East and West
Florida. University of Florida Press, Gainesville.

Rouse, Irving
1951 A Survey of Indian River Archeology. Yale University
Press, New Haven.

Rudolph, Teresa Perry
1981 Cultural Resources Survey: Pelican Island National
Wildlife Refuge, Florida. Southeastern Wildlife
Services, Inc. Athens, GA.

Schmalzer, Paul A.
1995 Biodiversity of Saline and Brackish Marshes of
the Indian River Lagoon: Historic and Current
Perspectives. Bulletin of Marine Science, 57:75-78.

Schmidt, Walter
1997 Geomorphology and Physiography of Florida. In The
Geology of Florida, edited by A.F. Randazzo and
D.S. Jones, pp. 1-12. University Press of Florida,
Gainesville.

Sigler-Eisenberg, Brenda and Michael Russo
1986 Seasonality and Function of Small Sites on Florida's
Central-East Coast. Southeastern Archaeology, 5:21-
31.

Solis de Meras, Gonzalo
1964 Pedro Menendez de Avilds: Memorial. Translated by
J. T. Connor. University of Florida Press, Gainesville.

Stahl, Jeremy D.
1986 An Ethnohistory of South Florida 1500 1575.
Masters Thesis, University of Florida, Gainesville.

Stauble, Donald K.
1988 The Geomorphology, Geologic History, Sediments
and Inlet Formation of the Indian River Lagoon
System. In The Indian River Lagoon: Proceedings

THE FLORIDA ANTHROPOLOGIST

2010 VOL. 63(3-4)

LOCATION OF THE PARAMOUNT TOWN OF THE AIS

of the Indian River Resources Symposium, edited by
Diane D. Barile, Marine Resources Council, Palm
Bay.

Swanton, John Reed
1922 EarlY History of the Creek Indians and their
Neighbors. Bulletin #73, Smithsonian Institution,
Bureau of American Ethnology, Washington D.C.

Tanner, Henry S.
1845 Florida (with) Plan of Tallahassee (with) Plan
of Pensacola [Map]. Tanner's Geographical
Establishment, New York. Electronic document,
www.davidrumsey.com, last accessed January 5,
2009.
1825 Map of Florida. Henry Tanner, Philadelphia.
Electronic document, www.davidrumsey.com, last
accessed January 5, 2009.
1823 A Map of North America, Constructed According to
the Latest Information. H.S. Tanner, Philadelphia.
Electronic document, www.davidrumsey.com, last
accessed January 5, 2009.

United States Coast and Geodetic Survey
1887 Coast Chart #162: From Cape Canaveral Southward
to Latitude 27 degrees 41 minutes, Florida
[Map]. Washington, D.C. Electronic document,
http://alabamamaps.ua.edu/historicalmaps/
navigationcharts/index.html, last accessed January 5,
2009.

Van Landingham, Kyle S.
1988 Pictorial History of St. Lucie County 1565-1910. St.
Lucie Historical Society.

Vignoles, Charles
1977 [1823] Observations Upon the Floridas. University
Presses of Florida, Gainesville.

Von Humboldt, Alexander
1811 Carte Du Mexique et des Pays Limitrophes Situes Au
Nord et a 'est Dressee d'apres la Grande Carte de la
Nouvelle Espagne [Map]. F. Schoell Paris. Electronic

document, www.davidrumsey.com, last accessed
January 5, 2009.

Weber, David J
1994 The Spanish Frontier in North America. Yale
University Press, New Haven.

Wenhold, Lucy L.
1936 A 17th Century Letter of Gabriel Diaz Vara Calder6n,
Bishop of Cuba, Describing the Indians and Indian
Missions of Florida. Smithsonian Miscellaneous
Collections, vol. 95, no. 16 Washington, DC.

Westfahl, Arline and George Keyes
2003 One Person Can Make a Difference: A Story of Paul
Kroegel and Pelican Island. National Wildlife Refuge
System, Washington, D.C. Electronic document,
http://pelicanislandfriends.org/centennialbrochure-
index.htm, last accessed June 9, 2010.

Wheeler, Ryan J.
1992 The Riviera Complex: An East Okeechobee
Archaeological Area Settlement. The Florida
Anthropologist 45:5-17.

Wheeler, Ryan J. and James Pepe
2002 The Jobe and Jeaga of the Palm Beach County Area.
The Florida Anthropologist, 55:221-241.

Williams, John Lee
1962 The Territory of Florida, or Sketches of the
Topography, Civil and Natural History of the
Country, the Climate and the Indian Tribes, from the
First Discovery to the Present Time. University of
Florida Press, Gainesville.

Zimmerman, Vera
1996 Archaeological Survey of King's Island, St. Lucie
County, Florida. Indian River Anthropological
Society, Cocoa.

BRECH AND LANHAM

Chapters of the Florida Anthropological Society

10 5

1. Ancient Ones Archaeological Society of North Central Florida
2902 NW 104"' Court, Unit A, Gainesville, FL 32606

2. Archaeological Society of Southern Florida
2495 N.W. 35th Ave.. Miami, FL 33142

3. Central Florida Anthropological Society
P.O. Box 947544, Maitland, FL 32794 3

4. Central Gulf Coast Archaeological Society
P.O. Box 1563, Pinellas Park, FL 33780

5. Emerald Coast Archaeological Society
c/o Indian Temple Mound Museum 4
139 Miracle Strip Pkwy SE, Fort Walton Beach, 32548

6. Gold Coast Anthropological Society
PO Box 11052, Fort Lauderdale, FL 33339

7. Indian River Anthropological Society 14
3705 S. Tropical Trail, Merritt Island, FL 32952

8. Kissimmee Valley Archaeological and Historical Conservancy
195 Huntley Oaks Blvd., Lake Placid, FL 33852 16

9. Panhandle Archaeological Society at Tallahassee
P.O. Box 20026, Tallahassee, FL 32316

10. Pensacola Archaeological Society
P.O. Box 13251, Pensacola, FL 32591 1

I1. St. Augustine Archaeological Association
P.O. Box 1301, St. Augustine, FL 32085

12. Southeast Florida Archaeological Society
P.O Box 2875, Stuart, FL 34995 .,

13. Southwest Florida Archaeological Society-
P.O. Box 9965, Naples, FL 34101 .

14. Time Sifters Archaeology Society
P.O. Box 25642, Sarasota, FL 34277

15. Volusia Anthropological Society
P.O. Box 1881, Ormond Beach, FL 32175

16. Warm Mineral Springs/Little Salt Spring Archaeological Society
P.O. Box 7797. North Port. FL 34287

GOTIER HAMMOCK MOUND AND MIDDEN
ON ST. JOSEPH BAY, NORTHWEST FLORIDA

NANCY MARIE WHITE

Department ofA',.id .. University ofSouth Florida, Tampa, FL 33620
E-mail: nmw@usf.edu

Project and Site Background

The Gotier Hammock site (8GU2) is a burial mound
and midden on the southeast shore of St. Joseph Bay in Gulf
County, northwest Florida (Figure 1), recently relocated and
investigated by University of South Florida (USF) field teams.
The mound is mostly destroyed but retained cultural deposits
radiocarbon-dated to A.D. 650. The associated midden
yielded Middle Woodland and later ceramics and radiocarbon
dates in the fourteenth and sixteenth centuries. This research
documents an unusual prehistoric site for the St. Joe Bay
region and provides some new data on Middle Woodland
mound ceremonialism.

Early History of the Site

Diggers. C. B. Moore (1902:210-11; Brose and White
1999:212-13) recorded the "mound in Gotier Hammock" a
century ago. He placed it 800 m northeast of Conch Island
(itself a prehistoric shell midden, 8GU20, a quarter-mile
offshore) and 800 m inland (Figure 2). The mound was "a
truncated cone of dark sand" 1.5 m high and 8 m in diameter,
but already plagued by looters when Moore arrived. He said
it was "famous for successful relic searches" and had been
"practically dug to pieces, one relic hunter or treasure seeker
filling the hole made by another." The several flexed or bundle
burials he uncovered were scattered around the mound, with
some in shallow pits below the mound base. Nearly a half-
century later, Willey (1949:253) noted that any intact ceramic
mortuary deposits) had largely been removed by Moore's
time, though Moore still obtained Weeden Island and Swift
Creek pottery.
One of the pre-Moore diggers in the mound was named
Floyd. According to Willey (1949:28, 256-57 [Willey's
footnote misspells the name as Lloyd, but later corrects
it]; Jones 2002:4), in 1893-94, C. H. B. Floyd sent to the
Smithsonian National Museum of Natural History (NMNH)
his collection of pottery, stone celts, and shell artifacts from a
mound 25 miles from the town of Apalachicola on St. Joseph
Bay. There is no other mound anywhere near this location, so
Willey was correct in calling it Gotier Hammock. But he gave
the collection another site number, "GU-6" (now corrected in
the Site File).

Willey (1949:256) cites discussion and illustrations of the
Floyd collection in William Henry Holmes's classic book on
eastern U.S. aboriginal pottery. Holmes (1903:111-112, Plates
LXXVIII and LXXVIII A) also misspelled Floyd's name, but
illustrated 10 finely-made Weeden Island Plain and Incised and
late-variety Swift Creek Complicated-Stamped pots. Museum
records show that these artifacts were actually recovered in the
fall of 1892 by S. A. Floyd and sold to the NMNH (accession
number 027333) a year or two later, for $25, by his son C. H.
B. Floyd, who was at the time 18 and in school in Savannah,
Georgia. Samuel Augustus Floyd was a Confederate veteran
who came to northwest Florida from Savannah, worked
in the timber industry and was elected representative to the
Florida House in 1877 and Franklin County sheriff in 1883.
Son Charles Henry Bourke Floyd (Harry or Harvey), "during
his school days......was somewhat erratic but very brilliant"
(Mathews 1998:69: Shores 2008:59); perhaps he sold the
collection to the Smithsonian for school money ($25 in 1894
was worth between $500 and $2000 today). He was later a
lawyer, tax assessor, justice of the peace in Apalachicola, and
also a state legislator. Both Floyds are buried in the Magnolia
Cemetery in Apalachicola (where Moore documented other
famous mounds!).
Moore claimed that he "completely demolished" Gotier
Hammock mound. But, judging from the status of other sites
he claimed to have dug completely in his many northwest
Florida travels, I assumed he left something. The site was not
relocated by Willey (1949:253-4), nor Florida State University
(FSU) archaeologists working in the panhandle in the 1960s
and '70s, perhaps because it was heavily forested.
Homesteaders. There is published information on the
historic use of the hammock area on which the mound sat,
as well as some papers in the Gulf County Library in Port St.
Joe. In the early nineteenth century, it was inhabited by the
Gautier family, descended from French and English migrants
who came to America around 1790 and lived in Georgia, then
northwest Florida. Peter Gautier had settled somewhere in the
area before 1827, when there is a record of his son Thomas's
birth (Gulf County Golden Anniversary Commission
1975:16). Peter William Gautier, Jr., another son, had owned a
hotel in Marianna, then another one in Apalachicola. In 1836
he migrated to the new boom town of St. Joseph, which was
just being founded, and apparently some time around then

VOL. 63(3-4) TI-mE FLORIDA ANTI-IROPOLOGISI SEPTEMBER DECEMBER 2010

VOL. 63(3-4)

THE FLORIDA ANTHROPOLOGIST

SEPTEMBER DECEMBER 2010

THE LORDA ATHRPOLGIST201 VOL 633-4

GULF COUNTY

St. Joseph

Bay

Gotier

Hammock

Figure 1. Location of Gotier Hammock and St. Joseph Bay in Gulf County, Florida.

built a home some four miles outside town on the hammock
that became named after him. He was active in politics, and
was the publisher of the St. Joseph Telegraph, which later
in 1836 became the St. Joseph Times until it ended in 1841.
He backed the successful attempt to hold the first Florida
Constitutional Convention in St. Joseph (now commemorated
in the Constitutional Convention Museum in Port St. Joe).
In 1841 he was elected Speaker of the Territorial House of
Representatives.
Though no record is known of what the Gautiers' plantation
home looked like, a historic note said to be based on hearsay
described an old mansion surrounded by majestic oaks. When
the 1841 yellow fever epidemic devastated St. Joseph, Peter
Gautier and his family fled to Texas (Gulf County Golden
Anniversary Commission 1975:16; Porter 1975:33-35). Many
others who had once enjoyed the good life in old St. Joseph
either died of the disease or fled, and the remainder of the town
was destroyed by succeeding hurricanes. Today's city of Port
St. Joe was founded decades later on roughly the same spot
(leaving old St. Joseph as an archaeological site itself yet to be
explored). Gotier Hammock was apparently unoccupied for
about a half-century.

Then brothers John and Dave Maddox and their family
came from Apalachicola in 1893. They built a house next to
the Gautier home ruins, which were apparently still standing,
and grazed cattle on the coastal grasses. According to local oral
history, they moved away afterjust a few years because of both
mosquitoes and better opportunities elsewhere (Gulf County
Golden Anniversary Commission 1975:16; Jones 2002:4). If
the reported timing is correct, perhaps Floyd's collection of
pots from the mound came just before the Maddoxes arrived or
at the time when their building may have disturbed the ground
and exposed artifacts. After the Maddoxes left, the land was
used for hunting, possibly cattle-grazing, and planted pine.
Some people in the area apparently have known about the
mound over the years, sometimes collecting artifacts. Local
historian and avocational archaeologist Herman Jones (2002)
wrote a newspaper article about this site and others Moore
visited in the region.
This hammock, a formation of higher ground amid the
bay shore lowlands, still bore the Gautier family name over 60
years after they left, when Moore recorded it, which must have
been after the Maddoxes left as well. Since Moore, a well-
educated man, misspelled the name, he may have had only

THE FLORIDA ANTHROPOLOGIST

2010 VOL. 63(3-4)

Figure 2. Location of Gotier Hammock mound and shell midden on aerial photo adapted from Google Earth.

local oral history and pronunciation ("Go-teer") to identify the
place (Jones 2002:1). Today there are no standing structures,
only diverse bricks, probably from both historic families'
buildings, littering the surface.

Recent Histoln'

For the last several decades, the land has been owned
by the St. Joe Paper Company, famous in Gulf County for
its timberlands and paper mill in Port St. Joe. They planted
pine there, creating the high furrows around the mound that
are characteristic of modern methods of pine plantation in low
wetlands (thanks to Neal Land and Timber Company director
Phil McMillan of Blountstown for explaining this to me). St.
Joe, the largest private landowner in Florida, is responsible
for remaking the landscape of much of the panhandle (Ziewitz
and Wiaz 2004:66), but they did not plant trees on the small
elevated hammock itself, leaving the old oaks. Possibly the
road that pushed through the mound was first made by the
Gautiers and over time cut deeper into the ground, expanding
disturbance to the mound. (the old highway from Apalachicola
to Port St. Joe ran nearby). St. Joe Paper Company is now
renamed the St. Joe Company, having closed the paper plant
and moved toward housing and other land "development"
(Ziewitz and Wiaz 2004; Herring 2009). But they still maintain
pine plantation around Gotier Hammock.

In 2001, after years of drought and a fire, artifacts
exposed in a firebreak plowed at this site were reported by a
St. Joe company official to Apalachicola National Estuarine
Research Reserve (ANERR) personnel, who brought me to the
firebreak (midden) area. Pottery and shell tools were exposed
on the ground surface, showing the site was near where
Moore had said, and that it covered a large area from close
to the paved highway eastward. From 2002 through 2004, I
conducted archaeological survey of the St. Joseph Bay State
Buffer Preserve lands (White 2005), immediately south and
north of Gotier Hammock. and became concerned about the
mound, one of the very few known in Gulf County. USF's
research program in the area always includes public outreach
with "archaeology day" programs to obtain/share data with
avocationals and other interested folks. In October 2003, a
local resident contacted through these programs took me and
my crew into what he thought was the actual mound, on the
higher ground of the hammock, southeast of the midden. A dirt
track appeared to have bisected the mound and exposed a few
artifacts. This collector had recovered a small piece of mica
here four decades earlier.
To determine if anything was left intact, I planned a
formal investigation. The St. Joe Company gave permission in
May 2008 for test excavations, and the St. Joseph State Buffer
Preserve and its Friends support group provided assistance.
Fieldwork and test excavation were conducted by the UiSF

THE FLORIDA ANTHROPOLOGIST 2010 VOL. 63(3-4~

site datum

Figure 3. Contour map of Gotier Hammock Mound area showing remains of
mound (three small elevations) and test excavation units (TUs).

student field school crew from 12-26 May (and also for a few
days in May 2009), joined on a couple days by personnel from
the Buffer Preserve.

Environmental and Archaeological Setting

St. Joseph Bay is unlike all other bays in the lower
Apalachicola River delta region in that it is a non-estuarine,
hypersaline lagoon, saltier than the Gulf of Mexico, since little
fresh water feeds into it. It is enclosed by a 24-km-long barrier
spit running north-south, connected to the mainland at the south
end by a shorter east-west arm of land (see Figure 1). The St.
Joseph Peninsula barrier spit is less than a km wide and made
of the pure white sand for which the region is famous (the state
park at its northern tip was declared top beach in the country
in 2002). The bay is 8 to 13 km wide, mostly landlocked, and
has one of the least-disturbed coastal bay systems in Florida.
Salt marshes and sea grasses contribute to its enormous
productivity. Though flowing tributary streams are scarce,
fresh water is sometimes available in swales between dune
formations (Davis 1997:166-67; Rupert 1991).

Inland from the bay shore the topography consists of sets
of parallel, low beach ridges that may merge at unusual angles.
For example, Figure 2 shows Depot Creek, here a shallow,
seasonal, linear wetland, originating as a long swale between
dune ridges and flowing south, then being pushed around by
newer ridges to go northeast toward the Apalachicola River.
The inland environment is one of low sandy flatwoods, today
planted in slash pine but originally in longleaf pine in open,
fire-maintained forest. Small oaks dot the understory of saw
palmetto, wax myrtle, and wiregrass (Schuster et al. 2001:20).
The elevated hammock cover is old oaks and sabal palms.
The archaeological landscape around St. Joe Bay is
distinctive for northwest Florida (Benchley and Bense 2001;
White 2005; White and Fitts 2001; White et al. 2002). As in any
coastal region, shoreline sites are usually shell middens, but the
saline bay waters harbor species of fish, shellfish, and turtles
otherwise obtainable only in the Gulf. So typical middens
here are characterized by large gastropods, lightning or left-
handed whelk (Busycon perversum [formerly contrarium or
sinistrum]) and horse conch (Pleuroploca gigantea the state
shell of Florida), along with the usual oysters, clams, and other

THE FLORIDA ANTHROPOLOGIST

2010 VOL. 63(3-4)

WHITE GOTIER HAMMOCK MOUND &ND MIDDEN

species. In addition to shell ecofacts in the dark sand middens,
there are often artifacts made from these big shells. Back
from the shoreline, prehistoric sites are small and scattered,
represented usually only by a few sherds or shell tools but no
shell middens. Lithic resources are fairly distant, at least 100
km to the north in chert outcrops (though some agatized coral
is occasionally available as beach rock). So it is not surprising
that hard, thick shells were made into tools (Eyles 2004; White
2005:136-38).
The earliest sites around the bay known so far, based on
ceramics recovered, date to Early Woodland Deptford times -
as early as 1000 B.C. (Mayo 2003; White 2005). Prehistoric
settlement patterns show heavier shoreline occupation and
scattered, probably seasonal use of inland areas for hunting
and gathering. The lack of fresh water is probably the reason
for the lack of large inland sites. This scenario made it all the
more fascinating to see Gotier Hammock mound 200 m inland
and about 200 m away from a tiny creek, and the bayshore
midden area with only scattered oyster and no large-gastropod
shell.

Fieldwork: Mound Investigation

Field Operations

Surface collection of all exposed areas produced only a
few prehistoric artifacts. Though project goals did not include
directly documenting the historic component, a sample
of bricks and other recent items was saved for any future
research. We also gathered all the modern trash, both to clean
up the place and to inventory the materials before discarding
them, in case such data might be useful for study of modern
hunting and logging activities (the inventory of modem stuff
was labeled the "trash-a-logue" by the students).
The site was mapped using a mechanical transit and
stadia rod, with the site datum set at the south end (Figure
3). The road appeared to have cut through the center of the
mound, leaving two small high areas. We set up 1-x-1-meter
Test Unit 1 on the west side high ground, and 1-x-2-meter
Test Unit 2 on the east side high ground. Though I originally
intended to dig only one-meter-square units, ANERR's Pat
Millender persuaded me to extend TU2 northward another
meter, promising to help with the additional labor (of course
that north square meter produced the complete pot!). The units
were dug in 10-cm arbitrary levels, with all soils dry-screened
through quarter-inch mesh. For each level, soil samples were
taken from the unit southwest quadrant: a 9-liter (30 x 30 x
10 cm) sample for flotation and another liter for permanent
storage/future research. All excavation continued to culturally
sterile soil, and all units were backfilled.
Test Unit 1. This square had a large pine stump in the
middle of it (Figure 4), making excavation difficult. Fire
ecologist Jean Huffman, manager of the St. Joe Buffer
Preserve, estimated the stump was from a tree cut between
1900 and 1910. The soil layers showed a recently buried brown
topsoil/forest duff stratum up to 8 cm thick above the stump,
separated from the current, similar topsoil stratum by up to 22
cm of mixed gray plow zone. This suggested more recent pine

Figure 4. Test Unit 1, with pine stump in center and buried
soil surface (Stratum la), but no undisturbed mound
stratum; view facing west.

planting or other disturbance had pushed soil on top of the
stump and old ground surface. Below this earlier old surface,
both the white (20 to 25 cm) and the light yellowish-brown
(into which we excavated another 48 cm) silty sand subsoil
strata produced a few pieces of prehistoric pottery, including
the only red-painted sherds recovered.
When the very tiniest remains recovered by flotation were
sorted under magnification, the extent of disturbance in this
unit was realized. A few historic items such as slivers of glass
came from Levels 3, 4 and 5, and a copper bullet tip from
Level 4 (see discussion of cultural materials below). Since
these levels also contained nearly all the prehistoric materials
as well, the interpretation is that disturbance from looting and
pine planting allowed such tiny items to travel downward. The
absence of the black mound stratum in this unit could mean
that it was obliterated or that the mound did not extend this far
west, despite the slightly higher elevation.
Test Unit 2. This rectangular unit (Figure 5) had a black
stratum of undisturbed mound deposits between the topsoil
and white sand natural subsoil. It also had more and larger
ceramics, including a complete plain bowl, on either side of
which were sherds of a Basin Bayou Incised jar, and other
Middle Woodland types. Some had an exterior deposit of a
yellow substance, as well as black, burned organic matter
(soot?). Samples of the yellow deposit were scraped off for
analysis, and a black soot sample was also scraped off and
radiocarbon-dated to A.D. 650 (discussed below).

GOTIER HAMMOCK MOUND AND MIDDEN

WHITE

H2010 VO. 63(3-4)

Figure 5. Test Unit 2 showing pottery being exposed; K. Hageman recording, J. Clevinger troweling, S.
Lonergan brushing, and E. Kimble photographing; view facing northeast.

Figure 6. Test Units 2A (left) and 2, view facing north, showing Feature 1 in cross-section, north walls with
dark mound stratum III), Basin Bayou Incised rim sherd still embedded in the uncleaned balk, large root
in west wall. Walls are not yet cleaned so topsoil stratum is obscured.

THE FLORIDA ANTHROPOLOGIST

0 102 Vot. 63(3-4)

WHITE Goiii. i~ HANI~iocK Mot ND AND MIDDEN

TU2A. Since the pottery was surrounded by dark, partially
disturbed soil, to understand the stratification, near the end of
the project we extended the north half of TU2 another meter
westward, calling it TU2A. This square uncovered more
sherds and a dark stain in the white subsoil, labeled Feature
1 (Figure 6), a small possible pit. The black mound stratum
appeared in this unit, but in places it was clearly cut into from
above and mixed with topsoil; tiny glass fragments (one may
be plastic) were recovered as deep as Level 4.
Features. Feature 1 consisted of the same black soil as
in the mound stratum. It was an irregular oval in plan view,
34 cm east-west by 19 cm north-south. In cross-section it had
a shallow basin shape, 15 cm at maximum depth, but either
intruding upon or intruded into by a flat-bottomed, straight-
sided apparent post mold 20 cm wide that extended another 15
cm from the bottom of the basin. Feature I was taken out in
north and south halves, and totaled 17.5 liters (weighing 2.58
kg). The contents recovered by flotation included charcoal,
charred seeds, a Swift Creek Complicated-Stamped sherd and
sand-tempered pottery crumbs.
Two other dark stains that showed up later in the unit walls
were not given formal feature numbers, but one suggested a
post mold similar to that in Feature 1. In the middle of the
east wall of TU2, this 20 cm wide, dark gray, straight-sided
possible postmold extended down 30 cm from the base of
disturbed, lighter gray, terminating with a flat bottom in the
thin black remnant of the mound stratum. A small pit feature

in the north wall of TU2 extended 16 cm from the bottom of
the mound stratum into the white subsoil but was filled with
mottled gray, white, and yellowish-brown sand.
These three features are hard to interpret. The disturbed
nature of the gray topsoil made it impossible to see whether
they originated in prehistoric or more recent times, since both
the mound builders and the mound looters had caused such
soil disturbances. On the other hand, unit profiles, especially
the north wall of TU2A, did show lighter-colored disturbances,
originating in recent times from the surface, which churned
tup the top of the black mound stratum without completely
penetrating it, and resembled typical shovel-tunneling done by
looters. If the features described above were historic they could
have been from shovelings that penetrated the black mound
layer but did not go through it. The best estimate is that the
lighter gray disturbed areas are from looters and the grayish-
black disturbed areas are fiom the mound builders, mostly
because the looters would probably not have left the whole pot
or large decorated sherds that we recovered.
Establishing site boundaries. Shovel testing was done as
we crashed through the thick, understory vegetation with a
30-meter tape measure while maintaining compass orientation.
The 50-cm square shovel tests (Figure 7, Table 1) averaged 1
m deep, though some hit water shallower. Shovel Test 1 was
50 m north of the datum, and Shovel Tests 2-8 were spaced
15-20 m apart moving westward from the datum. None of
these shovel tests produced any prehistoric cultural materials

Figure 7. Aerial photo of Gotier Hammock mound and midden areas and excavations, adapted from
Google Earth.

WHITE

GOTIER HAMMOCK MOUND AND MIDDEN

Table 1. Shovel Test Data from Mound Area, Gotier Hammock, 8GU2.

No. Location Max depth Stratigraphy (depths in cm; all soils = sands) Cultural materials
1 50 m N of site datum 98 cm 0-20 10YR 6/2 It brownish gray topsoil topsoil:
21-23 10YR 4/2 dk grayish brown iron skillet handle, 252 g; green glass, 4
24-26 OYR 5/1 gray g; 7 brick frags, 57 g; 1 metal frag, 7 g;
27-55 10YR 7/1 light gray 1 bullet casing, 5 g; oyster shell, 12 g;
56-60 1OYR 3/2 very dk grayish brown charcoal, 2 g
61-70 10YR 3/6 dk yellowish brown
71-82 1OYR 4/6 dk yellowish brown
83-98 1 OYR 7/3 very pale brown

2 14 mW of site datum 100cm 1-19 10YR 4/1 dk gray none
20-32 1 OYR 5/2 grayish brown
33-41 1OYR 3/2 very dk grayish brown
42-48 10YR 3/6 dk yellowish brown
49-71 10YR 5/6 yellowish brown
72-100 IOYR 7/4 very pale brown

3 18 m W of Shovel 98 cm 0-25 10YR 6/1 gray none
Test 2 26-31 1 OYR 3/2 very dark grayish brown
32-35 1 OYR 3/6 dk yellowish brown
36-58 1OYR 5/4 yellowish brown
59-98 OYR 7/3 very pale brown

21 mW of Shovel
Test 3

28 m W of Shovel
Test 4

25 mW of Shovel
Test 5
28 m W of Shovel
Test 6, 1 mE of
plowed firebreak
30 mW of Shovel
Test 7

99 cm

59 cm

55 cm

76 cm

0-31
32-52
53-72
73-99
0-6
7-30
brown,
31-59

10YR 5/1 gray
10YR 6/2 light brownish gray
1OYR 2/2 very dark brown
10YR 5/4 yellowish brown
I OYR 3/1 very dark gray
10YR 5/2 mottled with 3/2 grayish
very dk grayish brown
10YR 2/1 black

0-55 10YR 2/1 black

0-76

82 cm 0-82

SOYR 2/2 very dark brown

10YR 2/1 black

none

none

None

none

or shell as far as about 185 m west of the mound area. Historic
items were limited to those in Shovel Test 1, near the mound;
these included bricks, metal, glass, iron old garbage from the
historic occupations -as well as a few modern oyster shells.
Cores were attempted with a 4" bucket hand auger, but were
abandoned since the dune sand was so dry and loose it slipped
out of the core bucket.

Stratigraphy and Soils

The soil profile in the mound area consisted of the
following strata of fine to medium sand:
I. brown (10YR3/4) forest humus/duff layer, 10 to 20 cm
thick.
II. gray to grayish brown (10YR6/1 to 5/2) topsoil up to
20 cm thick that was a mixed layer of topsoil/plow zone
with or without mound soils and later historic materials;
mottling and some clear features originating near the
surface show historic disturbance.

III. dark brown (10YR3/3, 2/2) to nearly black, slightly
harder-packed zone of apparently undisturbed mound
deposits, only 7 to 20 cm thick, with charcoal flecks; this
stratum was what remained unmixed with lighter material
above it from looting or other disturbances. It was only
seen in Test Units 2 and 2A, and lensed out by the south
end of TU2. Its appearance agreed with Moore's statement
that the mound was a truncated cone of dark sand.
IV. mostly culturally sterile, natural white dune sand
(10YR8/1), averaging 20 cm thick.
V. light yellowish-brown (10YR5/6 to 6/8) silty sand
natural subsoil, culturally sterile; color of peanut butter,
may be 20 cm thick or greater; fades into next stratum.
VI. white to very pale brown (10YR8/1 to 8/2) coarse
wet sand near and at water table; color of butter pecan ice
cream.
The contrast was stark between the distinctive white
sand (IV) underlying the mound and the nearly black mound
deposits above it (see Figure 6). The sugar-white beach sand

i i -

THE FLORIDA ANTHROPOLOGIST

2010 VOL. 63(3-4)

Table 2. Shovel Test Data from Mound Area, Gotier Hammock, 8GU2.

Location

Max depth Stratigraphy (depths in cm; all soils = sands) Cultural materials

9 just N of E-W 60 cm 0-4 forest duff, reddish brown -4-50 cm:
plowed firebreak 4-20 10YR 5/1 gray topsoil 1 sand-t plain sherd, 2 g;
line 20-46 10YR 6/2-6/1; probably undisturbed 1 check-st rim sherd, 9 g; oyster shell frags
midden; (not saved)
46-57 10YR 7/1; undisturbed dune sand
57-60 10YR 4/4
10 -50 m S of TU 3 55 cm 0-21 dk gray sand topsoil none, no shell either
21-55 transition to light gray, water table at 55
11 -50 mN of TU 3 100 cm 0-2 brown forest duff none, no shell either
2-8 dk gray topsoil
8-42 medium gray, begin coring
42-60 dark brown (about 10YR4/2);water
table at 61
60-100 light yellowish-brown
12 -45 mS of creek, 74 cm 0-4 brown forest duff -30-50 cm:
20 m E of road, 4-30 It gray topsoil 1 grit-t plain rim, 46 g; 4 sand-t plain
250 m N ofTU 3 30-51 10YR4/4 brown midden sherds 65 g; 2 grog-t plain sherds, 14 g;
51-74 wet, about 10YR5/4 charcoal, 1 g
E wall: -48 cm:
1 cordmarked, grog-t sherd, 9 g;
1 sand-t plain sherd, 29 g; no shell
13 -200 m N of TU 3, 83 cm 0-6 brown forest duff 0-20 cm:
100 m S of creek 6-28 light gray topsoil, scattered shell 1 check-st sherd 4 g; 4 sand-t plain sherds 5 g;
28-39 light grayish brown charcoal 1 g; 9 oyster shells & frags, 147 g
39-62 hard-packed dark brown DATED to A.D. 1500
74-83 pale about 10YR8/1; water table at 81 -20-35 cm:
1 indet stamped sherd, 4 g; 1 grit & grog-t
plain sherd, 6 g; 3 grog-t plain, 12 g; charcoal
,2 g
-35-43 cm:
1 grog-t plain sherd, 1 g; charcoal, 4 g
-43-83 cm:
1 sand-t plain sherd, 2 g; charcoal, 9 g

-5 mN of creek

84 cm

0-5 brown forest duff
5-40 gray topsoil
40-84 dark gray

none, no shell either

naturally occurs beneath the topsoil. The clean line between
it and both underlying and overlying strata where there was
no disturbance may mean that this white sand was leveled or
otherwise prepared before mound construction.
The stratification described above but without the dark
mound layer was present in Test Unit 1 (see Figure 4), which
also had an additional brown forest-humus former ground
surface that was buried when the pine whose stump remained
in the unit was harvested a century ago.
Shovel tests (Table 1) showed that the above-detailed
strata are distinct to the hammock formation. Off the hammock,
the gray topsoil stratum, plowed to a depth of about 30 cm for
planting the pines, most often directly overlay the yellowish-
brown (peanut butter-colored) subsoil, which soon transitioned
into the lighter, nearly white (butter pecan-colored) sand near
the water table. In some tests the ground was so low that the
shallow water table colored the yellowish-brown sand nearly
black, typical of wetland deposits.
An interesting aspect of the site was the absence of bone,
human or otherwise, from the surface or the excavations, in
this supposed burial mound, except for crumbs (usually <.1 g)

recovered from soil samples in the fine screen after flotation.
Perhaps the skeletons of people buried in this mound were long
ago removed by looters or left to decay on the surface after
exposure. Surface bone exposed to the elements disappears
quickly in Florida. One collector said he may have seen bone
fragments lying around when the area was first bulldozed
in the 1960s. The crumbs recovered may be from burials or
faunal remains, but are too tiny to identify without extensive
DNA testing.

Fieldwork: Midden Area Investigation

Just west of the line of shovel tests extending westward
from the mound, we returned to the plowed firebreak closer
to the bay, where evidence of the midden, including scattered
oyster shell, was exposed on the surface. This occupation
area was not recognized or associated with the mound until
recently, probably because it was covered in thick forest until
the firebreaks were cut into it. We assumed it was the living
area on the bay shore for the people who utilized the burial
mound, especially since it produced Middle Woodland pottery

WHITE

GOTIER HAMMOCK MOUND AND MIDDEN

THE LORDA AI UOPOLGIS 201 VO. 633-4

from the surface. It is closer to a couple of small, intermittent
streams (see Figure 7), including the one (apparently unnamed)
creek big enough to be now channeled into St. Joseph Bay.
This midden is 185 to 200 m west of the mound, near the paved
road (C30A) that skirts the bay shore. It extends some 350 m
north-south, with the north end at the creek. It was originally
probably no more than 50 m wide but was spread and damaged
by the construction of the paved road.

Field Operations

Shovel testing to establish midden area boundaries.
Shovel Test 9 was excavated in 2008 north of the east-west
firebreak, over 100 m south and 185 m west of the mound
(Table 2), near what we learned was the south end of this
linear, discontinuous midden. It was placed in an area with
surface oyster shell and black sand, and produced a check-
stamped rim and a sand-tempered plain sherd. During the
2009 season we returned briefly to determine the midden's
extent and integrity with four more shovel tests (Figure 7),
and to obtain material for dating.
Fifty meters north and south of Shovel Test 9, Tests 10
and 11, respectively, contained no cultural material, not even
shells. But farther north, approaching the creek, the shovel
tests produced plain and check-stamped pottery, and a few
charcoal bits. The nine sherds in Shovel Test 12, near the creek,
included a cordmarked one as well. The dozen sherds in Shovel
Test 13, about 100 m from the creek, were also accompanied
by prehistorically-collected oyster shells. A sample of 75 g of
this shell from 0-20 cm below the surface produced an AMS
radiocarbon date ofA. D. 1500 -some 850 years later than the
date obtained for the mound (see discussion below).
Just to be sure of site boundaries, we returned in 2010
to shovel-test north of the creek. In March, our efforts were
mostly useless since this is the height of rainy season and we
hit water within 20 cm. Another attempt in early September
was more successful in that we reached a depth of 84 cm,
but no cultural materials were seen. So the south bank of the
stream can be established as the northern boundary of the site.
Test Unit 3. This was a more controlled, 1 x 1-m excavation
adjacent to Shovel Test 9, dug to obtain stratigraphic data that
could tie the midden area to the mound. It had the same forest
duff top stratum overlying gray disturbed topsoil, and then a
stratum of the white sand mixed with cultural materials that
produced a light gray midden zone from 10 to 20 cm thick.
Under that, the natural subsoil was very dark brown (10YR2/2),
unlike the light yellowish brown hammock subsoil. The dark
color was probably due to the low elevation and proximity of
the water table. This unit produced 64 sherds, check-stamped
and plain and one indeterminate incised, as well as a few tiny
chert flakes.
In the northeast corner of the unit, a portion of a prehistoric
pit feature was exposed. Filled with darker gray mottled soil
(from 10YR3/2 to 6/1), it clearly originated from the midden
stratum and was a large round basin perhaps originally a meter
in diameter and about 28 cm at maximum depth. Removed
separately, the soils of the south side of this feature were

dry-screened, and the north portion (9.5 liters, over 10 kg)
processed through flotation. Feature contents included a
couple of sherds and charcoal radiocarbon-dated to A.D. 1350
(see discussion below), some 700 years later than the mound
date.

Midden Area Summary

From the shovel test data and surface materials in the
many firebreak segments in the area we estimated the extent
of the midden and saw that the shell was not continuous but
patchy. People lived along the shore but did not only or always
collect oysters here or camp in the same spot. The Middle
Woodland ceramics demonstrate that they were here at least as
early as the time the mound was built. If the two radiocarbon
dates are correct, they indicate that people returned during
later prehistory. Mound builders must have stayed on this
shoreline, since food and water were dependable, as compared
with the deer and seasonal acorns available back in the forest.
As noted, also interesting in the midden area was the absence
of the large-gastropod food garbage that characterizes other
shell middens around St. Joe Bay, though some shell tools
were recovered from the surface.
In sum, the archaeological picture is very interesting. The
mound is isolated atop the natural hammock that rises less
than a meter above the surrounding flatwoods, relatively far
200 m from fresh water and from the bay shore. The midden
extends some 300 m north-south and at most 50 meters back
from the water.

Data and Materials Recovered

Materials and data processing in the lab was accomplished
from fall 2008 through fall 2009. Flotation of the 9-liter soil
samples was done with screen sizes as follows: A fraction =
1/4" (6.35 mm), B = #20 geological screen or .034" (.86 mm),
and C = #50 geological screen or .0166" (.29 mm). Recovery
was good; a test with the standard 100 charred poppyseeds
in one soil sample resulted in the later recovery of nearly
all of them (an exact count is difficult since some became
fragmented). All materials, notes, maps and other data are
curated in the USF archaeology lab. A summary report (White
2008a) and a comprehensive final report (White 2010) were
prepared to fulfill grant and permission requirements.
All cultural remains from the USF investigations, as well
as my reconstructions of those obtained by Moore (1902:210-
11) and Floyd (Holmes 1903:111-112), aided by information
from the Smithsonian NMAI and NMNH collections, are
summarized in Table 3 (except for the historic artifacts, which
were only recovered on the mound area surface and in Shovel
Test 1 [reported in Table ]). One can only idly speculate about
what might have been removed from Gotier Hammock before
(or after) Floyd and Moore got there. But Table 3 lists all
prehistoric cultural materials known from the site. The first
two columns are the old collections, the next four are from our
recent investigations of the mound and the last three from the
midden area.

THE FLORIDA ANTHROPOLOGIST

2010 VOL. 63(3-4)

Wi-i ITE GOTIER HAMMOCK MOUND AND MIDDEN

Figure 8. Weeden Island Plain compound bowl (plan view)
from C.H.B. Floyd collection, Smithsonian NMNH (cat.
no. A155329-0) recovered in 1893-4 from Gotier Hammock
mound; drawing by J. Du Vernay based on photo in
NMMH online collection.

Ceramics

Mound pottery included both plain and fancy Middle
Woodland types. The midden area produced mostly non-
diagnostic check-stamped and plain ceramics, but some
Middle Woodland sherds.
Floyd's Ceramics. The NMNH Floyd collection is in
11 separate catalog numbers (A155318-0 through 155328),
comprising the following: 10 ceramic vessels (all apparently
shown in the Holmes [1903: Plate LXXXVII] illustration),
two labeled as "small" and one as "four-cornered"), 1 pottery
"pan," and 8 sherds.
The only one of these items with accompanying photos
in the online collection is the "pan," which is a Weeden Island
Plain compound vessel (Figure 8 drawn from the online
photo). This 5-chambered, shallow open bowl has four rounded
lobes (one clearly reconstructed, differently colored and
shaped) surrounding a central rectangular chamber. Calculated
based on the photo scale, the central rectangular opening is
10.6 x 15 cm, and the vessel's widest point, between the tips
of the two opposing original lobes, is about 46 cm, so this is
a large vessel. It has an irregular "kill" hole in the base, and
the paste looks yellowish. The photo of the underside shows
the label "Franklin County," which would have included in
Floyd's time what was later to be Gulf County.
Compound bowls of this type are common as Weeden
Island funerary offerings. Moore (1903:457) recovered
another one from the Chipola Cutoff mound on the other side
of Gulf County. It is smaller, painted red, and measures about

20 cm at the widest point; it has only three lobes around an
interior rectangular chamber (2.5 x 6.5 cm) with raised sides. I
like to think of these vessels as prehistoric chip-and-dip bowls,
but it is unknown if they were actually for serving something
or had some other function (paint pots? offering trays?).
Holmes's (1903:111-112; Jones 2002:1) discussion of
the collection obtained by Floyd included photos and some
description of the 10 other vessels in the NMNH collection,
summarized as follows:
-an apparently plain, flattened globular bowl with a
curvilinear incision running around the vessel and looping
from the neck down around the body of the bowl
-a Swift Creek Complicated-Stamped late-variety
flattened globular bowl with a short, straight neck and
the concentric-teardrop pattern stamped in a narrow band
around the base of the neck
-a plain, red-painted globular jar with a bird-head effigy
adorno looking inward on the thickened rim "flat on the
upper surface and nearly an inch wide"; basal perforation
was not knocked out after firing but is a 1-inch-diameter
circular hole made apparently when the clay was wet
-a plain (apparently) shallow open bowl
a Swift Creek Complicated-Stamped late-variety
globular bowl with squared neck at top, incurving in
profile; the pattern is stamped in a narrow band below the
folded rim
-a Swift Creek Complicated-Stamped late-variety
globular jar with a folded rim and a long, tapering neck
and the pattern stamped in a wide band around it, covering
almost half the vessel
-a small, plain, very flattened jar or bowl with incurving
neck and thickened, folded rim
-an unusual Weeden Island Incised jar, 16.5 cm tall, of
reddish paste, with a straight long neck; squarish body as
viewed from above, with wing-shaped molded protrusions
at each corer that were incised in teardrop and other
curved patterns; and a narrower cylindrical base. itself
incised and punctated in interlocking scrolls and other
patterns. Holmes (1903:111, Plate LXXVIII A) illustrated
this jar also in a separate, larger figure, with rollout
drawings of the incised and punctated patterns. He called
this a remarkable vessel and thought it demonstrated links
with the aboriginal pottery of the Caribbean and Yucatan,
though it was "as a whole, essentially Floridian"
-a small plain globular bowl with incurving rim and
(apparently) one incision relatively far below the lip
-a plain jar with a small globular body and long curved
neck and folded rim
Holmes ibidd.) said all vessels were of siliceous, fine-
grained paste (so, sand-tempered -but would he have
recognized grog?) and some had mica flecks (typical of
Apalachicola delta ceramics). The paste was a "warm gray"
except for one pot, which had a reddish paste (unclear if this
was the one painted red also). Plain surfaces were polished.
All had basal perforations or "kill" holes. He notes another
specimen not illustrated: a rim sherd with another animal
effigy head. This piece may be included in the NMNH catalog
entry labeled 8 sherds.

0 5cm

GOTIER HAMMOCK MOUND AND MIDDEN

WHITE

THI Fi ORIDA ANIHR0POLOGISI 2010 VOL. 63(3-4)

Willey (1949:256) remarked on the heavy rim
reinforcement on the 3"' and 7'' of these pots listed above as
being reminiscent of the Weeden Island 1 (Middle Woodland)
type Oklawaha Plain in north-central and northeast Florida.
He also noted that the Swift Creek Complicated-Stamped was
of his Late Variety since the decoration was confined to a band
on the upper vessel.
Moore's Ceramics. Moore's (1902:210-11) specimens,
based on his prose and NMAI records, are as follows:
-a Weeden Island Plain "rude, undecorated, imperforate
toy bowl," the only piece of pottery with a burial; possibly
the basal perforation was not done because the bowl was
too small
-a Weeden Island Plain "coarse, undecorated pot of
about three pints capacity, with basal perforation, [which]
lay alone." Since he said "pot" and not "bowl" this may
have been of a different shape, perhaps another jar
"A four-sided cup with flat base, of about 1 pint
capacity, [which] lay in the sand alone"; this is a classic
squared-neck small Weeden Island Plain vessel with a
folded rim, about 10 cm wide; it is of "brown ware" and
has a perforated base (Figure 9).
a Swift Creek Complicated-Stamped "perforate vessel
of about 3 quarts capacity, semi-globular body, upright
and slightly flaring rim"; it had no association, and was
found near the base of the mound. Willey (1949:253, 429-
435) called it late-variety since the stamp was around the
rim only.
a Weeden Island Incised sherd Moore (1902:Figure
140), illustrated with a drawing of the stylized bird
decoration, which "lay with others in undisturbed sand";
it is a bowl rim with a folded lip and yellowish paste
(observable in a photo in the NMAI online collection,
catalog number 174942.000).
-a Swift Creek Complicated-Stamped rim sherd,
illustrated by Moore (1902:Figure 141) in a photo, but
not enough to tell whether it is of the early or late variety;
its pattern seems unusual and asymmetrical.
The only other facts extractable from Moore's (1902:211)
account concerning ceramics are that. "While all vessels from
this mound were of most inferior quality, numbers of sherds
were of excellent yellow paste and decorated with crimson
paint or with incised designs, showing that the aborigines who
built the mound could hold their own in pottery making with
any in this region." Moore made more than one trip to the
Apalachicola delta/lower Chattahoochee Valley region (Brose
and White 1999) seeking what he considered to be the most
beautiful finds, Middle Woodland pots; he liked what he called
"yellow ware." Since he was more interested in whole vessels,
his disappointment that the only four whole ones he found at
Gotier Hammock were plain and ugly is understandable (he
apparently did not know about the Floyd pots).
Ceramics from USF Investigations. To permit comparison,
I describe pottery from the mound area for each ceramic type,
then the midden pottery. The majority of the cultural materials
from all four test units (Tables 4-7) and the surface of both
areas (Table 3) are ceramics. The total ceramic assemblage
from the mound area consists of 118 specimens, weighing

CM

Figure 9. Small Weeden Island Plain vessel recovered by
Moore at Gotier Hammock Mound, NMAI collections (cat.
no. 174013.000). Photo detail (background cropped by
author) courtesy of the National Museum of the American
Indian, Smithsonian Institution; photo by NMAI Photo
Services Staff.

2368 g (including the whole bowl and partial jar). From the
midden area we obtained 249 sherds weighing 2055 g. Minus
the bowl and jar from the mound, ceramic sherds from both
areas weighed an average of about 8 g each.
Test Unit 1 (Table 4) was at the disturbed periphery of
the mound and only produced 12 sherds (87 g). It contained
nothing cultural in the first two levels, which comprised the
buried topsoil and plow-over that covered the pine stump
about a century ago (strata I, II, 1-A in Figure 4). There was no
stratum III, the dark mound layer. The disturbed strata IV and V
contained the artifacts (down to Level 7, 70 cm deep), probably
because plowing for pine planting, house construction, or
even earlier looting disturbed whatever (probably thin) mound
stratum might have once existed. In TU2 and 2A (Tables 5
and 6) the shallowest materials were right below the surface
(possibly thrown up there by pushing the road through), but
most pottery was encountered in Levels 4 and 5 (30-50 cm
deep), in the mound stratum (111; see Figure 6).
The Basin Bayou Incised jar (FigurelO, and seen in situ
in lower left of Figure 5 and right-center of Figure 6) was
reconstructed from sherds in the mound stratum in the north
end ofTU2 and northwest corner baulk ofTU2-2A (though it is
listed only on Table 5 as one vessel). It is sand-tempered, with
an interior diameter at the rim of 18 cm, exterior of 19.5 cm,
and total weight of 857.6 g. The bottom of the vessel, which
was plain-surfaced, is mostly missing, but most of the incised

THE FLORIDA ANTHROPOLOGIST

2010 VOL. 63(3-4)

Table 3. Prehistoric cultural materials from Gotier Hammock site (wts in grams, rounded up).

Floyd, Moore, Surface, TU1 TU2 TU2A TU3 Surface, Shovel TOTALS
mound mound mound midden Tests,
midden

Type N N N Wt N Wt N Wt N Wt N Wt N Wt N Wt N Wt
CERAMICS
Sw Cr Comp-St 3* 2* 4 17 1 12 3 17 1 33 12* 73+
Crkd R Comp-St 1 35 1 35
Basin Bayou Inc++ ?** 1* 858 1 858++
Weeden I Inc 1* 1+?** 2 3 3 3+
Weeden I PI red-pt I* ** 1 5 3 16 5 17+
Weedcn I Plain 6* 3* 9
Indian Pass Inc 2 26 2 26
Keith Incised 1 4 1 4
Carrabelle Pune 1 22 1 22
indetinc 1 10 1 5 1 21 1 8 1 1 1 2 6 47
cordmarked 2 40 1 13 1 4 1 9 5 66
check-st 1 2 21 186 81 813 2 13 105 1014
indet st 3 6 2 15 '
sand-tpl 19 78 4 20 28 310 23 308 22 67 40 284 11 103 147 1170
grit-tpl 5 17 1 <1 8 28 10 108 1 46 25 200
grog-t pl 4 46 2* 525 11 56 19 128 6 27 42* 782
grit &grog-t pl 4 20 1 6 1 9 1 16 1 6 8 57
Total ceramics 11* 3+ 38 190 12 87 37 1748 31 343 64 347 162 1500 23 208 381* 4423+
LITHIC MATERIALS
point tip 1 9 1 9
chert flakes 1 9 10 <1 10 <1 1 15 22 25
ground stone celt 2 1 299 3 299+
mica frags 2 1*** <1 <1 4
red sandstone <1 <1 SHELL ARTIFACTS
columella 2 21 818 4 432 27 1250
scoop/scraper 5 76 2 74 7 150
whelk debitage 20 168 3 36 23 204
ECOFACTS
oyster shell 17 III 147 275
unident bone frag charcoal 16 120 37 88 17 279
charred seed <1 < I <1 <1
* includes whole or nearly whole vessels)
? unclear which type represented by Moore's descriptions
** Moore's numbers of sherds
+ actually greater but no data for Moore's/Floyd's materials
++ all from 1 jar, though Moore may have had some of this type as well
** private collection

GOTIER HAMMOCK MOUND AND MIDDEN

WHITE

Table 4. Cultural materials by level, Test Unit 1 (1 x 1 m; 10-cm levels; wts in grams, rounded up).

Type L3 L4 L5 L6 L7 TOTALS

N Wt N Wt N Wt N Wt N Wt N Wt

Weeden I Plain red-painted 2 13 1 3 3 16

indet incised 1 5 1 5

sand-t pl 2 20.6 1 14 1 <1 1 <1 4 20

grog-t pl 4 46 4 46

TOTALSHERDS 9 85 1 14 2 4 1 1 13 104

chert flake 4 <1 4 <1 2 <1 10 <1

oyster shell 12 5 17

charcoal 3 6 5 2 16

historic items (glass slivers, copper bullet tip) 2 <1 2 2 1 <1 5 3

Table 5. Cultural materials by level, Test Unit 2 (1 x 2 m; 10-cm levels; wts in grams, rounded up).

Type L L2 L3 L4 L5 L6 L7 L8 L9 TOTALS
N Wt N Wt N Wt N Wt N Wt N Wt N Wt N Wt N Wt
Basin Bayou Inc 1* 858 1 858
Sw Cr Comp-St 1 12 1 12
indet incised 1 21 1 21
cordmarked 1 13 1 13
check-st 1 2 1 2
sand-tpl 1 5 2 24 11 211 8 60 3 8 1 <1 1 <1 1 <1 28 310
grit-t pl 1 <1 1 <1
grog-t pl 1 27 1** 498 2 525
grit & grog-t pl 1 6 1 6
TOTAL SHERDS 2 15 1 5 2 24 11 204 9 1416 4 8 1 1 1 I 1 1 32 1675
chert point tip 1 9 1 9
mica <1 <1 < 1
red sandstone <1 <1 <1

bone frags <1 <1 < 1
charcoal 13 10 34 28 26 5 3 <1 120
charred seeds <1 <1 <1 <1 <1 <1 <1 <1
*many sherds all from one (partial) jar, extending into TU 2A, with soot deposit DATED to A.D. 650
** includes entire bowl

design remains around the neck. This design is a pattern of
slanted or leaning, opposing loops lying on top of one another,
with the spaces around them filled in with curving parallel
lines and the spaces within the loops and within the smallest
curving lines filled in with straight parallel lines oriented in an
opposing direction to the flow of the curve. Though this partial
jar is the only example of this type recognizable in the Gotier
Hammock assemblage, other sherds labeled indeterminate
incised may be from other vessels of this type. Pieces of the

bottom ofthe jar might be unrecognizable among the rest of the
sand-tempered plain sherds, though we examined everything
carefully to try to restore the whole jar.
Basin Bayou Incised (Willey 1949:374-76) is the sand-
tempered, Florida version of Marksville Incised, a Lower
Mississippi Valley type with predominantly grog ("clay"
and some [crushed] "sherd") temper but including some
sand temper, with designs including line-filled "meanders"
but vessel forms of bowls (Phillips, Ford, and Griffin 1951).

THE FLORIDA ANTHROPOLOGIST

2010 VOL. 63(3-4)

Table 6. Cultural materials by level, Test Unit 2A (1 x 1 m; 10-cm levels; wts in grams, rounded up).

Type L L2 L3 L4 L5 L6 Mixed Fl TOTALS
N Wt N Wt N Wt N Wt N Wt N Wt N Wt N Wt N Wt
SwCr Comp-St 2 14 1 3 3 17
indet stamped 2 3 1 3 3 6
poss cordmarked 1 4 1 4
indet incised 1 8 1 8

sand-t pl 3 11 2 1 10 265 5 4 1 10 1 <1 22 292
TOTAL SHERDS 5 14 3 4 11 273 9 38 1 10 2 4 31 343
bone frags <1 <1
charcoal 1 1 2 3 1 1 28 37
charred seeds <1 <1 <1 <1 <1 <1
red sandstone <1 <1 <1
glass, plastic? 1 45 2 1 1 1 4 47

Table 7. Cultural materials by level, Test Unit 3 (1 x 1 m; 10-cm levels; wts in grams, rounded up).

Type LI L2 L3 L 4 L 5 L6 L 7 F 09-1 TOTALS
N Wt N Wt N Wt N Wt N Wt N Wt N Wt N Wt N Wt
indet incised 1 1 1 1
check-st 4 22 12 112 5 52 21 186
sand-t pl 6 14 11 20 3 17 1 5 1 11 22 67
grit-t pl 3 15 4 12 1 <1 8 28
grog-t pl 6 42 4 10 1 4 11 56
grit&grog-t pi 1 9 1 9
TOTAL SHERDS 19 93 32 155 10 82 1 5 2 12 64 347
chert flake 3 <1 7 <1 10 < 1
sandstone 1 <1 1 <1
bone frags <1 <1 <1 <1 <1 <1
charcoal 1 43 24 9 <1 <1 9 <1* 88
charred seeds <1 <1 <1 <1 <1 <1 < 1
* charcoal from feature sent for radiocarbon DATING; returned result of cal. A.D. 1290 to 1420

Wimberly (1960:93-98) recognized Basin Bayou Incised in
south Alabama, though he included punctations in the range
of decoration, not just incisions (thus making it overlap
with Weeden Island Incised). His illustrations of the type
(1960:Figures 54, 55) show resemblances with the Gotier
Hammock jar.
The loopy design on our jar is reminiscent of the kinds of
patterns seen on slightly earlier and contemporaneous Swift
Creek Complicated-Stamped vessels, e.g., Willey's (1949:433;
1966:Figure 5-42c) illustrations of stamped patterns with lots
of loops, parallel curving lines, parallel straight lines filling in
other spaces, and decoration only on the neck of the vessel.

This resemblance of incised patterns to complicated-stamped
designs has also been noted by Ashley Dumas (2008; Price
2008:156), who found similar Basin Bayou Incised pottery
in coastal Alabama at Plash Island, some 250 km westward
along the coast from Gotier Hammock. Moving forward
in time, the broad-line scroll-shaped patterns incised on the
Middle Woodland Gotier Hammock jar clearly foreshadow
the running scrolls on Fort Walton Incised and other later
prehistoric and protohistoric ceramic types. All these designs
can be interpreted as birds, snakes, waves, or just curvy
patterns of unknown significance; our Gotier Hammockjar fits
the last. However Moore's (1902:210) Weeden Island Incised

WHITE

GOTIER HAMMOCK MOUND AND MIDDEN

THE LORDA ATHRPOLGIST201 VOL 633-4

Figure 10. Basin Bayou Incised jar from TU2 NI/2 and TU2A balk (cat nos. 08-39, -44, -49, -51,-99, -107); sherds make up
complete upper portion but are too fragile to stay glued together; note encrusted soot, AMS radiocarbon-dated to A.D. 650;
rollout of incised design below was done by hand-tracing.

sherd from Gotier Hammock clearly shows a bird head figure
among other stylized elements.
The Basin Bayou Incised jar (and all the other pottery)
has the micaceous paste typical of this region, deriving from
the natural inclusion of this mineral in the soils. It also has
caked-on (baked-on?) black deposits that are not dark firing
clouds but solids adhering to the surface after heating or some
other process. Less than a gram of this soot (or whatever its
proper name may be) was sliced off with a scalpel and AMS
radiocarbon-dated to A.D.650+40 (see discussion below). The
small standard deviation suggests the date is very reliable for
the vessel and the mound. The jar could be older, perhaps
kept by family members who were descendants of its original
owners or makers, and only buried in the mound at some
special occasion. Or it could have been made specifically for
some mortuary ceremony, which may also have resulted in the
soot deposit. The Plash Island materials noted above in coastal
Alabama, including similar Basin Bayou Incised ceramics,
were likewise dated to as late as A.D. 650, though the Porter
Phase recognized there had previously been thought to extend
only to about A.D. 400 (Dumas 2008; Price 2008)

The few indeterminate incised sherds recovered by USF
operations (three from excavated contexts) for now must be
assumed to come from either Weeden Island Incised or Basin
Bayou Incised vessels, as no other incised types are known
from the mound. We did obtain from the surface of the
mound area two very small sherds of Weeden Island Incised
(Figure 11 bottom), one of which also has a black, baked-on
deposit. Swift Creek Complicated-Stamped sherds from the
mound are few, but interesting. Besides Moore's (1902:Figure
141) single, complex-patterned example, our investigations
produced eight, including one (Figure 11 top) with more of the
baked-on black deposit and two (probably from the same pot)
with a ladder-like design (one is in Figure 11, middle right)
from TU2A. Three mound-area sherds had to be classified
as indeterminate stamped since their surfaces were obscured
(eroded or smoothed).
The red-painted sherds from TU1 are probably all from
the same vessel and have the red pigment on the interior. They
are classified as Weeden Island Plain, since this type (Willey
1949:409) includes sand-tempered plain vessels with red paint.
Moore and Holmes also both found red-painted pots here. The

THE FLORIDA ANTHROPOLOGIST

2010 VOL. 63(3-4)

WHII L GO~IIER HAN1~tocK IVI()1JNI) ~ND MIDDIN

0 1 2 3 4 5

cm

Figure 11. Artifacts from mound area: clockwise from top,
complicated-stamped sherd in fine-line pattern with soot
deposit, from TU2 N1/2 L4 (08-24), complicated- stamped
sherd with ladder-like pattern from TU2A L5 (08-92); 2
Weeden Island Incised sherds from surface (03-1); pinkish
(probably thermally altered) point tip from TU2 S1/2 L3
(08-23).

most notable of the plain-surfaced ceramics is the whole bowl
(Figure 12), which is grog-tempered, with a typical Middle
Woodland folded rim and yellowish paste, and weighing
766.7 g. It had been carefully exposed and recorded in situ
(Figure 5), but when it was lifted from the soil it fell into two
halves, from old damage. The bottom had been knocked out
irregularly, and pieces of the rim fold had broken off. Though
what appears to be a firing cloud darkens the lower part of the
bowl exterior, a baked-on black deposit (which could be dated)
coats this area as well. The exterior bowl diameter is 19.75 cm
and interior, 17.75 cm. By definition it is classified as Weeden
Island Plain, just like the small plain vessel Moore recovered,
which it closely resembles in style, especially the folded rim
(though Moore's was a square cup); it is also similar to some
of Floyd's pots.
Another interesting plain specimen is a sand-tempered
ceramic disc from TU2, L 5, nicely cut to have a beveled
edge. It is a near-perfect circle, 5.7 in internal and 6.9 external
diameter, weighing 50.3 g (though part of it is broken off).
Such disks have been interpreted as gaming pieces or other

Figure 12. Weeden Island Plain bowl from mound, exterior
side and interior top views; note black soot deposit on
exterior and irregular "kill" hole; cat. no. 08-40.

kinds of functional/ceremonial items, but also may just be the
pieces left when the bottom is cut/knocked out of a sacrificed
vessel during the burial or other ceremony (assuming the
person performing the ceremony does a neat job, not like the
irregular bottom piece missing from the complete bowl in
Figure 12).
All the plain sherds are given generic labels by temper
because they may be from plain vessels or pieces of other
types from portions where the decoration was not applied to
the surface. Among the plain sherds listed on the tables are
various extremely tiny crumbs recovered from flotation and
only able to be examined for temper. They usually weigh less
than a gram or even a tenth of a gram, so do not add much to
the totals; but their presence indicates something cultural even
at deep levels (though they may have filtered down).
Though the tempers of the whole vessels in the Moore
and Floyd collections are mostly unknown, some interesting
trends in temper can be documented with the controlled data
(Table 8). In the mound area, most of the pottery is sand-
tempered (81% by number of specimens and 5-1".. by weight),

WHITE

GOTIER HAlMMOCK MOUND AND MIDDEN

THE FLORIDA ANTHROPOLOGIST 2010 VOL. 63(3-4)

Table 8. Tempers in plain-surfaced ceramics from Gotier Hammock, USF
investigations.

M(
STemper N
sand 74
i ... ..... ... ....
Sgrit 6

grog 6
i grit & grog 1 5
TOTALS | 91
TOTALS j91

)UND AREA

5% 26

1000% 1330

100%

MIDDEN AREA

100%

with lesser amounts of grit, grog, and mixed grit and grog
temper. However, the one complete plain grog-tempered
bowl, while counted as a single specimen, brings the amount
of grog-tempered in the mound area to 43% by weight. Grog is
understood to be crushed clay particles, perhaps fired, perhaps
from crushed sherds, of many colors, from pale off-white to
gray or brown to red. The sand-tempered plain has occasional
particles of grit (or perhaps large sand grains). Several sherds
from TU2 glue together to make the side of what must have
been another simple plain sand-tempered bowl, also notable
because it, too, is caked with a baked-on, datable, black
deposit.
In the midden area, sand-tempered pottery is still the
majority, at little over half by both count and weight, but grog
temper characterizes a quarter of the sample, and there are
considerably more grit-tempered plain sherds as well (Table
8). Such greater diversity may reflect the fact that the midden
area occupation took place at several different time periods
in prehistory, as compared with the single-component late
Middle Woodland mound. In summary, concerning temper,
Middle Woodland folks apparently preferred sand, but used
grit and grog as well in differing amounts for both plain and
all other ceramics (for unknown reasons).
Cordmarked sherds (Figure 13) were recovered from the
mound area, two from the disturbed road surface, and one
from TU2. All are sand-tempered and have impressions of
S-twist cords about 1.5 to 2 mm thick, set 1.5 to 4 mm apart
on the vessel. The rim sherd showed that the cord impressions
run nearly vertically on the vessel exterior; the top of the rim
is folded over and smoothed a bit to cover the tops of the cord
marks. The sherd from T1U2 had cord impressions covering
only a portion of the surface, leaving the rest plain. This type
of pottery could be considered fancy or utilitarian depending
upon one's impression of what constitutes "decoration," as
opposed to a surface treatment for some functional purpose
(thermal properties, ability to grip a rougher surface better, or
some other reason).
A cordmarked sherd (Figure 13, right) was recovered
from the midden area, from 48 cm deep (embedded in the
wall) in Shovel Test 12. It is tempered with grayish-white grog
particles as well as fine sand. It has impressions of S-twisted,
about 2 mm-thick cords close together enough that the sherd
was originally identified as sloppy check-stamped. The twists
of the cord, 2 to 3 mm apart, are visible in the sherd and
perhaps more so in a clay positive impression, which shows

a faint hint of thinner strands woven in between those cords,
which may actually make this sherd more accurately classified
as fabric-marked.
Check-stamped pottery is important for understanding
this site and has an unusual distribution. Only one small
sherd of this type, weighing 2 grams, was recovered from
the mound area, and it was not deep, near most of the mound
ceramics, but in TU2, Level 1. All the other check-stamped
(104 sherds, 1012 g) came from the midden area (Figure 14).
Check-stamped pottery is non-diagnostic; panhandle Florida
natives began making it during Deptford (Early Woodland)
times, around 1000 B.C., and kept making it for another three
millennia into protohistoric times (Marrinan and White 2007).
The type associated with Middle Woodland, by definition,
is Gulf Check-Stamped (Willey 1949:387-88), but this type
is only recognizable by its notched or scalloped rim, not in
body sherds. There are no such rims in the entire ceramic
assemblage from Gotier Hammock. Check-stamped pottery
from the midden area, tempered with grit or grog in about equal
amounts, probably represents several different prehistoric
time periods based on the radiocarbon dates (see below).
Understanding the midden area requires noting the other
ceramic types present (Table 3, Figure 14): two complicated-
stamped sherds, one of the Crooked River type; two Indian
Pass Incised; one Keith Incised; one Carrabelle Punctate; and
one indeterminate incised, as well as all the plain specimens
described above. All are general Woodland types.
In sum, the ceramics at Gotier Hammock mound are typical
of Middle Woodland assemblages from both mounds and
habitation sites in the Apalachicola delta. A lot of the pottery
is caked with the black, baked-on deposit that produced the
single radiocarbon date, so additional dates could be obtained
to support the interpretation. Ceramic tempers are variable but
seem not to correlate with type or anything else; they may have
been just what was convenient. Ceramics from the midden
area are diagnostic of only generic Middle- to Late-Woodland
habitation, but the check-stamped sherds, more numerous than
plain sherds, could also be associated with the Fort Walton
occupations indicated by the radiocarbon dates.

Stone Artifacts

Few stone artifacts were found at Gotier Hammock,
probably a function of the lack of local rock for artifact
manufacture. People would have had to go up the river and

100%

THE FLORIDA ANTHROPOLOGIST

2010 VOL. 63(3-4)

Wi-i lIE GoIIER 1IXNI~IOCK MotNEl ~~ND MIDDFN

cm
Figure 13. Cordmarked pottery: left top, rim sherd from dirt road surface (cat no. 08-1); bottom, sherd from TU2 L1(08-6);
right, from midden area, Shovel Test 13 (09-8), with clay positive impression of the sherd below it, showing twisted cords and
possible interwoven strands on left side.

other streams some 80 km to chert outcrops in Calhoun or
Jackson County to get raw material for chipped-stone tools,
and at least equally far for quartzite and other cobbles on the
lower Chattahoochee gravel bars for ground-stone artifacts.
From the mound area, only two chipped-stone items of
typical size were recovered: a biface fragment and a piece
of debitage. The former, a pink chert projectile point tip (see
Figure 11) weighing 8.5 g, came from TU2, L 3. Its rosy color
and lustrous appearance mean it was thermally altered to
make flaking easier. A single flake of local whitish chert (8.9
g, nearly 6 cm long) came from the dirt road surface. It is a
secondary decortication flake, indicating later stages of tool
manufacture, or even sharpening a tool that may have still had
cortex on it. The flake has use wear on three sides of the wider
end and so the narrower end may have been kept as a handle.
It is an expedient tool, possibly kept longer in a region where
stone is scarce.
Chert micro-flakes appeared in the remains recovered by
flotation and sorted under the microscope or magnifying lamp.

In TU Levels 3, 4, and 6 had a total of 10 tiny secondary
flakes from bifacial thinning. One or two of these flakes even
exhibited use wear or retouch; together all weighed less than
.1 g. They might have been produced just by sharpening
some tool. Unfortunately they were accompanied by the
glass noted for this unit that indicated modern contamination.
Whether these tiny items migrated downward naturally or not
is unknown. The other units on the mound did not produce
such flakes. Also during sorting of flotation remains a few tiny
crumbs of red sandstone were found recovered from TUs 2
and 2A; they could be from a material used for pigment, orjust
natural inclusions in the soil.
The midden area also produced very few stone artifacts.
From the surface at the south end came one large chert
secondary flake with a little use wear and the ground-stone
celt. This unusual celt has some narrowing at the butt (left end
in Figure 14) apparently for hafting, and it is of raw material
that is not the typical greenstone but olivine-rich granite
(identified by FSU geologist Joe Donoghue), an igneous rock

GOTIER HAMMNIOCK MOUND AND MIDDEN

WHITE

= "~~"C~,~~, ~----

m9 *r -rsiiB~~

0 1 2 3 4 5

cm

I

Figure 14. Artifacts from the disturbed surface of the midden area: top, right to left, Keith Incised, Carrabelle Punctate
rim (with huge punctations), two Indian Pass Incised rims, Swift Creek Complicated-Stamped, all cat. no. 01-1; middle,
four check-stamped sherds (a rim, two body sherds, another rim); bottom, ground stone celt, two pointed Busycon shell
columella tools with broken bases, all cat. no. 08-02 except middle right sherd and shell tool below it, 08-111.

clearly foreign to northwest Florida. With a chip gone from
the bit and a general battered appearance, it appears to have
been an ax head heavily used before being lost or discarded.
Since the granite probably came from as far away as the
north Georgia mountains, perhaps down the Chattahoochee/
Apalachicola system, the celt may have been an "expensive"
and valued tool and therefore kept until it was worn out or lost.
Chert excavated from TU3 totals 10 flakes so tiny that
together they do not even weigh a gram (similar to the micro-
flakes from TUI noted above); all are from the flotation
B-fractions. Of the three flakes from Level 2, the largest, at .1
g, is a secondary decortication flake; another tiny one is block
shatter and another is a secondary flake from bifacial thinning
or sharpening. All seven tiny flakes in L 3 are secondary except

one, which is both secondary decortication (with a little cortex
on it) and bright red, indicating thermal alteration.
The Floyd collection at the NMNH includes 2 polished
stone celts (cat. no. A170270-0) but the collection information
does not say raw material or any other data. It also includes
an entry for mica (catalog number A170272), indicating there
were two pieces, but no further data, and also the notation
"[Removed]"; so this mica may also be lost to further
research. Doubtless these mica pieces were cut or broken
fragments, since the catalog would have indicated if they
were in some recognizable shape. A mica fragment recovered
by a local avocational archaeologist is amorphously shaped
and measures roughly 5 x 4.5 cm. It is silvery-yellowish and
probably much broken and eroded, since it came from the

THE FLORIDA ANTHROPOLOGIST

2010 VoL. 63(3-4)

WHITE GOTIER HAMMOCK MOUND AND MIDDEN

surface. Though mica occurs naturally in the alluvial sands of
the region, a piece this big would have had to be obtained at a
quarry, probably in north Georgia. Small mica flakes were in
the flotation recovery from TU2, Levels 3 and 4, apparently all
layers of the same piece, which measured about 1.1 cm long.
They (and all the other pieces) may be crumbs that flaked off
larger, fragile mica artifacts.

Shell Artifacts

Moore did not record any shell artifacts from Gotier
Hammock. The Floyd collection at the NMNH includes two
"spiral shells" (catalog number A170271-0) but the records
do not indicate what species or if they were fashioned into
artifacts. Also, the notation "[Removed]" with these items may
indicate they are lost to further study. They had to be artifacts,
however, since gastropod shells, which they must be, would
not occur naturally 200 m away from the bay where the mound
sits. The spiral description and the rest of the shell artifact
assemblage strongly suggest they were columella tools.
Prehistoric whelk and conch shell artifacts are common
in this region, where the raw material is so abundant from
the salty bay waters (White 2005). The USF investigations
recovered 32 shell implements from Gotier Hammock, 25
from the mound and 7 from the midden area. They are listed
in Table 9 (which does not include the pieces of shell debitage
in Table 3, since those are not finished tools). For years we
have been trying to establish a typology of shell tools for
the Apalachicola delta region, where they are rare except at
coastal and estuarine sites. Though we have made progress
(Eyles 2004; White 2005), the type names used below should
still be considered provisional.
The majority of the specimens from the mound area are of
Busycon perversuin, lightning or left-handed whelk: 9 pointed
and 10 bipointed tools. Three pointed columella tools (Figure
15j-1) are of Pleiroploca gigantea, horse conch, recognizable
by the parallel curved grooves and right-hand direction of the
spiral (the naturally pointed base might cause some workers to
classify them as bipointed tools). Another pointed columella
is so heavily shaped that the original shell species is not
observable, except that it cannot be lightning whelk since it
opens to the right, like most large gastropods (it may be of
some other Busycon species, or horse conch, or even tulip
shell). Six columella tools have the apex end bifurcated into
a distinctive U-shaped notch (Figure 15b, c, g, h, o) for some
function. Two columella tools have much of the whorl left
around an interior, sharpened apex, as if to shield the point of
the tool (Figure 15e). Another conch columella is a hammer or
cutting tool (Figure 15a) with some of the whorl left, possibly
as a handle. Many of these columella tools have a distinctive
right-angle cut part of the way above the base that seems to be
for more than just removing all the whorl from the apex so as
to make the point. More columella tools with such right-angle
cuts were recorded at other sites in the immediate area in the
St. Joseph State Buffer Preserve (White 2005). Five lightning
whelk artifacts are scrapers, mostly squarish, with at least
one smooth edge. Two of them are very small; one of these
is smooth on all edges and has a narrower side that may have

been for hafting or grasping (Figure 15q), and the other has a
wide (possible) notch out of one edge and a long narrow notch
cut into another edge (Figure 15r).
Midden area shell artifacts are also predominantly pointed
columella tools. Those of lightning whelk are a rectangular
scoop with a smoothed edge and narrower, probable handle
area, and five pointed columella tools (Figure 14). One of the
latter has a sharply pointed apex inside a cut-away section of
whorl that shields the point. Most of the shell is still present on
this specimen, and also on a horse conch pointed columella of
similar design. Like the two described above from the mound
area (Figure 15e), this may be some previously unrecognized
tool type, a punch or awl with a tip perhaps cushioned from
damage during transport. One additional shell tool from the
midden area is a rectangular scoop of clamshell that retains
a hinge fragment and a possible handle area for grasping. It
is unusual in that it is not a gastropod but a clam, probably
Chione cancellata, cross-barred venus. Usually clams and
oysters are not suitable for tools in this region because their
shells are too thin, but this one is fairly thick.
The shell artifacts from both mound and midden are
similar. Many are expedient tools, squarish fragments cut
from the whorl used as scrapers such that at least one side
is smoothed from use wear. Others are the carefully shaped
pointed columellae. Some had to be multi-purpose tools with
more than one working end. The whole assemblage is very
different from a typical shell artifact assemblage in south
Florida (e.g., Luer 1986; Marquardt 1992), which would
contain large, hafted whelk hammers, net sinkers, adzes and
other cutting tools, as well as beads and pendants. Possibly
the greater availability of stone in northwest Florida accounts
for the different kinds of tools here, but some other factors
may be at work. For example, the marine quahog or venus
clam (Mercenaria campechiensis) shells, large and thick,
were frequently made into tools in south Florida, and I have
seen occasional tools made from them in sites along the
Apalachicola in the riverine interior; but none were found at
Gotier Hammock. This species occurs off the Apalachicola
delta barrier islands, but may not have been available in St.
Joe Bay in the past.
The fact that the shell artifacts are all tools, with no
decorative items, is notable. An important wider research
question for this whole region has been why there is little
evidence for processing and exchange of decorative, ritual, or
sacred objects made of the big gastropod shells, which had
such great significance throughout the eastern U.S. in Middle
Woodland burial mound ceremonialism (Florida whelks
with Ohio Hopewell burials, for example), as well as later
Mississippian times. This is an especially pertinent question
for Gotier Hammock, a Middle Woodland burial mound.
In addition to finished tools, 23 shell fragments classifiable
as debitage were recovered, 20 from the mound area and three
from the midden area. Though only a few have cut marks
on them (see Table 3), they are probably all from artifact
manufacture. All are of lightning whelk except for one horse
conch fragment from the mound area. All are whorl fragments
except for one columella piece from the midden area. Some
are roughly square, for example, the cut piece in Figure 15m.

WHITE

GOTIER HAMMOCK MOUND AND MIDDEN

THE FLORIDA ANTHROPOLOGIST

2010 VOL. 63(3-4)

Table 9. Shell Artifacts from Gotier Hammock (all surface finds, all whelk, Busycon perversum, except as indicated).

SCat No. Type Size (cm) iVt (g) Comments

MOUND AREA SHELL ARTIFACTS

03-2 bipointed columella

Sbipointed colutella
Sbipointed columella

bipointed columella

Pointed columella

pointed columella

pointed columella

scraper

Scraper

scraper

08-54 colunella hammer/
cutting tool

bipointed columella
bipointed columella

pointed columella
.. .... . ---... . .
scraper

scraper

S bipointed columella

S08-87 pointed columella

S pointed columella

pointed columella

108-102 pointed columella

09-1 bipointed columella

bipointed columella

09-2 bipointed columella

pointed columella

L=-14.5

L-=9
L:=9

L=-12

L-=5

L-4.7

7 x 7.5

j3.3 x 4

'2.5 x 2

L-18.8

i--,-
L=11.8

I-
L=9.9

iL-10

14.9 x 4
I-::
4 x 5

L=8.2
L=10.4
S-- 4,... .
L=9.3

iL16.6

L=3.8

L 4.9

SL=4.6

IL=12.8

L-3.7

iCat No. Type I Size (cm) Wt (g)

1 MIDDEN AREA SHELL ARTIFACTS

:01-1 I pointed coluiella jL 7.4 29.7

rectangular scoop L= 0. 134.5
SW=5.5

rectangular clamshell 8. 39.9
Scoop I W=ave. 4.5
01-2 pointed columella L=23 274.5
; I !

08-2 pointed columella

!08-1 I1 pointed colunella

09-7 pointed columella

iL=8.5

L=10-
,L=14

31.1 i apex blunt, base sharp

17.8 apex sharp inside longer whorl fragment to make u-shaped end, broken base
S--i ---------- --- ____ -. ____________--------
S7.0 apex tiny, sharp; base possibly shaped

|55.4 apex sharp inside whorl fragment; base sharp

20.7 apex blunt, probably unmodified base with whorl cut at right angle; of horse conch,
S Ple/uroploca gigrante

S3.0 apex cut & pointed, base probably unmodified: of horse conch. Pleuroploca giganteI
.. .-.~ ..--
12.5 small, young shell, apex pointed, base probably unmodified; of horse conch. Pleuroploca
gigantea
- j--_- ^^ _- ..-
26.3 squarish cut whorl fragment with I straight edge smooth, worn
---j- --- f - -- ------ ------- ----------
.3 6.3 squarish cut whorl, small: 2 straight edges worn smooth. 1 with wide shallow notch:
Another, long, thin notch cut into 3"' edge

.7 2.9 trapezoidal whorl. very small, all edges worn smooth; shape suggests section for grasping
----- 7' '-- hall~---~-- ) :, .. ..----------
190.2 some of whorl left (handle?), broken apex and base; 3.5-mm diam hole partially drilled into
l thickest part of whorl (but is not a pendant); of horse conch, Pleurop/oca gigantea

41.5 apex pointed, with U-shaped notch; base unmodified

S3.6 apex and base sharply pointed: U-shaped notch at apex and part of whorl surrounding it is
Ssquared-off and smooth from use

112.9 tiny point of apex surrounded by whorl, base broken

.6 I22.7 squarish, 2 edges somewhat smoothed
-- - .... .--- --i
17.6 squarish. 1 smoothed edge

16 1 apex broken but has U-shaped notch; base more chisel-shaped

21.3 | all very smoothed, apex U-shaped. base pointed

23 5 apex broken but poss. once pointed; base pointed, right angle portion of whorl left above
base

S188.5 apex poss. once pointed inside much larger whorl fragment, base sharply pointed
S1.2 tiny, unmodified base of shell, sharp tool tip

3. I U-shaped notch at apex

3.2 apex has chisel-like point

|25.0

5.2 ] all smoothed, pointed apex. unmodified whorl around base; not B. perversumn; could be

tulip shell, horse conch, other Busvcon

Comments

chisel-pointed base, apex rounded (for pounding?)

has handle area, smoothed edge

rectangular with hinge fragment present, all smooth and worn with one sharp edge, possible
I handle are for grasping: probably cross-barred venus Chione cainceclllta

apex pointed and whorl cut away. sides cut, rest of shell present: of horse conch.
I Pleuroploca gigantea

sharply pointed apex, base broken

Broken base

sharply pointed apex inside cut-away whorl that surrounds it: much of shell still present;
Base unmodified

-4 -- -
20.1

125.0

j 108.1

WH lIE. GorIER HAMNIOZK MOUND AND N'IIDDLN

Figure 15. Shell tools from mound area: a) conch columella hammer with part of whorl (handle?); b-d) 3 bipointed whelk
columellae with notched apices; e) whelk with broken base, tiny pointed apex (all these cat. no. 08-54); f-i) 4 whelk bipointed
columellae (2nd and 3rd with notched apices); j-1) conch pointed columellae (all these cat. no. 03-2); m) cut whorl debitage
(08-54); n) small bipointed columella; o) small pointed columella with notched apex (both cat. no. 09-1); p) bipointed
columella (cat. no. 09-2); q-s) 3 whelk scrapers, showing worn edges at bottom (all cat. no. 03-2).

All the shell artifacts are from the surface of the site.
Perhaps they were still there because looters prefer to take
pottery and stone tools but find shell to be of less interest,
since they live near the bay that produces so many such shells.

Biotic Remains: Ecofacts

Unmodified faunal remains were few at Gotier Hammock
(Tables 2-7). Oyster shell at the mound area was minimal;
a few were seen scattered on the ground surface but not
collected, and the recovered oyster was all from the disturbed
upper levels in TU1 and shovel test 1. This shell has to be from
modern hunters, who sit and wait for deer and traditionally
often eat oysters (and drink beer). By contrast, the midden area
has ancient oyster shells, but they are scattered only loosely
and intermittently across the whole 50-m-wide-by-350-m-
long area. Besides surface shells exposed and moved around
in firebreaks, oyster was uncovered in Shovel Tests 9 and 13,
at opposite ends of the linear bayshore midden (Table 2).
Most shell midden sites around St. Joseph Bay, from
Early Woodland through historic Native American times,
as noted, are characterized by large gastropod shells that
presumably resulted from harvesting these creatures for food,
since they are so easily available in the salty bay. There are
typically also oyster, clam, and other shells and animal bone
better preserved than usual since the shell cuts the soil acidity
that causes decay. (In addition, there are typically shell tools,
as described above). So an important research issue at Gotier
Hammock is the very different nature of the midden, with its
sporadic oyster and no large-gastropod ecofacts. It is unknown
why the prehistoric inhabitants of different time periods did

not harvest these species like typical campers, and why they
did not harvest a lot of shell in general. Perhaps they only
stayed for brief visits, and mostly ate fish.
Another fascinating research question concerns the
oyster shell in the midden area and in other large-gastropod
shell middens around the bay. Since oyster-shell middens are
common along the Gulf, it never occurred to me over the years
of this research to ask where the St. Joe Bay prehistoric people
got their oysters. But experts have recently pointed out to me
that this bay is far too salty for oysters. So people had to go
either around the delta to Apalachicola Bay for the oysters, or
else farther north and west to other bays with fresher water.
An alternative explanation is that St. Joe Bay was less salty at
some times) in the prehistoric past (discussed more below).
Only the tiniest crumbs of bone were recovered, from both
mound and midden, and always from the excavated levels'
soil samples after they were processed through flotation and
the remains sorted under magnification. Even identification
of some of these crumbs as bone is uncertain, not to mention
what species they might be. Far more evidence is required
before we can discuss prehistoric animal use.
Charcoal was also recovered from mound and midden
excavations. Most of it is small pieces recovered in flotation
samples, and most looks like wood charcoal. Some of
the carbonized material in both mound and midden areas
(specifically in TU 1, L5, and TU3, L3) looks like bubbly burned
sugar and may be charred resin fragments, perhaps from pine.
Though modern natural materials identified in the flotation
remains, such as insect carapaces and roots, have been omitted
from the tables, some of the things included, such as seeds,
may indeed be modern. Given this caution, it is important that

WHITE

GOTIER HAMMOCK MOUND AND MIDDEN

2010 VOL. 63(3-4I

many clearly charred seeds were recovered. For example. TU3
Level 2 flotation remains in the B- and C-fractions included
hundreds of charred seeds of various shapes and sizes. Some
are spherical, between 1-2 mm in diameter; perhaps some
of these are fern spores. Others are oval and fluted, some 3
mm long, and resemble the forms for different species of
bulrushes (Martin and Barkley 1961:90-91). Still others are
sub-rectangular or bi-lobed or other shapes. More research
could be done here by a paleoethnobotanist to get useful data
on past environments and human use of them.
Though it may not be biological in origin, the carbonized
deposits on the mound pottery are worth noting again. Besides
the dating, they could be analyzed for composition. The
yellow soil deposit also on the jar, which appeared silty and
slightly slimy at the time the sherds were first uncovered in
the field, was analyzed further. After drying, this deposit was
not as bright a shade of yellow (10YR6/6, brownish yellow)
as when first exposed (10YR7/6 to 8/6, yellow). The ceramic
paste of the jar is yellowish too, ranging from 10YR7/4 and
8/4 (very pale brown) to 8/6 (yellow), to 7.5YR7/6 (reddish
yellow). Also the natural subsoil under the white sand is light
yellowish brown silty sand (10YR5/6 to 6/8). But the yellow
deposit was brighter than the subsoil or the ceramic paste.
It was investigated by Christian Wells, USF archaeological
soils expert, and grad student Kara Rothenberg. Table 10
summarizes the characteristics manifested in the two samples
analyzed, some yellow soil adhering to the jar exterior, and
some mixed gray and yellow soil on the interior.
Results indicate this yellow deposit is sandy soil with
relatively high clay content, and very high in phosphate,
indicating the decomposition of organic remains. With soil
organic matter of less than 1 percent, the high phosphate is
not natural, but likely anthropogenic. The soil conditions (20-
30 percent clay and slightly acidic) are ideal for long-term
preservation. The yellow deposit is likely from some organic

Table 10. Analysis of Yellow Soil on Basin Bayou Incised Jar in Mound.

substance included in the pot. It might also be from the clay
of the ceramic pot itself wearing off under moist conditions
during burial.

Dating the Site

The three radiocarbon dates obtained for the Gotier
Hammock site (Table 11) require discussion. Dating the black
deposit on the Basin Bayou Incised jar is the same as dating
the mound. Even if this deposit came from burned wood, food,
or some other substance that was older than the jar, or younger,
probably the age difference would be in years or tens of years,
not centuries. Two notable aspects of this A.D. 650 date are
that the small standard deviation makes it fairly tight, and it is
fairly late for a Middle Woodland, Swift Creek-early Weeden
Island mound. However, as noted, similar results are being
obtained for late Middle Woodland elsewhere along the Gulf
Coast.
The two dates on the midden include one on oyster shell,
sometimes considered less accurate. But if correct, either or
both of these dates indicate, not unexpectedly, that people
were staying on the shores of St. Joseph Bay and obtaining
its resources at many different times in the past. The check-
stamped and plain pottery could fit well with the dates, and be
associated with both prehistoric and protohistoric Fort Walton.
There is nothing in the midden artifact assemblage to prevent
the dates from being correct, but this assemblage is mostly so
generic as to be near-useless for indicating cultural affiliation
except for the Middle Woodland sherds that do indicate a
component contemporaneous with mound use.

Historic Artifacts

Since this project is concerned with the prehistoric
component of Gotier Hammock, historic materials from the

Texture

Soil Organic
Matter

Beta
262381

Hydrogen
Potential

Ave. Phosphates

PO4 P
Sample Sand (%) Silt () Clay (%) SOM (%) pH P205 (ppnm)
(PPm) (ppm)
IG (yellow, exterior) 73.30 6.67 20.03 0.91 6.55 437.75 584.38 191.25
2 (mixed, interior) 60.00 10.00 30.00 0.74 6.34 337.88 452.63 146.63

Table 11. AMS Radiocarbon dates for Gotier Hammock site.
Provenience Material Cat Date No. Radio Calibrated Calibrated 95% Cultural affinity
No. carbon Intercept(s) probability
yrs (BP) age range
TU2 NW baulk&TU2-A, NE baulk, sherds in situ off Basin Bayou 257057 -40 (Swift Creek-early
153 cm depth, in mound Incised jar Weeden Island)
stratum exterior
TU3, Feature 09-1, NW half, ca. 2 g charcoal 09-31 Beta 610+40 A.D. 1320 A.D. 1290 to 1420 must be Fort Walton
from flotation. B-fraction (.86 262382 A.D. 1350
mm mesh) A.D. 1390

Shovel Test 13.0 to 20 cm
depth, with check-stamped
and plain sherds

75 g oyster shell 09-12

prototohistoric and/or late
Fort Walton

THE FLORIDA ANTHROPOLOGIST

A.D. 1500

830+40

A.D. 1450 to 1580

0 102 VOL. 63(3-4)

Wut IF GOrIER HAMMOCK MOUND AND MIDDLN

mound area are not examined here, though they are included
in the database of site materials. These artifacts characterize
what could be considered three components. The early and late
nineteenth-century occupations and probably the early looters'
activities are represented by bricks of various types, and
domestic items such as some of the glass, whiteware ceramic
sherds, porcelain, and metal fragments, including an old skillet
handle (see Table 1). Early twentieth-century turpentining
produced sherds of Herty cups used for collecting pine resin.
Recent visitors to the site left glass. metal, and plastic. No
historic items came from the midden area excavations, and
surface items were all modern garbage this close to the paved
road.
Mound and Midden Interpretations

Mound Natural and Social Context

The Middle Woodland was the time of the height of
burial mound ceremonialism in the eastern U.S., and the
lower Chattahoochee-Apalachicola drainage basin was a
major heartland for it. Though Gotier Hammock mound is
not large or impressive, it probably served well what may
have been a relatively remote corner at the southwest end of
the river delta. Typical of the rest of this valley, people used
both Swift Creek and early Weeden Island ceramics. More
interaction northward into the interior than cast-west along the
Gulf is indicated. While the Basin Bayou Incised jar suggests
westward connections, it is very interesting that there is no
Santa Rosa pottery, which would be characteristic of Middle
Woodland closer to the Pensacola area. This is not Santa
Rosa-Swift Creek Middle Woodland but clearly Swift Creek-
early Weeden Island Middle Woodland (Willey 1949). The
late date could reflect the amount of time that burial mound
building took to get here if transportation networks included
intennittent, seasonal streams or, more likely, that Middle
Woodland hung on longer here after things were changing
elsewhere.
This location is today away from major transportation
routes, except for easy (but long) movement by water around
the bay and Gulf shores. Technically St. Joe Bay is not even
part of the Apalachicola drainage system at present, since no
streams connect it directly to the river. But it is part of the main
delta formation of this big river, which originates hundreds of
km away in north Georgia. The river might have been reached
by way of Depot Creek, some 1500 m to the cast of the mound
(Figure 2), which may have been more navigable 1350 years
ago. Figure 16 shows the lower Apalachicola region and this
potential connection. Today in its upper reaches, Depot Creek
is a wide, shallow, often dry depression, but it does flow some
20 km northeastward into Lake Wimico, which flows east into
the Jackson River, which goes east into the Apalachicola River.
The Apalachicola was farther west earlier in time. It has
been migrating eastward since the end of the Pleistocene,
pushed by rising sea level. Archaeological sites demonstrate
human responses in settlement pattern to this fluvial change
over time (Donoghue and White 1995). Our survey of Black's
Island, in St. Joseph Bay just 3 km offshore from Gotier
Hammock, provides supporting evidence. The 11-acre Black's

Island (see Figure 2) is a multicomponent prehistoric site.
Faunal remains recovered there associated with Woodland
and/or Fort Walton components included bones of freshwater
fish (Mayo 2003:75). While this could mean people went far
from the salty bay to get food, more likely it means that more
fresh water was closer during the past than it is today. Depot
Creek itself may be former river channel. Its upper course,
with a near 180-degree meander, indicates it has been heavily
influenced by the natural formation of successive beach ridges
that run north-south on its west side, as well as those south of
it that run east-west.
Gotier Hammock mound's small size and absence of a
nearby large contemporaneous village suggest a relatively
small population bringing the honored dead to be buried but
not staying long. The mound location, back from the bayshore
on higher ground, may indicate pains taken to find a suitable
place more secure from flooding or storms, or more apart
from living areas. The higher ground of the hammock itself,
covered in oaks and palms amid the fire-maintained pine
flatwoods around it, may have been appealing exactly because
of its topography. In a natural fire or flood the hammock would
not be damaged, perhaps avoiding disrespect to those buried
there. While camp ground could easily be moved in response
to adverse conditions, the mound could not. Returning in the
winter of 2009 to take more photos of the site, we saw how a
recent controlled-burn fire had blackened all the pine woods
but not even touched the oaks on the mound.
Florida's pine forests are now often dense because they
are planted that way, but before modern wildfire suppression
the landscape was very open, with understory plants that were
sun-loving, not shade-adapted. The thick tangle of forest in the
midden area today is a direct result of secondary growth that
has not been burned, either naturally or by human intent. Any
fire in recent times will go out at the first ditch or firebreak,
and not regularly burn off the undergrowth. But in the past, the
natural land around the mound and higher hammock would
have been low, open pine savannah. Thus we might imagine
that at the time it was being used, the mound might have been
very visible from the habitation area nearly 200 m away.

Midden Location and Water Relationships

The isolated mound was perhaps removed from living
areas because of its nature as sacred space. People who used it
apparently came from far away and did not live at the mound.
They probably did camp on the bayshore, the most reasonable
living area. Short-term stays, only long enough to bury the
dead, might be indicated by small zones within the midden
that our shovel testing and especially our selected couple
of dates just did not hit. On the other hand, while the single
check-stamped sherd from TU2 at the mound could be of the
Middle Woodland type Gulf Check-Stamped (only identifiable
from rim sherds), it could also be from later Woodland or Fort
Walton occupants of the midden who went in to the mound
(hunting inland? visiting an old holy place?).
The mound is today some 120 m from the nearest fresh
water, an intermittent creek to the north-northwest. The larger
creek that joins it (both apparently unnamed), about 300

WHITE

GOTIER HAMMOCK MOUND AND MIDDEN

THE FLORIDA ANTIIRoPoI.ocis I 2010 VOL. 63(3-4~,

Figure 16. The lower Apalachicola River delta and St. Joseph Bay region, showing connection between Gotier Hammock
and the main river via Depot Creek (white dots), Lake Wimico and the Jackson River; Gulf-Franklin County line shown
in light gray and tiny gray dots (main river channel); adapted from Google Earth (whose individual images leave squarish
patches around the islands).

m northwest of the mound, may have been the reason the
hammock was chosen for mound building, if the larger creek
was a more permanent water source. People could camp on its
banks at the bay shore and get fresh water and seafood while
staying to use the mound. This larger stream, which is now
channelized at its mouth (visible in Figures 7 as a backwards
black Z shape northwest of the midden), seems today to be the
only fresh water feeding into the south end of the bay.
The configuration of these streams during prehistoric
times may never be known. Certainly they feed into or from the
swales between the dune ridges. Depot Creek flows southward
down a swale (see Figures 2, 16) until it hits the east-west-
trending old shoreline ridges and is abruptly turned north to
flow north into Lake Wimico. But the small stream drainages
to the west of the Gotier Hammock mound were probably
closer and may have provided the fresh water necessary for
people to stay there a while; they may even have connected to
Depot Creek.
Researching another site, I found interesting historic
information about drainage patterns here. In 1718 the French
constructed a fort on the mainland at the north end of the bay
opposite the point of St. Joseph peninsula. Though the fort
was abandoned after only two months, its historical location
appears on early maps that also show nearby streams, which

the French (unlike the Spanish) noted carefully and used to
their advantage (Rogers 2009; Weddle 1991:208-10). Jean
Beranger's map, made when he came to St. Joseph Bay, dated
May 1, 1718 (Bdranger 1718; Weddle 1991:Figure 11), and
Jean Baptiste Bourguignon d'Anville's map of Louisiana
(excerpted in Figure 17), completed in 1732 and published
in 1752 (University of Alabama 2010), both show the fort's
location at the north end of the bay next to a stream described
as "Ruisse l'au dousse" and "Eau douce," respectively,
meaning freshwater creek. So this was a good reason for the
French to be at that spot, to have a source for drinking water,
and a small creek still flows in that vicinity today. The only
other stream on both maps emptying into the bay is drawn
at the middle-south end, and certainly looks like it could be
the unnamed stream near Gotier Hammock. But it is labeled
"Ruisseaux salle" and "Eau salee," respectively, on these two
maps, indicating a saltwater creek.
Full of mistakes for much of northwest Florida and
the Apalachicola valley, these maps are nonetheless fairly
accurate for the St. Joseph Bay area, though the cartographer
for the later map probably copied the earlier one and had
not actually been there. The maps even indicate with little
white ovals the locations of Black's Island (labeled "Island
of Turtles") and Conch Island (labeled "Island of Savages").

THE FLORIDA ANTHROPOI.OGISTI

2010 VOL. 63(3-4)

WHITE GOTIER HAMMOCK MOUND AND MIDDEN

Figure 17. Detail from the 1732 Bourguignon d'Anville
map (adapted from University of Alabama 2010) showing
St. Joseph Bay, stream near Gotier Hammock labeled
"Eau salve" (salt water); black dot is location of Gotier
Hammock; note dotted line from that stream going
northeast to a big river.

Both islands themselves are archaeological sites, and the
names are interesting. The bay is full of sea turtles, which were
easily obtained by native inhabitants and are well-represented
in shell middens of the region (White 2005). Conch Island is
small and so visible as a white shell midden that the French
may have recognized it was actually built by the native
"savages." The most interesting aspect of the Bourguignon
d'Anville map here is the dotted line connecting the saltwater
creek near Gotier Hammock with a larger stream named the
"Calistobole River" (origins of that name are so far unknown).
Since the Apalachicola is yet another river east of this stream,
perhaps the map shows an interpretation of Depot Creek
leading to the Jackson River. It all indicates a navigable or at
least an established route, whether all by water or not, between
St. Joe Bay and the big river. Other old maps suggest a similar
connection, so more research on this topic is planned.
The larger creek near Gotier Hammock today, from which
the shell midden extends southward, is definitely a freshwater
stream (I tasted it). But it has been altered a great deal in
modern times. Saltwater intrusion in the past could have been
possible from storms, tides, seasonal effects or other causes,
making what is today fresh water not available in the past.
What few streams there are in this region originate usually
as intermittent swales between the dune and beach ridges
that constitute the whole area. Fresh water accumulates and
in rainy periods may swell the linear trough until it can flow
out to the bay and become a running stream (this is clear on
St. Vincent Island to the south, which we surveyed in 2009,
where archaeological sites were located at the mouths of such

streams). During dry times, or perhaps severe storms that cause
encroachment of saltwater, such an intermittent stream might
become salty. People may have sampled it before they decided
to camp in a given year/century, knowing it was changeable.
Prehistoric natives undoubtedly knew exactly what
season, even which years the water would be running deep
and fresh. The midden area was probably inhabited during
the fall and early winter rainy season. By late winter it might
have been too wet; in March 2010, I saw the landscape from
the mound westward to the midden was in shin-deep mud
and water after typical winter rains. Middle Woodland people
may have conserved the bodies or skeletons of their dead (a
common practice in the aboriginal Southeast) until the right
season to bring them to bury in the mound, when they could
be assured of adequate camping conditions. There are other
reasons for using shell middens seasonally (Meehan 1982;
Waselkov 1987), ranging from availability of different plant
and animal resources to the avoiding the summer insects. Both
the historic residents of Gotier Hammock, as noted above, and
the archaeology crew testified to the latter!
The Gotier Hammock midden area, extending 350 m
along the bay shore, was inhabited during late prehistoric and
protohistoric times, according to the radiocarbon dates, as
well as the Middle Woodland, according to the ceramics. It is
a shallow, low-density midden with atypical scattered oyster.
These facts may indicate that it was only used sporadically
over prehistoric time, perhaps because the fresh water source
was sporadic. People using Gotier Hammock mound could
also have stayed on another piece of shore around the bay, or
even on Black's or Conch Island. But the Gotier Hammock
midden was the closest motel with groceries, drinks, and a
nice view! Perhaps some more long-lasting saltwater infusion
made the nearby creek unusable and ended the appeal to Middle
Woodland mound builders before they could get much of a
settlement going. Later, short-term, overlapping occupations
may have resulted when the water turned fresh/flowing again.

The Mound in Regional Context

It is important to understand Gotier Hammock mound
within its wider archaeological context. Compilation of
information from the USF northwest Florida archaeological
database (Frashuer 2006:80-81) shows Middle Woodland
mounds are distributed all along the Apalachicola valley
and the portion of the lower Chattahoochee valley (about 25
river navigation miles, or 40 km) in Florida (Figure 18). All
but four of the 30 known mounds have both Swift Creek and
Weeden Island pottery. Three have produced only Swift Creek
ceramics and one only Weeden Island ceramics (though these
numbers could change with future investigations).

Related Nearby Sites

Contemporaneous Middle Woodland mounds closest
to Gotier Hammock are two others in southern/coastal Gulf
County. One is Richardson's Hammock (8GU10), on St.
Joseph peninsula, on the opposite (west) shore of St. Joe Bay
from Gotier Hammock. It does have the large gastropods

WHITE

GOTIER HAMMOCK MOUND AND MIDDEN

THE FLORIDA ANTHROPOLOGIST 2010 VOL. 63(3-4~

A early Weeden I only

Figure 18. Distribution of Middle Woodland mounds in the Apalachicola-lower
Chattahoochee Valley; most have both Swift Creek and early Weeden Island ceramics
(adapted from Frashuer 2006).

more typical of the region, and a superimposed Fort Walton
occupation (White et al. 2005, White and Fitts 2001; White et
al. 2002). The other mound is at Indian Pass peninsula (8GU 1;
Moore 1902:211-14; Willey 1949:253), southeast of Gotier
Hammock; it is the only one in the valley with no known Swift
Creek pottery (triangle on Figure 18). Indian Pass mound is
on the central ridge of the peninsula and has an oyster shell
midden (8GU17) some 200 m west of it on the bay, similar
to the situation at Gotier Hammock. The next-closest Middle
Woodland burial mound is at Eleven Mile Point (Moore
1902:214-16) on the Apalachicola Bay shore even farther

eastward; its oyster shell midden is adjacent to the mound but
also has Deptford and Fort Walton components.
Recently archaeologists with the National Park Service
Southeast Archaeological Center in Tallahassee have been
investigating prehistoric sites on Tyndall Air Force Base in
Bay County, on a barrier peninsula that forms East Bay, the
next drainage system west (northwest) of St. Joseph Bay. They
relocated Moore's (1902:196-7; Willey 1949:247-8) Mound
Near Baker's Landing, 8BY29 (the white square in Bay County
on Figure 18), which had eluded many previous twentieth-
century archaeologists. At 1.6 m high and 22 m diameter, this

THE FLORIDA ANTHROPOLOGIST

2010 VOL. 63(3-4)

WHITE GOTIER HANINIoCK MOIND AND MIDDEN

sand mound was comparable to the one at Gotier Hammock. It
was also looted, and produced one St. Andrews Complicated-
Stamped and at least two Swift Creek Complicated-Stamped
vessels but no Weeden Island ceramics. A circular oyster shell
midden sits on the bay shoreline 10 m from the water, while the
mound is inland 70 m to the southwest (Russo 2009; Shanks
2009). Not only did the Swift Creek pottery include at least
one tetrapodal base, but also the radiocarbon dates obtained
were on the midden and were A.D.290 and 380. These data
suggest an Early Woodland occupation earlier than the mound
or an early Middle Woodland mound.
Shanks (2009:12) notes that Hare Hammock mound,
about 5 miles from Baker's Landing, has a shell midden dated
to A.D. 80 to 300, but also a later midden dating to A.D. 400
and 425, with ranges overlapping the ranges for the Swift
Creek-only midden at Baker's Landing. So caution is needed
in the region, since middens may not be necessarily associated
with nearby mounds, and later folks may have continued to
use mounds or middens.

Apalachicola Middle Woodland

Middle Woodland Swift Creek-early Weeden Island
ceramics in the Apalachicola-lower Chattahoochee valley
region "mark an aesthetic high point in Eastern prehistory"
(Willey 1966:288). Along with fancy pottery, this time period
sees, even at habitation sites, the use of the widest variety of
stone raw materials and other exotics of any prehistoric time
period. Middle Woodland begins early and lasts a long time
here, and includes both Swift Creek and early Weeden Island
ceramics. Despite the tendency for some archaeologists to think
those two ceramic series represent successive archaeological
"cultures," there is so far no supporting evidence for this idea.
A common dilemma occurs where earlier researchers
equated specific dates with archaeological "cultures" and
recognized both the culture and the time period by the
presence or absence of specific marker types of artifacts or
other characteristics. New dates and other information may
not fit exactly within these established "culture-periods."
The awkward terminology of archaeological writing includes
many constructed sequences in culture history that surely
would have bewildered the past peoples themselves and that
now may be seen as more variable than originally thought.
Middle Woodland culture means the time of the burial mound-
building throughout the eastern U.S. and the production of
elaborate grave goods. But the exact dating varies from region
to region.
In the Apalachicola delta, where Middle Woodland
sites are almost always characterized by a mixture of both
Swift Creek and early Weeden Island ceramics, the former
may appear slightly earlier. Gotier Hammock mound fits
well within the majority of these mound centers, with both
Swift Creek and Weeden Island ceramics. Swift Creek
pottery without any early Weeden Island types does appear
at habitation sites perhaps as early as A.D. 200 in what we
could call late Early Woodland, and extends through Middle
Woodland. Documenting a Swift Creek occupation and

possible platform mound on a creek in the Chipola River valley
(lower Chattahoochee-upper Apalachicola drainage) at the
Waddell's Mill Pond site (8JA65), Tesar and Jones (2009:716-
717) obtained a date of 1780+80 radiocarbon years, which can
be calibrated (at one-sigma; Cologne Radiocarbon Calibration
& Paleoclimate Research Package 2007) to A.D. 243+98. The
Overgrown Road site (8GU38), a small Swift Creek camp I
tested in the lower valley, produced a raw radiocarbon date of
1650+50 (White 1992:24) which, similarly calibrated, comes
out to A.D. 407+76.
Data and materials from USF's research at the Otis
Hare site (8LI172) in the middle valley (White 1991) are not
yet completely processed, but radiocarbon dates from this
multicomponent freshwater shell midden suggest a similar age
range for Swift Creek without accompanying Weeden Island
ceramics. Two pit features from the earliest occupation of the
site, containing only Swift Creek Complicated-Stamped and
plain ceramics, produced raw radiocarbon dates of 1580+80
(Feature 15) and 1480+70 (Feature 22), calibrated at A.D.
470 and 547, respectively. After this, the early Weeden Island
sherds begin to appear in the stratigraphic sequence.
Thus, the full-blown expression of Middle Woodland
culture seems to be centered around A.D. 500-650 in this
valley. After that, there is a relatively slow slide into Late
Woodland, late Weeden Island, characterized mostly by
check-stamped and plain pottery, with the fancy Weeden
Island Incised, Punctated, and red-painted, and most of the
Swift Creek Complicated-Stamped disappearing and only
the more (to us) mundane types Keith Incised and Carrabelle
Incised and Punctated remaining as a remembrance of fancier
things past. Late Woodland also has far fewer exotic artifacts/
raw materials in general, and the mound building apparently
diminishes or stops. Willey (1966:250) recognized that the
elaborate ceramics and mound building lasted longer on the
coast. Milanich (1994) was less willing to call anything Middle
Woodland if it was later than the Hopewellian manifestations of
the Midwest, with their own elaborate burial mound traditions,
and so sidestepped the issue of labeling. But his excavations
at the McKeithen site in north Florida demonstrated that burial
mound building and elaborate pottery production probably
extended from A.D. 300-700 (Milanich et al. 1997:186).

Sacred, Seculal; or Both

Long ago William Sears (1973) came up with the idea of a
dichotomy in the prehistoric Southeast between the sacred and
the secular. This meant, most immediately, that the ceramics
in mounds were "ceremonial" and sacred, in other words,
fancier (more highly decorated, better made), not to mention
imbued with special meaning, while those in domestic areas
were plainer, less important, for everyday use. Why this idea
has hung around for so long is a mystery. The evidence from
Moore's mound explorations published 70 years before Sears
wrote this fails to support the idea. Brose (1979:142) long
ago and others more recently (e.g., Tesar and Jones 2009:22)
have pointed out how many of Moore's works mention plain,
undecorated, "inferior" ceramic wares he came across in the

WHITE

GOTIER HAMMOCK MOUND AND MIDDEN

THE FLORIDA ANTHROPOLOGIST 2010 VOL. 63(3-4)

many mounds he dug; these ceramics were so mundane he
rarely found them even worth describing.
Gotier Hammock mound provides more evidence that
we should rethink the sacred-secular concept. Unadorned,
poorly made or otherwise unremarkable pots were recovered
by Floyd and Moore and by our investigations a century later
in the mound. The fanciest pots the Basin Bayou Incised
jar, the Weeden Island Incised and Swift Creek Complicated-
Stamped sherds as well as the plain bowl all had irregular,
dirty-looking, baked-on deposits, suggesting their use for
cooking (and people not washing dishes afterwards!). Other
mounds in this region have also produced fancy Middle
Woodland ceramics with baked-on black deposits (e.g., Tesar
and Jones (2009:240). Of course the cooking could be for
special occasions such as burial ritual, but it might have been
the actual use, not the pot design, that made it meaningful.
Since the majority of the pots recovered from Gotier
Hammock mound, plain or not, have kill holes in the base, the
suggestion is that all these vessels were part of some (probably
funerary) ritual and that non-material considerations led to their
being sacred, not just or not necessarily any fancier decoration.
Plainness or ugliness or any other characteristic we bestow
upon these pots is a judgment based on our own esthetic (and
other) standards. The sacredness of a pot or any other artifact
may have depended on characteristics completely unknowable
in the material record, not scientifically discernible- just like
holy water is indistinguishable from any other water and only
considered sacred because of religious ritual associated with
it, only materially identifiable as holy because it is found in
special containers, and so forth.
Stephenson and Smith (2008) suggest that Middle
Woodland vessels in mounds are perhaps more associated
with service and individual consumption, then killing and
caching apart from burials, whereas domestic-use vessels
may indicate more a utilitarian function such as cooking.
But the high amount of cooked-on sediment on some of the
Gotier pots suggests such utilitarian use. Perhaps the shell
columella tools had some ritual importance, but they also look
very utilitarian. Many were recovered from the mound, what
should be a sacred site. But others came from the midden area,
and all resemble typical artifacts from habitation sites all over
the St. Joe Bay area.
Many traditional societies, including Native American
groups, did/do not separate in their minds or activities their
ideology from their perceptions and understandings of
everything else in daily life. Spirituality pervades many
cultures all day long. Even in secular, western society, there
are groups (in monasteries, convents, religious schools) and
individuals for whom every act, practice, and thought is
part of living in a spiritual as well as a material world. The
material record does not reflect this, of course, since there are
churches and places of worship and then there are domestic,
secular areas, though sometimes religious artifacts reside in
the domestic zone too. We cannot know if spirituality and a
concern for the sacred permeated Middle Woodland society or
was only a part of life during mound ritual or other important
ceremonies, since "emic" archaeology is impossible when
dealing with prehistoric peoples in the very distant past.

Culture Process

The remoteness of the lower Apalachicola delta may have
continued into later prehistory and even historic times. While
Late Woodland peoples in the riverine interior were beginning
horticulture, then intensifying it to become agriculturalists
during the Fort Walton times that followed, contemporaneous
coastal and estuarine sites show no evidence of food production
but only continual dependence upon aquatic resources (White
1994, 2005). Middle Woodland may have lasted as late as it
did at Gotier Hammock because people found little reason to
change.
That burial mound building possibly lasts longer on the
coast than upriver in the Apalachicola delta and other parts
of the interior Gulf Coastal Plain is a testable hypothesis. The
coast, especially away from river and stream mouths, is more
distant, possibly less accessible, lower in elevation, often
inundated. Overland trails or other routes may wash away
more rapidly in lower ground, and small streams that were
also communication pathways may dry up seasonally or for
multi-year periods. Inland riverine cultures may have devoted
more time later in the Middle Woodland or Late Woodland
to horticulture. But coastal salty air and poor, white sandy
soils are not good for farming. Food production is more work
that may have been unnecessary for people who got all they
needed by just sitting in the boat with a net or wading into
shallow water (or sending the kids to get it). People inland
took advantage of fertile, annually-renewed alluvial soils and
the particularly productive crop called maize by Late Weeden
Island times. For poorly understood reasons they intensified
subsistence activities and became true agriculturalists. Coastal
groups maintained full-time collection of wild resources.
Whether this resulted in different sociopolitical organization
for them, as compared with their gardening and farming
cousins inland, is still unknown, though of course a topic of
continual investigation.

Summary and Contributions of this Work

Northwest Florida Archaeology

The season and a half of field investigations at Gotier
Hammock and the additional intermittent visits, lab work, and
other research over the past nine years provide what I hope are
some useful contributions to the knowledge of the archaeology
of northwest Florida and the Southeast in general.
Lost Mound Found. A lost C. B. Moore mound has been
investigated in detail, the first in Gulf County. Though Gotier
Hammock mound is nearly gone, some of it remains. Re-
excavation and reanalysis are important in archaeology, and
collections data (online or otherwise) can add hugely to the
record of what we thought we knew.
Middle Woodland in the Apalachicola Delta. The solid
date on this mound helps define the time frame for Middle
Woodland in the region. The data confirm the tight association
of Swift Creek and early Weeden Island ceramics in the whole
definition of Middle Woodland here.

THE FLORIDA ANTHROPOLOGIST

2010 VOL. 63(3-4)

WHITE GOTIER HA~1ocK MOUND AND MIDDEN

Big Shell on St. Joseph Bay. St. Joseph Bay-area typical
shoreline shell middens are piles of large gastropod shells, in
addition to oysters and other species. They illustrate a good
example of how ecological-functional explanations still work
well in archaeology. People usually (but not always) do what
is most convenient: saltwater shellfish are what's for dinner
(and toolmaking) because of the atypical bay. On other bays
in the region, oyster, which tolerates less salt, predominates
at shell middens. Farther inland in estuarine settings Rangia
(marsh clam) shell predominates, a species that needs more
fresh water, such as at stream mouths.
The artifact data from Gotier Hammock, both mound and
midden, follow the St. Joe Bay pattern with large-gastropod-
shell tools. However, the bayshore midden is atypical, with
no conch or whelk ecofacts. Perhaps people did not stay long
enough to gather these big gastropods to eat, but just came for
brief burial ceremonies. Or perhaps they did not want them
for food (to me, eating conch is like eating shoe leather), or
they came at the wrong season to get them, or at a time when
these resources were less available because the bay was less
salty. Probably combinations of these reasons were in effect
for Middle Woodland and later peoples.
Prehistoric and Protohistoric on St. Joseph Bay. Besides
the Middle Woodland component, this work documents later
occupation on the bayshore near Gotier Hammock. Though no
diagnostic Fort Walton ceramics were recovered, the plain and
check-stamped sherds could easily be associated with a Fort
Walton camp during the 1300s. The later date probably means
additional Fort Walton; groups with this material culture lasted
in the region until about 1700 (Marrinan and White 2007).
Sacred May Be Ugly. The Gotier Hammock mound data
do not support the once-sacred idea of a "sacred-secular"
dichotomy in ceramic types. The mound's many plain ("ugly")
pots were, according to the original definition, in sacred
contexts. Modern esthetic values lead us to see intricately-
patterned incised and punctated or stamped pots as being
something special for the grave, but material correlates for
this are not evident at Gotier Hammock. Such concepts may
indeed have existed in past people's conceptions of ritual and
spirituality. But the mound data described here suggest that
it was not what they had, but what they did with whatever
they had that counted. Sacred may not be beautiful or even
specially-made, but just whatever is used in the important
ceremony that then gives it new meaning.

Public Archaeology'

Modem culture process at Gotier Hammock mound
is also worth noting. This research shows that, even after
disturbance by historic occupants, erosion, ill treatment by
nineteenth-century looters, "demolishing" in 1902 by C. B.
Moore, twentieth-century disturbance by pine-planting and
harvesting machinery, and additional attention from recent
collectors and indignities from modern looters, a site such as
this mound retains small intact portions that constitute enough
for scientific investigation.
Professional archaeologists are not the only ones searching
for Moore's mounds. The Apalachicola-lower Chattahoochee

valley region is so rich in prehistoric sites and what many
consider beautiful pottery and other elaborate artifacts that
Moore came back several times during his decades of work,
and thousands of other collectors have been active for well
over a century. Even with modern insults and injuries, sites
are not necessarily completely destroyed. Furthermore, they
can be relocated far more easily with the help of those who
live in the area, perhaps collect artifacts, know and use the
land, and graciously share their knowledge with professionals
(White 2008b).
This project resulted from a happy collaboration of
professional and student archaeologists, the private landowner,
local collectors, public and private land managers, and other
area residents. More than just benefitting science, such public
archaeology also helps tell the story of people from the distant
past who inhabited and enjoyed the St. Joe Bay region that is
so beloved today.

Acknowledgments

Many are thanked for their help in this research. The St.
Joe Company first reported the artifacts uncovered by their own
firebreak plowing; they later graciously gave me permission
for fieldwork. Grant support was provided by the Friends of
the St. Joseph Bay Buffer State Preserve and USF archaeology
alumna Dorothy Ward. The Preserve also gave us crew
lodging and other support. Preserve and Apalachicola National
Estuarine Research Reserve staff Jean Huffman, Jimmy Moses,
Pat Millender, and Neil Jones helped with planning, mapping,
digging, and other tasks. Student crew members Kristen
Braun, Jenna Clevinger, Nancy Fairchild, Joe Frezza, Kevin
Hageman, Ryan Harke, Jake Jones, Caitlin Kelley, Stephanie
Lonergan, Bart McLeod, Erik Palm, Maura Peterson, Erin
Rosenthal, Liz Usherwood, and graduate student crew chief
Elicia Kimble were excellent fieldworkers, even under rough
conditions. Kimble, Fairchild, Lonergan, Rosenthal, and Jana
Futch processed materials in the lab over long hours of the
academic year. Graduate students Julie Rogers and Jeff Du
Vernay provided additional research data and Du Vernay read
the manuscript to find problems. USF archaeologist Christian
Wells and student Kara Rothenberg studied soil deposits on
the pottery. Gulf County experts Herman Jones and Wayne
Childers shared historical background material on the region.
Offspring Tony White made the site map. Archaeologist
Mike Russo and colleagues at the National Park Service in
Tallahassee shared data on their excavations of similar sites at
Tyndall Air Force Base west of Gulf County.
Keith Stephenson of the South Carolina Institute of
Archaeology and Anthropology, Savannah River Site, provided
the AMS radiocarbon dates, which were done at Beta Analytic,
Inc., in Miami. At the Smithsonian Institution National
Museum of the American Indian, Supervisory Collections
Manager Patricia Nietfeld, Photo Archivist Lou Stancari, and
Supervisory Photographer Cynthia Frankenburg provided
information on Moore's artifacts curated there and the photo
in Figure 13. At the Smithsonian National Museum of Natural
History, Archaeological Collections Specialist James Krakker
provided data on artifacts in the Floyd collection.

WHITE

GOTIER HAMMOCK MOUND AND MIDDEN

THE FLORIDA ANIHROPOLOGIST 2010 VOL. 63(3-4)

References Cited

Benchley, Elizabeth D., and Judith A. Bense
2001 Archaeology and Historv of St. Joseph Peninsula
State Park. Phase I Investigations. University
of West Florida Archaeology Institute Report of
Investigations No. 89. Pensacola.

Beranger. Jean
1718 Plan de la Bare de St. Joseph tire par Jean Beranger
cc le. May 1718. Bibliotheque Nationale de France,
Department des Cartes et Plans. Accessed online
December 2009 as record #7965, Newberry Library
Cartographic Catalog, at http://www.biblioserver.
com/newberry.

Brose, David S.
1979 An Interpretation of Hopewellian Traits in Florida. In
Hopewell Archaeology, The Chillicothe Conference,
edited by D. Brose and N. Greber, pp. 141-149. Kent
State University Press, Kent, Ohio.

Brose, David S. and Nancy Marie White (editors)
1999 The Northwest Florida Expeditions of Clarence
Bloomfield Moore. University of Alabama Press,
Tuscaloosa.

Cologne Radiocarbon Calibration & Paleoclimate Research
Package
2007 CalPal Online at http://www.calpal-online.de/,
accessed 22 September 2009.

Davis, Richard A., Jr.
1997 Geology of the Florida Coast. In The Geology of
Florida, edited by A. Randazzo and D. Jones, pp.
155-168. University Press of Florida, Gainesville.

Donoghue, Joseph R., and Nancy Marie White
1995 Late Holocene Sea Level Change and Delta
Migration, Apalachicola River Region, Florida.
Journal of Coastal Research 11 (3): 651-663.

Dumas, Ashley A.
2008 New Data for the Middle Woodland Period on the
Alabama Gulf Coast. Poster presentation at the
annual meeting of the Southeastern Archaeological
Conference, Raleigh, NC, November.

Eyles, Eric
2004 Prehistoric Shell Artifacts fivm the Apalachicola
River Valley Area, Northwest Florida. M.A. thesis,
Department of Anthropology, University of South
Florida, Tampa. Online at USF library at http://etd.
fcla.edu/SF/SFE0000498/eeylesthesis.pdf.

Fewkes, Jesse Walter
1924 Preliminary Archaeological Explorations at

Weeden Island, Florida. Smithsonian Miscellaneous
Collections 76 (13): 1-26.

Frashuer, Anya C.
2006 Middle Woodland Mound Distribution and
Ceremonialism in the Apalachicola Valley, Northwest
Florida. M.A. thesis in anthropology, University
of South Florida, Tampa. Online at USF library at
http://digital.lib.usf.edu:8080/fedora/get/usfldc:E14-
SFE 0001504/DOCUMENT.

Gulf County Golden Anniversary Commission
1975 Golden Anniversary' Celebration, Gulf County,
Florida. Great American Publishing Company,
Chattanooga, Tennessee.

Herring, Hal
2009 The Panhandle Paradox. Are the St. Joe Company's
Development Plans....... Miller-McCune magazine,
September-October 2009. Miller-McCuneCenter for
Research, Media and Public Policy, Santa Barbara,
California.

Holmes, William Henry
1903 Aboriginal Pottery of the Eastern United States.
Bureau of American Ethnology 20'" Annual Report,
1898-99:1-237. Smithsonian Institution, Washington,
D.C. Reprinted in The Archaeology of William Henry
Holmes, edited by D. Meltzer and R. Dunnell, pp.
466-666. Smithsonian Institution Press, Washington,
D.C., 1992.

Jones, Herman
2002 The Voyage of the Gopher Uncovering the Secret of
the Mounds. Panhandle Beacon 4 (8):1, 4, 6, 8, 14.
(March 2002).

Luer, George M. (editor)
1986 Shells and Archaeology in Southern Florida. Florida
Anthropological Society Publications No. 12, Special
issue: The Florida Anthropologist 39 (3, Part 1).

Marquardt, William H.
1992 Shell Artifacts from the Caloosahatchee Area. In
Culture and Environment in the Domain of the
Calusa, edited by W. Marquardt, pp.191-221.
Monograph No. 1, Institute of Archaeology and
Paleoenvironmental Studies, University of Florida,
Gainesville.

Marrinan, Rochelle A., and Nancy Marie White
2007 Modeling Fort Walton Culture in Northwest Florida.
Southeastern Archaeology 26(2):292-318.

Martin, Alexander C. and William D. Barkley
1961 Seed Identification Manual. University of California
Press, Berkeley.

THE FLORIDA ANTHROPOLOGIST

2010 VOL. 63(3-4)

WHITE GOTIER HAMMOCK MOUND AND MIDDEN

Mathews, Marguerite Marree
1998 Floyd, History and Lineage. Charles Floyd, John
Floyd of Virginia, South Carolina, and Camden
County, Georgia. Online document accessed 9-29-
2009 at http://www.camdencounty.org/floyd.pdf.

Mayo, Karen L.
2003 Archaeological and Historical Investigations on
Black' Island, Gulf County, Florida. M. A. thesis,
Department of Anthropology, University of South
Florida, Tampa.

Meehan, Betty
1982 Shell Bed to Shell Midden. Humanities Press, Atlantic
Highlands, New Jersey.

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

Milanich, Jerald T., Ann S. Cordell, Vernon J. Knight Jr.,
Timothy A. Kohler, and Brenda J. Sigler-Lavelle
1997 Archaeology of Northern Florida, A.D. 200-900:
The McKeithen Weeden Island Culture. University
Press of Florida, Gainesville. Originally published
as McKeithen Weeden Island, University of Florida
Press, Gainesville, 1984.

Moore, Clarence B.
1902 Certain Aboriginal Remains of the Northwest Florida
Coast, Part II. Journal of the Academy of Natural
Sciences 12:127-358. Philadelphia.
1903 Certain Aboriginal Mounds of the Apalachicola
River. Journal of the Academy of Natural Sciences
12:441-492. Philadelphia.

Phillips, Philip, James A. Ford, and James B. Griffin
1951 Archaeological Survey in the Lower Mississippi
Alluvial Valley, 1940-1947. Papers of the Peabody
Museum of American Archaeology and Ethnology
25. Harvard University, Cambridge, Massachusetts.
Reprinted by the University of Alabama Press,
Tuscaloosa, 2003.

Porter, Louise M.
1975 Lives of St. Joseph. Great American, Chattanooga,
Tennessee.

Price, Sarah E.
2008 Phase IIlArchaeology atPlash Island, Archaeological
Site 1BA134, in Baldwin County, Alabama. Center
for Archaeological Studies, University of South
Alabama, Mobile.

Rogers, Julie H.
2009 Archaeology of the Early Eighteenth-Century Fort
San Jose in Northwest Florida. Paper presented at

the annual meeting of the Florida Anthropological
Society, Pensacola.

Rupert, Frank R.
1991 Geology of Gulf'County, Florida. Florida Geological
Survey, Tallahassee.

Russo, Mike
2009 Presentation on archaeology at Tyndall Air Force
Base at the symposium entitled Rethinking Weeden
Island: A Pottery Potlatch. February 2009, Weedon
Island Preserve Cultural and Natural History Center,
St. Petersburg, Florida.

Schuster, Joseph N., Kenneth W. Monroe, Leland D. Sasser,
Robert E. Evon, Henry J. Ferguson, Matthew W. Havens,
Robert N. Pate and Dale G. Sprankle
2001 Soil Survey of Gulf County, Florida. U.S. Department
of Agriculture, Natural Resources Conservation
Service, Washington, D.C. Accessed online on 4
October 2009 at http://soildatamart.nrcs.usda.gov/
Manuscripts/FL045/0/Gulf.pdf.

Sears, William H.
1973 The Sacred and the Secular in Prehistoric Ceramics.
In Variations in Anthropology: Essays in Honor
of John C. McGregor, edited by D. Lathrap and J.
Douglas, pp. 31-42. Illinois Archaeological Survey
Publications, Urbana.

Shanks, Jeffrey H.
2009 Baker's Mound (8By29) a Swift Creek Sand
Mound and Ring Midden. Paper presented at the
annual meeting of the Southeastern Archaeological
Conference, Mobile, Alabama, November.

Shores, Elizabeth Findley
2008 On Harper s Trail: Roland McMillan Harper,
Pioneering Botanist of the Southern Coastal Plain.
University of Georgia Press, Athens.

Stephenson, Keith, and Karen Y. Smith
2008 Middle Swift Creek/ Weeden Island I Ceremonialism
in the Interior Coastal Plain of Georgia. Poster
presentation at the annual meeting of the Society
for American Archaeology in Vancouver, British
Columbia, Canada.

Tesar, Louis D., and B. Calvin Jones
2009 The Waddells Mill Pond Site (8Ja65): 1973-74 Test
Excavation Results. Florida Department of State,
Bureau of Archaeological Research, Tallahassee.

University of Alabama
2010 Department of Geography, Alabama Maps, Historical
Map Index. Accessible online at http://alabamamaps.
ua.edu/index.html.

GOTIER HAMMOCK MOUND AND MIDDEN

WHITE

THEFLRID ATHRPOOGIT 010VO. 6(34'

Waselkov, Gregory A.
1987 Shellfish Gathering and Shell Midden Archaeology.
In Advances in Archaeological Method and Theory,
edited by M. Schiffer, pp. 93-210. Academic Press,
New York.

Weddle, Robert S.
1991 The French Thorn. Rival Explorers in the Spanish
Sea, 1682-1762. Texas A&M University Press,
College Station, Texas.

White, Nancy Marie
1991 Woodland Ceramics and Subsistence in the Middle
Apalachicola Valley, Northwest Florida. Paper
presented at the Southeastern Archaeological
Conference, Jackson, MS.
1992 The Overgrown Road Site (8Gu38): A Swift Creek
Camp in the Lower Apalachicola Valley. The Florida
Anthropologist 45:18-38).
1994 Archaeological Investigations at Six Sites in the
Apalachicola River Valley, Northwest Florida.
NOAA Technical Memorandum NOS SRD 26,
National Oceanic and Atmospheric Administration,
U.S. Department of Commerce, Washington, D.C.
2005 Archaeological Survey of the St. Joseph Bay State
Buffer Preserve, Gulf County, Florida. Report to the
Apalachicola National Estuarine Research Reserve
and the Division of Historical Resources, Tallahassee.
Department of Anthropology, University of South
Florida, Tampa.
2008a Archaeological Investigations at the Gotier
Hammock, 8Gu2 Gulf County, Florida Summary
Report to the Friends of the St. Joseph State Buffer
Preserves and the St. Joe Company. Department of
Anthropology, University of South Florida.
2008b Looking for Clarence, or, How We Find C.B. Moore's
Lost Mounds. Paper presented at the annual meeting
of the Southeastern Archaeological Conference,
Charlotte, NC, November.
2010 Archaeological Investigations at Gotier Hammock
Mound and Midden (8Gu2) on St. Joseph Bays
Northwest Florida. Report to the St. Joe Company,
the St. Joseph Bay State Buffer Preserve, the
Apalachicola National Estuarine Research Reserve,
and the Division of Historical Resources, Tallahassee.
Department of Anthropology, University of South
Florida, Tampa.

White, Nancy Marie, and Mary Beth Fitts
2001 Richardson s: Hammock Site, 8GulO, Gulf County,
Florida: Report of the 2000 Archaeological
Investigations. Report to the Apalachicola National
Estuarine Research Reserve and the Division of
Historical Resources, Tallahassee. Department of
Anthropology, University of South Florida, Tampa.

White, Nancy Marie, Nelson D. Rodriguez, Christopher
Smith, and Mary Beth Fitts
2002 St. Joseph Bay Shell Middens Test Excavations, Gulf
County: Florida, 2000-2002. Report to the Florida
Division of Historical Resources, Tallahassee.
Department of Anthropology, University of South
Florida, Tampa.

White, Nancy Marie, Terrance L. Simpson, and Suella
McMillan
1992 Apalachicola Valley Archaeology. Booklet published
by the W.T. Neal Civic Center, Blountstown, Florida.
Reprinted 1998 by the Apalachicola National
Estuarine Research Reserve, Apalachicola, Florida.

Willey, Gordon R.
1949 Archaeolog, of the Florida Gulf Coast. Smithsonian
Miscellaneous Collections 113. Reprinted by the
University Press of Florida, Gainesville, 1999.
1966 An Introduction to American Archaeology. Volume
1, North and Middle America. Prentice-Hall, Inc.,
Englewood Cliffs, New Jersey.

Wimberly, Steve B.
1960 Indian Pottery from Clarke County and Mobile
County, Southern Alabama. Alabama Museum of
Natural History Museum Paper 36, University,
Alabama.

Ziewitz, Kathryn, and June Wiaz
2004 Green Empire. The St. Joe Company and the
Remaking of Florida's Panhandle. University Press
of Florida, Gainesville.

THE FLORIDA ANTHROPOLOGIST

2010 VOL. 63(3-4)

2010 FLORIDA FIELD SCHOOL SUMMARIES

2010 New College of Florida Survey of Galilee Cemetery:
Sarasota, Florida
by Uzi Baram

Combining experiential learning and community service,
the Survey of the Galilee Cemetery: Race, History, and
Community project began in February 2010 and is an ongoing
program in civic engagement in Sarasota, Florida. Ever since
James Deetz (1996) seriated the motifs on the grave stones of
New England, historical archaeologists have used cemeteries
for experiential studies of material culture. While studying
cemeteries for their material histories, the challenges faced by
burial grounds have become clear; recently newspaper stories
across the United States (e.g., Vitello 2009, Charles 2009)
note that community-managed cemeteries face a crisis of
maintenance as an older generation that cared for the grounds
pass on. Sarasota has many cemeteries with great significance
to members of the local community and with regard to history
and heritage. Two such cemeteries are community-owned
because of the legacy of segregation.

It was regarding one of these cemeteries that I was
contacted by the City of Sarasota in April 2009. The City
was interested in having New College conduct a new survey
of the Rosemary Cemetery, located just north of Sarasota's
Newtown neighborhoods (see Case Studies for Preservation
and Protection of Historic Cemeteries in Florida n.d.). The
cemetery is owned by the Woodlawn/Galilee Cemetery
Restoration Task Force, a community-based non-profit
organization that is dedicated to the respectful maintenance of
the burial grounds. I began the project by attending meetings
of the Task Force and proposing an above-ground survey of
the ceremony in the spirit of collaborative research. The goals
of research at the Galilee Cemetery intertwine anthropological
research, community education, and historic preservation.
In January 2010, the Woodlawn/Galilee Cemeteries
Restoration Task Force decided to close the Galilee Cemetery,
determining that purchased deeds will be honored but no new
plots will be offered (Cox 2010). Consequently, January 2010
marked an interesting transition, previously undocumented in
the literature on heritage. Thus, with a particular interest in

-

Figure 1. The Woodlawn/Galilee Cemetery Restoration Task Force and the Survey Team at the cemetery February 2010.
Photograph by the author.

VOL. 63(3-4)

SEPTEMBER DECEMBER 2010

THE FLORIDA ANTHROPOLOGIST

THE LORDA ATHRPOLGIST201 VOL 633-4

Figure 2. Liz Usherwood and Chelsea Montgomery, of New College of Florida, recording
grave markers. Photograph by the author.

observing the transition from an active to an inactive burial
ground, the survey of the Galilee Cemetery is focused on
documenting the grave markers in support of the Task Force's
goals of conservation, exploring the history of the cemetery as
an entry to the heritage of the surrounding communities, and
beginning the process of historic preservation.
From February to May 2010 and from October to
December 2010, I trained and supervised more than a dozen
students from New College of Florida and State College of
Florida at the Galilee Cemetery in historical preservation, with
field directors Kacie Allen (spring 2010), James Birmingham
and Monica Tambay (fall 2010), and with the support from the
Division of Social Science at New College. The goal of the
survey of the Galilee Cemetery is to reveal, document, and
organize the cemetery's history by studying the grave markers,
conducting archival research, and collecting oral histories.
Students record grave markers through written descriptions,
sketches, and photographs, noting the inscriptions, motifs,
marker material, grave condition, and spatial relationship to
other graves and landmarks, in an ongoing process of research
and representation of the historic cemetery.
The survey of the Galilee Cemetery is an example of
community education and public archaeology (even though
there are no excavations), and an opportunity to contribute
to the rich heritage of the region. Beyond the important
pre-Columbian burial grounds in Sarasota, the city also has
several historic cemeteries: the Whitaker Pioneer Cemetery
was established in 1879 and the Rosemary Cemetery in 1886.
As Annie McElroy (1986:121) noted, using the terminology
of the early twentieth century, in 1905 land near Overtown
was sold to the trustees of a "colored cemetery" that would

become the Oakland Cemetery (also known as the Woodlawn
Cemetery). When the African-American community moved
northward to Newtown in the 1930s, the Galilee Cemetery
was established, on what is now U.S. Highway 301 between
Myrtle Road and Dr. Martin Luther King, Jr. Way. While
only a few dozen inscriptions have been recorded by local
genealogical societies, the cemetery contains hundreds of
burials. The original plat has not been maintained, making
collecting information in a systematic manner a benefit to both
the Task Force and to the documentation of Sarasota's history.
Annie McElroy's 1986 volume is the only published
history of the African-American community in Sarasota,
a rich heritage that has only started to be connected to the
overall regional history. Since the first burials at the Galilee
Cemetery were interred in the 1930s, it is interesting to note a
contemporary description in the WPA Guide to Florida (Federal
Writers Project 1984:269): "The local Negro settlement, east
of the railroad, has it shops, churches, recreation centers, and
rows of shacks. The majority of inhabitants, 30 per cent of
the city's total population, are engaged in agricultural pursuits,
and a few find employment as hostlers and roustabouts with
the circus, returning to Sarasota in the fall to pick up odd
jobs in canning factories, packing houses, and as gardeners."
The cemetery is the resting place for many of the people who
lived and worked in Sarasota and its history can contribute
to understanding and commemorating the lives of those
described. But as the description suggests, the subject of race
haunts the history of Sarasota.
For the students, the survey of the Galilee Cemetery has
three pedagogical layers. The first is learning the techniques of
historic preservation and conducting the level of detailed work

THE FLORIDA ANTHROPOLOGIST

2010 VOL. 63(3-4)

2010 FIELD SCHOOL SUMMARIES 185

Figure 3. Kacie Allen (University of Florida) and Michael
Waas (New College of Florida) recording grave markers.
Photograph by the author.

required in anthropological research. The second is exploring
the challenging issues surrounding the history and legacies
of segregation in Sarasota while engaging with community
members. The third focuses on the materialization of memory,
the anthropological concern for material culture, community,
and commemoration; through participant-observation, they
have been part of the commemoration of the cemetery. By
spending hours in the cemetery recording each and every
grave marker, reflecting on the place and its meanings for
descendants and other family members of those buried at the
Galilee Cemetery, and recognizing the legacies of race, the
students are producing a record of the cemetery and wrestling
the legacies and longer-term social implications of the past
(Baram 2009). The collaboration with the Task Force has
encouraged several of the students to document the beauty
of this forgotten history through photographs, drawings, and
descriptions.
Grave markers are still being mapped and information
from the survey is being entered into a database; at this time no
research conclusions on the materialization of memory at this
location have been reached yet. The public outreach has been
successful, including a Sarasota Public Schools' Education
Channel half-hour video titled "Galilee Cemetery Beauty
in a Forgotten Place" (Stocker 2010) and plans for further
outreach as the survey concludes. But even at this stage of the
research, a place easily missed while driving on Highway 301
is providing insights into Sarasota's history.

References Cited

Baram, Uzi
2009 "Learning Service and Civic Engagement: A Historic
Cemetery as a Site for Grappling with Community,
Politics, and Commemoration" In Archaeology and
Community Service Learning, edited by Michael
Nassaney and Mary Ann Levine, pages 110-121.
University Press of Florida, Gainesville.

Figure 4. A grave marker from the Galilee Cemetery,
Sarasota. Photograph by the author.

Case Studies for Preservation and Protection of Historic
Cemeteries in Florida
n.d. Florida Division of Historic Resources. http://
www.flheritage.com/archaeology/cemeteries/index.
cfm?page=Case_Studies Accessed November 28,
2010

Charles, Nadege
2009 Lemon City Cemetery Awarded Historic Designation
from Miami. The Miami Herald http://www.
bradenton.com. Accessed December 20, 2009.

Coz, Billy
2010 Interments Suspended at Two Historic Cemeteries.
Sarasota Herald Tribune January 21, 2010 http://
www.heraldtribune.com/article/2010012 1/
ARTICLE/1211050. Accessed January 21, 2010.

Deetz, James
1996 In Small Things Forgotten. Anchor, New York.

Federal Writers Project
1984 WPA Guide to Florida: The Federal Writers' Project
Guide to 1930s Florida. Pantheon, New York.

McElroy, Annie M.
1986 But Your World and My World: The Struggle for
Struggle: A Partial History of Blacks in Sarasota
County, 1884-1986. Black South Press, Sarasota.

Stocker, Kimberly
2010 Galilee Cemetery: Beauty in a Forgotten Space.
Education Channel, Sarasota.

Vitello, Paul
2009 With Demise of Jewish Burial Societies, Resting
Places Are in Turmoil. New York Times August 3,
2009 http://www.nytimes.com/2009/08/03/myregion
/03bury.html? r=l. Accessed August 3, 2009.

L AT G2010 VOL. 630-41

2010 The Ohio State University Field School Excavations:
Fort Center, Glades County, Florida
by Hannah R. Morris and S. Margaret Spivey

The Fort Center archaeological site (8GL13) lies at the
southwestern edge of Lake Okeechobee, along Fisheating
Creek in South Florida. A persistent place on the landscape,
the site has a demonstrated occupation lasting from the Late
Archaic period through the historic period (Sears 1982). The
site is named for the fort that was first constructed at the
location during the second Seminole War. Native Americans
during the prehistoric era constructed large-scale earthworks
of nearly every Belle Glade culture type, including a charnel
pond, linear earthworks, and at least fifteen mounds. One of
the most distinctive features of the site is a circular earthwork
365 meters in diameter called "the Great Circle."
Major excavations were previously carried out by
Fairbanks (1966-1967) and Sears (1966-1971). Fairbanks's
work remains largely unreported, while Sears (1982)
published a major monograph on the site. Sears's often cited
but controversial explanation for key features of the site
proposes an early date for the adoption of domesticated maize
in Florida.
In summer 2010, our crew excavated four test units and
performed resistance, ground penetrating radar (GPR), and
topographic surveys across various portions of the Great Circle.
This work was completed through a field school conducted
through the Ohio State University (OSU) and supplemented
by a grant from the National Geographic Society. Both the

field school and grant were directed by Principal Investigator
(PI) Dr. Victor Thompson (OSU). Dr. Thomas Pluckhahn
(University of South Florida) served as the co-PI. Amanda
Roberts Thompson was the assistant field director, while
Hannah Morris (OSU) and S. Margaret Spivey (Washington
University in St. Louis) served as graduate student assistants.
Our first test unit re-excavated a portion of Fairbanks's
excavation, positioned in the western edge of the Great Circle.
Test Units 2 and 4 were placed in midden contexts adjacent to
Fisheating Creek, while Test Unit 3 was located on the edge
of the ditch that comprises the Great Circle. The midden units
(2 and 4) were particularly rich in faunal material. Rather than
attempting to field-sort artifacts from 1/8" screen, any debris
remaining in the screen (i.e., roots, leaves, unidentifiable
material) after large artifacts were removed was bagged to be
sorted in the lab. Archaeology students are currently sorting
these remains in the OSU Archaeology Lab. This work is
revealing an astonishing amount of faunal remains, as well as
many other small artifacts, including numerous sharks' teeth.
The excavation of Unit 3 provided a detailed profile of the
construction of the Great Circle in reference to the occupational
sequence at the site. Finally, a 25-by-25-cm column unit was
excavated directly adjacent to test units two, three, and four
to provide samples for microbotanical and macrobotanical
analysis.
The resistance survey, using a RM15 resistivity meter,
was performed over portions of the Great Circle and
surrounding areas. Despite several non-fatal stabbings in the
foot with the equipment and the intense heat, our crew worked

Figure 5. Seth Osnowitz and Josh Pammer perform resistance survey near the Seminole Fort. Photo by S. Margaret Spivey.

THE FLORIDA ANTHROPOLOGIST

0 102 Vot. 63(3-4)

2010 FIELD SCHOOL SUMMARIES

Figure 6. Brittan Krichbaum excavates Unit 1. Photo by
Rob Eichorst.

tremendously hard. We were able to survey approximately
14,000 square meters. The northernmost portion of the Great
Circle was surveyed, including both eastern and western edges
of the feature. In addition, a large area covering the inner and
southern outer edge of the Great Circle was also surveyed.
These investigations revealed that the current feature is
actually one of four to be constructed at the site. Previously,
it had been hypothesized that the Great Circle was the third
feature of its kind at the site.
While performing the resistance survey over the
northwestern portion of the circle, an interesting pattern of
results led us to investigate the area further with GPR. We
surveyed three 20-by-20-m blocks with the GPR, and believe
we have located the corner of the Seminole Fort that was
constructed at the site during historic times. An 1853 quarter,
dating to the occupation of the fort, was also recovered from
the surface in this area.

The materials recovered from the test unit excavations
(bone, lithics, ceramics, and historic artifacts) are currently
being sorted in the Laboratory of Archaeology at OSU.
Thompson is currently analyzing the ceramics from Fort
Center. He has found ceramics dating from the Late Archaic,
Belle Glade, and historic periods. In addition, Morris has
begun processing column samples from Units 2, 3, and 4 for
macrobotanical analysis. Thus far, she has identified pokeberry
and goosefoot.
Column samples for microbotanical analysis have been
submitted to PaleoResearch Institute for processing, and we
are awaiting results. Materials for radiocarbon dating have also
been submitted, and these dates reflect a period of occupation
from the Early Woodland into the historic period.
We look forward to the results of these analyses, and the
insights we will gain concerning the history of Fort Center. We
would like to thank the staff of the Fisheating Creek Wildlife
Management Area for their support, the undergraduate
students who participated in the field school- Chris Goodrich,
Alexander Karaffa, Ben Keller, Brittany Krichbaum, Ryan
Oergcl, Seth Osnowitz, Josh Pammer, Diana Sickles, Juli
Six, Hannah Solomon, Kimi Swisher, Krissy Warne, and
Natasha Zientek- for their hard work, as well as Rob Eichorst
and Theresa Schober for facilitating our exciting adventures
to nearby archaeological sites. Without the whole-hearted
excitement of everyone who was involved, our fieldwork at
Fort Center this season would not have been possible. And
there would be many, many more saw palmettos remaining at
Fort Center.

Reference Cited

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

Figure 7. The 2010 OSU Field School field trip to the Big Mound City site.

THE FLORIDA ANTHROPOLOGIST 2010 VOL. 63(3-4)

2010 University of Florida St Johns Archaeological Field
School:
Silver Glen Run, Lake and Marion counties, Florida
by Kenneth E. Sassaman

The freshwater shell deposits along Silver Glen Run
in Lake and Marion counties, Florida has been the summer
venue of the University of Florida's St. Johns Archaeological
Field School since 2007. When Jeffries Wyman visited the
area in the early 1870s, he observed a massive, U-shaped
accumulation of shell where Silver Glen meets the southwest
corer of Lake George, along with an "amphitheater" of shell
encircling the spring boil 600 m to the west. Both deposits
were commercially mined in the early twentieth century,
but subsurface aspects of both have survived, and a series
of related shell-bearing sites remain largely unmolested.
Collectively, the mounds, ridges, shell middens, and feature
assemblages of Silver Glen Run entail a remarkable record
of ancient human experiences spanning 6000 to 4000 years
ago (Figure 8). Field school efforts aim to document the social

and ecological circumstances attending the deposition of shell,
notably the series of cultural transformations coincident with
changes in the purpose, timing, and magnitude of deposition.
The results of prior field schools-including a five-year stint
on and around Hontoon Island-provide compelling, albeit
controversial, evidence that many of the shell deposits along
the St. Johns were emplaced for purposes other than refuse
disposal. Asa Randall's 2010 UF dissertation makes this
case for our Hontoon Island observations, and current Ph.D.
students Zackary Gilmore and Jason O'Donoughue are
expanding this line of research with diverse sets of data from
Silver Glen Run.
Efforts to document post-mining remnants of the
U-shaped deposit Wyman observed at the mouth of the run got
a boost this year through the generosity of FPAN's Richard
Estabrook. Ground penetrating radar (GPR) was deployed by
field school students under Rich's supervision to locate the
edge of the southern shell ridge, and to search for evidence of
features beneath the shell. Our observations on Mount Taylor
shell ridges dating ca. 7000-5000 years ago show that shell

Figure 8. Partially reconstructed topography along Silver Glen Run (8LA1) as it would have been witnessed by Jeffries
Wyman in the 1870s, illustrating the arrangement of an Orange period U-shaped mound, a Mount Taylor period shell ridge,
and associated sites. Excluded from this reconstructed landscape is the pre-mining configuration of the "amphitheater"
Wyman observed encircling the spring basin (figure courtesy of Asa R. Randall).

THE FLORIDA ANTHROPOLOGIST

2010 VOL. 63(3-4)

2010 FIELD SCHOOL SUMMARIES

Figure 9. Prelimnary GPR results of the South Ridge locus of 8LAI-East, courtesy Rich Estabrook and FPAN. This "time
slice" reflects reflectance levels roughly 45-55 cm below the surface. The lightest shades of gray signify highest reflectance,
which one-meter-interval coring showed to correlate best with density and compaction of subsurface shell. An overall arc of
high-density shell with a projected diameter of -80 m provided some hope for a circular compound of houses, but subsurface
testing to date remains inconclusive (figure courtesy of Rich Estabrook and Asa R. Randall).

was sometimes emplaced over abandoned villages, leading
us to expect house floors and related features beneath any
shell that appears to be deliberately emplaced. However, the
deposit at the mouth of Silver Glen apparently dates to the
Orange period, roughly 4000 years ago, and its U-shaped
configuration bears greater resemblance to shell rings of the
coast than it does Mt. Taylor ridges. Given the Orange-period
proclivity for circular living, we were thus excited to receive
GPR results that indicated an arc or ring of subsurface features.
A composite of five survey blocks showed an arcuate array of
subsurface anomalies with a projected diameter of about 80
m (Figure 9). A program of coring and limited test excavation
to groundtruth the GPR results provided little evidence for
house floors, hearths, and pits, and suggested instead that shell
was emplaced along the south ridge over an assemblage of
nondomestic features.
A second major focus of the 2010 field school was
continued testing at a shell-bearing site on the south margin
of Silver Glen Run. Integral to the dissertation research of

Zack Gilmore, 8LAl-West Locus B contains a well preserved
midden and feature assemblage spanning the late Mt. Taylor
and Orange periods. Initial testing in 2007 revealed complex
stratigraphy consisting of stacked, buried surfaces, emplaced
shell, and large, deep pits. Block excavation in 2009 showed
that shellfish steaming pits up to one meter wide and 1.5 m
deep were dug during the Orange period, which proved to be
divisible into early and late components, the latter signified
by the Tick Island variety of decorated fiber-tempered pottery.
Block excavation this past year expanded on this feature
assemblage and also revealed deeply buried Mt. Taylor
deposits rich in vertebrate fauna and occasional shell, bone,
and stone tools (Figure 10). Zack is investigating the idea that
mass processing of freshwater clams in the Orange-period
pits was driven by activities taking place at the U-shaped
monument to the east. The shift from Mt. Taylor to Orange-
period use of Locus B suggests a rupture in the continuity of
local communities, perhaps the influx of coastal communities
with early pottery and a tradition of circular settlements.

Silver Glen Run East (8LA1E)
Preliminary GPR Results
August 2010

- iaeter

2007
f 2x2-m unit

0 5 10 20
meters
GPR SLICE (GRID 1 47-55CM; GRID 2-5 45-53CM)

THE FLORIDA ANTHROPOLOGIST

- '.,-

C)

Figure 10. One of many large pit features attributed to the Orange period
at Locus B, Silver Glen Run (8LAl-West) (photo by the author).

Zack is conducting petrographic analysis of Orange pottery
from Locus B and the U-shaped deposit to explore possible
nonlocal origins.
A third project of the 2010 field school was made possible
by plans of the U.S. Forest Service (USFS) to improve the
public use facilities of Silver Glen Spring. The "amphitheater"
of shell surrounding the spring was mined long before the
site was incorporated into the Ocala National Forest. Still,
portions of this deposit and associated sites remain in place
and required assessment before improvements are made to the
parking lot, bath facilities, and access trail to the spring. In
partnership with USFS, Asa Randall, Jason O'Donoughue, and
field school students conducted shovel testing in areas targeted
for renovation. Limited but pervasive subsurface deposits
were encountered in locations upslope from the spring, while
remnants of the mined deposits were encountered in the
approximately 75 m encircling the spring. The partnership
with USFS continues beyond field school to allow for some

additional testing in locations of support facilities and to
profile the mining escarpment to document the intact, basal
component of the "amphitheater," which by all accounts
appears to be Mt. Taylor in age. The St. Johns Archaeological
Field School will return to the shores of Silver Glen Run in
2011 to continue testing of the U-shaped monument and Locus
B, and to initiate testing of a St. Johns II period village on a
ridge nose overlooking the spring boil. Two sand mounds in
the vicinity of this ridge nose village attest to the enduring
cultural significance of this locale to native peoples over five
millennia. Despite the devastation brought by commercial
mining to these and similar sites in northeast Florida, field
school efforts underscore the unquestionable significance that
even badly impacted sites have to furthering our admittedly
sketchy knowledge about cultural variation and change in the
St. Johns region. Our thanks to the U.S. Forest Service and our
field school hosts, the Juniper Club of Louisville, Kentucky,
for making all this possible.

2010 VOL. 63(3-4)

X.V1L,

2010 FIELD SCHOOL SUMMARIES 191

2009-2010 University of North Florida Mocama
Archaeological Project:
Black Hammock Island, Duval County, Florida
by Keith Ashley

The University of North Florida (UNF) Archaeology Lab
has recently initiated a long-term study of the Contact and
Mission period Mocama-speaking Timucua of northeastern
Florida. To this end, UNF has spent the past two summers
(2009-2010) shovel testing and excavating sites at the southern
end of Black Hammock Island, just north of the mouth of the
St. Johns River. With the support of the National Park Service,
we have focused our attention on the Cedar Point (8DU81) and
Cedar Point West (8DU63) sites, both of which lie within the
boundaries of the Timucuan Ecological and Historic Preserve
in Jacksonville.
As a result of the 2009 and 2010 UNF summer
archaeological field schools, 406 shovel tests have been dug
at the Cedar Point West site (8DU81), indicating a site size
of 800 x 500 m. Cultural components span the Late Archaic
Orange through the American Plantation periods. In addition,
45 1-x-2-m units and two 1-m squares have been excavated
in areas well represented by Contact (San Pedro) and Mission
(San Marcos) period pottery. Archaeological and documentary
evidence points to this site as the location of the ca. A.D. 1600
Mocama settlement known to the Spanish as Vera Cruz.
At the nearby Cedar Point site (8DU81), we spent 2 weeks
in 2009 expanding upon an excavation block first excavated
by UNF in 2007. These excavations focused on a later Mission
period component believed to be associated with the relocated
Mission of San Buenaventura de Guadalquini (A.D. 1684-
1696). The extensive amount of work performed by the UNF
lab over the past two summers is attributable to the enthusiasm
and hard work of UNF students and volunteers. What follows
are overviews on the 2009 post-field school testing at the Cedar
Point site and our 2010 field school at the Cedar Point West
site, each written by a UNF undergraduate student directly
involved in fieldwork.

2009 University of North Florida Summer Field School:
The Cedar Point Site (8DU81), Mission Santa Cruz de
Guadalquini, Duval County, Florida
by Michael Stull

The findings of the 2007 Block B excavations at the Cedar
Point site (8DU81) were significant and highly encouraging.
Large amounts of San Marcos pottery, colonoware (ceramics
with Spanish forms produced by natives using their own
materials and techniques), Spanish olive jar along with
carbonized corncobs, a peach pit, and a sacred heart of Jesus
finger ring were recovered. These items are all indicative of a
mission period site that can date to between 1587 and 1702.
The recovery of post-1650 Spanish majolica types helped
narrow down the dates for this site. Moreover, there was a
near absence of San Pedro pottery, the dominant contact and
early mission period pottery type found on Mocama sites.
The high ratio of San Marcos to San Pedro points to the
presence of a late seventeenth-century mission period site.

These archaeological findings combined with supporting
documentary evidence suggest that this is the site of Mission
Santa Cruz de Guadalquini, a relocated Mocama community
from St. Simons Island, Georgia formerly known as San
Buenaventura de Guadalquini.
These findings came at the very end of the 2007 field
season. The anticipation from the end of one field season until
the beginning of the next is likely a common occurrence for
most archaeologists. In this instance, the anticipation would
last more than one year because the 2008 field school was
already planned for a different site. For me, however, this delay
would prove to be a fortunate twist of fate, for my fieldwork
career began in 2009. Following the completion of the 2009
field school at the Cedar Point West site, the decision was
made to revisit 8DU81. Several field school students spent an
additional two weeks working at 8DU81.
A series of 1 x 2 m units (Block C) were laid out expanding
upon the earlier 2007 excavation block. As excavations began,
it became apparent that the 2007 findings were no fluke. Large
quantities of San Marcos and colonoware were uncovered as
were more charred corncobs, peach pits, and pig bones. As
work progressed to a depth of 30 cm below surface, several
discrete shell-filled features appeared (Features 4, 7, 9, 10, and
13). Upon closer examination it became clear that these shell
features were in fact postholes. What also stood out was that
these postholes were aligned in a linear fashion that formed a
right angle (Figure 11).
Feature 13 was among the most well-preserved and
artifact-rich postholes. This feature was first observed in plan
view as a roughly circular shell deposit with a very conspicuous
core void of shell. Near the top of the shell deposit was a large
utilized (hafted) whelk shell. Feature 13 was removed very
meticulously and the excavation was thoroughly documented
through continual mapping and photography. During this
process Feature 13 was bisected in order to get a profile view,
which provided considerable insight into the construction
techniques used. First, a conical-shaped hole was dug then a
post was inserted. The feature profile revealed a flat-bottomed
postmold suggestive of a square cut post, clearly a European
design. Next shell was dumped in and tamped down as
evidenced by crushed shell. This was followed by more shell
which was not crushed like the shell at the bottom of the
feature. What was really significant about the shell fill around
the posts was that it contained large quantities of pottery and
faunal remains. In essence these postholes acted like very
short-lived middens. Included within the shell fill were 20 San
Marcos sherds (several very large), a piece of Spanish Puebla
polychrome majolica, a few colonoware sherds, one storage
jar fragment, one glass bead, one dark green bottle fragment,
one hand wrought nail, and modified bone and shell. Also
found was a very large piece of Spanish olive jar, almost the
size of a human hand. The size and placement of this olive jar
sherd in particular is intriguing. This sherd might have been
intentionally placed near the post, possibly to help stabilize
it. Also recovered from other features in Block C were four
kaolin pipe stems, a gunflint, and 11 glass beads.
The high frequency of San Marcos sherds, colonoware
(with San Marcos styles and pastes), and majolica all served to

Figure 11. Block C South showing three posthole features, including Feature 13 (center of photo) with its shell free core still
visible.

date this site to the late Spanish mission period. The presence
of mission period artifacts, however, does not unequivocally
indicate that a mission was present there. The structural
evidence along with the types of artifacts and animal bone
recovered from Block B-C suggest the presence of a domestic
structure, possibly a kitchen. Square cut posts aligned together
in the form of a square or rectangular building all indicate a
European design. When all the archaeological evidence is
put together and combined with information gleaned from
Spanish documents, the Cedar Point site (8DU81) stands as
the likely location of Mission Santa Cruz de Guadalquini
(A.D. 1684-1696). To confirm this, much work remains to
be done, including defining the full size and nature of the
structure found. The 2010 field season did not include a return
to 8DU81. Perhaps fate is awaiting the next student whose
fieldwork career has yet to begin.

2009-2010 University of North Florida Field School:
Cedar Point West (8DU63), Duval County, Florida
by Sharon Wester Davis

During 2009 and 2010, the UNF Archaeology Lab
conducted a summer field school at the Cedar Point West
site (8DU63), located along the southwestern tip of Black
Hammock Island. The site is on high ground bounded by salt
marsh and tidal creeks, which would have provided an excellent

source offish and shellfish for the native population. Previous
archaeological investigations in the area had discovered
evidence for a Contact and/or Mission-period settlement,
possibly the visit of Vera Cruz. Historical records show that
Vera Cruz was one of nine outlying Mocama villages tied to
the nearby mission San Juan del Puerto in 1602.
I was a member of the 2010 UNF field school. The earlier
2009 field school succeeded in excavating 212 shovel tests
in the northern part of the Cedar Point West site. Subsequent
block excavations, totaling 49 n', focused on areas of Contact
period San Pedro pottery and uncovered a variety of features
dating to the Contact period and earlier. The 2010 field season
continued the shovel test grid to the south, resulting in the
excavation of an additional 194 shovel tests. To our surprise,
shovel testing revealed a distinct concentration of San Marcos
pottery in this part of the site. Because San Marcos pottery was
manufactured during both the early and late Mission periods,
it was unclear whether this San Marcos concentration related
to the early mission visit of Vera Cruz or the later, relocated
mission of Santa Cruz de Guadalquini. Also surprising was the
amount of Late Archaic Orange pottery found in the southern
part of the site. This fiber-tempered ware is some of the earliest
ceramics in the Americas, and its presence suggests that the
site was occupied by Native Americans for more than four
thousand years.

2010 FIELD SCHOOL SUMMARIES 193

Within the San Marcos concentration we also recovered
several olive jar sherds and a glass bead. We decided to focus
on this area for our large excavation units to look for the
artifacts, structures, or other evidence that could help give
us cultural and temporal context for this part of the site. In
all, 39 square meters were excavated within an 18 x 14 mn
area. Student teams dug 19 1 x 2 meter units. After shovel
testing, I spent most of my time working in Block E which
measured roughly 3 x 5 meters (Figure 12). Beneath the
mission period midden, we found a large sherd concentration,
which we referred to in the field as a "pot bust," that contained
the remains of at least two pots, likely dating to early Swift
Creek time (A.D. 200-500). Below the pot bust, we exposed
a Woodland-era shell feature that extended a good 1.5 meters
below ground surface. The ability to work beyond our original
research question was invaluable. The excavation of the sherd
concentration, the lower shell midden, and shell feature, which
were all much earlier in time than our research focus, allowed
us to gain valuable knowledge about native life in Florida
during the Woodland and the Late Archaic periods.
A wide array of artifacts was found, ranging from
plantation period (mid to late nineteenth century) refined
earthenwares to Orange sherds. While only small amounts
of San Pedro pottery were recovered, we did find Mission

period colonoware and olive jar; the latter are large European
storage vessels that were originally used to carry wine, olive
oil, or water. Some of the most fascinating finds included a
small concentration of carbonized corncobs possibly used
for smudging or as insect repellant, and a glass bead from a
rosary. We also found kaolin pipe stems and a gun flint, which
suggest that the San Marcos concentration was connected to
the later Mission of Santa Cruz and not the earlier visita of
Vera Cruz. This indicates that households associated with the
main mission compound to the cast were spread over a much
larger area than we originally thought.
After five weeks in the field, research continued in the
lab, where we were able to do more in-depth analysis of the
pottery. The extensive process of identifying and cataloging
the artifacts emphasized the meticulous documentation we
learned in the field. During the field school, we received
valuable field experience excavating units and features,
screening, mapping, and identifying artifacts; in other words,
putting the methodology to work. Working on a field problem
and following it up with the laboratory analysis is not a skill
that can be learned in a classroom. The dedication of our
professors instilled an appreciation for the methodology and
the art of archaeology.

Figure 12. UNF 2010 field school. Shaza Davis and Amber Shelton in foreground.

THE FLORIDA ANTHROPOLOGIST 2010 VOL. 63(3-4~

2009-2010 University of South Florida Field School:
St. Vincent Island and Jackson County, Florida
by Nancy White

During summer 2009, the University of South Florida
archaeological field school, directed by Nancy White and
graduate student supervisor Elicia Kimble, conducted a survey
of St. Vincent Island in northwest Florida. This large triangular
barrier island on the southwest side of the Apalachicola delta
is a National Wildlife Refuge, with many miles of bayshore
covered in oyster shell midden sites. Historically, it was a
private hunting preserve, with a wealthy owner who introduced
exotic animals; Asian sambar deer and other species are still
present. One project goal was to update 25-year-old records
on all prehistoric sites and see what is left after storm damage
and severe erosion.
Sixteen previously known prehistoric sites, and one newly-
discovered site were documented. Shovel testing throughout

the whole island confirmed that prehistoric occupation was
limited to the bay shores on the north and east side, the oldest
beach ridges (nobody wanted to live on the Gulf shore until
modern times). Test excavation was conducted at the St.
Vincent 5 site (8FR364), with the thickest midden, to see
if the deposits were intact or washed over and disturbed. A
1-x-1-mn unit exposed a meter of intact midden with Woodland
ceramics. Animal remains recovered by flotation of 9-liter soil
samples from each of the eleven 10-cm levels were analyzed
during the academic year by Florida State University professor
Rochelle Marrinan and her students in Paleonutrition class.
Besides oyster and a dozen other shellfish species, this faunal
assemblage was dominated by fishes, especially mullet, drums,
and catfishes, and also included turtle, bird, chameleon, deer,
rabbit, rat, and even whale.
Another aspect of the research was geological sampling at
the Paradise Point site (8FR71), done with FSU geologists Joe
Donoghue and Frank Stapor, to study sea level fluctuations.

Figure 13. Beginning Test Unit A at the St. Vincent 5 site, May 2009, with Elizabeth Ush-
erwood measuring, Elicia Kimble recording, Maura Peterson troweling; many students
wore head-nets in response to the diabolical mosquitos.

THE FLORIDA ANTHROPOLOGIST

2010 VOL. 63(3-4)

2010 FIELD SCHOOL SUMMlARIES 195

Figure 14. Excavating Shovel Test 12 at Gotier Hammock midden (8Gu2), with Joe Frez-
za troweling, Erik Palm digging, Caitlin Kelley bagging, June 2009; note more head-nets
in response to fiendish mosquitoes.

Cores were taken to recover datable sediments from the shell
midden strata above and below the greenish clay stratum
indicating higher sea level than at present. A final goal of the
work was to develop a site-monitoring program for visitors to
the island so that they can participate in archaeology without
picking up artifacts illegally from this federal land. Kimble
will develop both the research and public archaeology aspects
of the project in her M.A. thesis now in progress. Additional
activity during 2009 was investigating a curious pile of shell
in the yard of a private residence on Indian Pass peninsula on
the mainland across from St. Vincent Island. Shovel testing
demonstrated it to be apparently a modern deposit -possibly
someone's discarded and very large shell collection! We also
dug a few more shovel tests at the midden associated with the
Gotier Hammock mound site on the shore of St. Joe Bay, to
determine the extent of the site and obtain material for dating.
In 2010, the USF field school again combined research
and community-based public archaeology, this time in Jackson
County, on the upper Apalachicola. Longtime residents of a
large agricultural area invited White and crew to investigate
the sites on their land near Ocheesee Pond, which lies in
between the Apalachicola and Chipola rivers. Test excavation
revealed at least one area with Early Archaic remains in situ
(two Bolen points) below the plow zone. Artifact collections
totaling over 1000 items, amassed by three generations of the
family and others, were documented. Diagnostics indicate
near-continuous occupation from late Paleo-Indian (Dalton)
through at least Middle Archaic times, then a small amount
of Woodland and Fort Walton ceramics suggesting short-term
use in later prehistory. Doubtless the area was a hunting and
fishing center, as it remains today. Graduate student Caitlin

Kelley will write up this project for her M.A. thesis. During
the field season the crew also helped community leaders
setting up a new clinic and learning center in the nearby town
of Marianna. Survey and shovel testing around a historic
general store in an African-American neighborhood turned up
fascinating artifacts from the early twentieth century, when
typical containers were not of plastic but thick glass.

Figure 15. Students Bart McLeod, Caitlin Kelley, and Jake
Jones photograph and classify private artifact collections
using the Florida, Georgia, and Alabama point guides, re-
spectively, outside a family homestead in Jackson County,
northwest Florida, May 2010.

~jr;i~ ~

THE FLORIDA ANTHROPOLOGIST

2010 University of West Florida Maritime Archaeology
Field School:
Pensacola, Florida
by John Bratten and Greg Cook

The University of West Florida's 2010 maritime
archaeology field school began with several goals and research
projects planned. A major focus involved continued excavation
on the stern of the 1559 Tristin de Luna wreck site (dubbed
"Emanuel Point II"), as well as further investigations on a
wooden-hulled vessel near the Pensacola waterfront known
as the "B-Street Schooner." We also planned remote sensing
surveys of areas in the vicinity, both for training purposes as
well as to ascertain whether other significant submerged sites
lay undiscovered nearby. Pensacola's rich maritime heritage
makes it a great place to conduct maritime archaeological
investigations, with sites ranging from early attempts by
the Spanish to colonize "La Florida" (the Luna shipwrecks)
through nineteenth- and twentieth-century working boats and
fishing craft laying preserved in the sediments of the bay.
After a week of scientific diver training at UWF's Marine
Services Center overseen by our new Dive Safety Officer

Fritz Sharar, students began working in the murky waters
of Pensacola Bay, setting up grids, water induction dredges
and other equipment for underwater excavation. Several 1
x 1 units were opened up on the stern of Emanuel Point II,
which ended up exposing two of the vessel's gudgeons or
hinge assemblies that would have secured the ship's rudder
to the sternpost. Additionally, divers exposed the interior bow
area of the B-Street Schooner, allowing the recording of key
structures in the front of the ship.
A key concern this year related of course to the Deepwater
Horizon oil rig explosion that had occurred in April of 2010.
Field school directors kept in contact with local scientists
sampling the inshore and offshore waters of Pensacola Bay,
as well as federal agencies in order to monitor the situation
and cancel diving operations if oil was spotted in the bay. Staff
and students also evaluated several known historic wreck
sites to get an idea of site conditions prior to any oil impact
on these resources. Unfortunately, in June oil from the spill
made its way into Pensacola Bay, prompting the field school
directors to cease diving operations. After further evaluation,
dive operations for the field school were moved into inland
waterways that suffered no impact from the oil spill, and

Figure 16. UWF graduate student Jacob Shidner checks in for another day at the office during the 2010 field school in un-
derwater archaeology.

2010 VOL. 63(3-4)

2010 FIELD SCHOOL SUMMARIES

Figure 17. Site mapping by UWF undergraduate student
Amanda Dahlberg.

research continued on known wreck sites in the Blackwater
River, as well as additional survey to search for new sites.
Students recorded hull structure on abandoned lumber
schooners off of Shields Point, and mapped in the remains of
two previously unrecorded sites, known as the "Swingbridge
Wreck" and the "Centerboard Wreck". In addition, side scan
sonar surveys revealed the location of two new sites, which
will be evaluated further as part of a field school in 2011.

I

Figure 18. Mosaic of the profile of Block 2.

2010 University of West Florida Field School Pensacola
Colonial Frontiers Project:
Mission San Joseph de Escambe, Escambia County,
Florida
by John Worth and Nonna Harris

During the summer of 2010, the University ofWest Florida
returned to Molino, Florida, to follow up on archaeological
fieldwork done in 2009 at the site of the eighteenth-century
Spanish mission San Joseph de Escambe. The terrestrial
field school took place over ten weeks from May 17''' until
July 23'd. Field work was overseen by Dr. John Worth and
Norma Harris assisted by two graduate field directors, Jennifer
Melcher and Rachel DeVan. The field school consisted of four
graduate-student supervisors and eight full time terrestrial
undergraduates, along with a rotating crew of twelve combined
terrestrial and maritime students split into two groups, each of
which worked on the site for a five week period.
Mission San Joseph de Escambe is thought to have been
established in 1741 under the leadership of Tallahassee-born
chief Juan Marcos Fant with an initial population of roughly
thirty Apalachee Indians, and with the assistance of Franciscan
friar Marcos de Hita, based at Presidio Isla de Santa Rosa.
After nearly two decades with a single Spanish friar and
only a few Spanish soldiers with Apalachee wives, early in
1760 a Spanish cavalry detachment was garrisoned in the
mission, including an officer and fifteen men. The mission
was destroyed, however, on April 9th, 1761 when 28 Creek
warriors raided and burned the settlement in a night attack,
killing and wounding several soldiers and capturing the rest
of the small garrison. In the context of growing hostilities
between Creek and Spanish populations during the French
and Indian War, Apalachee residents abandoned the site and
retreated southward to join Yamasee refugees from another
burned mission, and taking up residence near Presidio San
Miguel de Panzacola in present-day downtown Pensacola
before evacuating to Veracruz, Mexico in 1763.
In 2010, excavations started with additional shovel testing
to further delineate the boundaries of the site found in 2009.
This was followed by opening three large block excavations

THE FLORIDA ANTHROPOLOGIST

placed adjacent to productive test units from the 2009 field
season in the core of the mission. These blocks were opened in
order to further examine several significant structural features
found in 2009. Additionally, a new test unit was opened on the
northern margin of the site around an isolated positive shovel
test with mission-era debris.
Excavations in Block I provided further information on a
very large structure which had initially been discovered during
the 2009 field season as a substantial, heavily-engineered post-
on-sill wall-trench structure. Excavations in 2010 revealed the
structure measured a minimum of thirteen meters in length.
The large size and substantial construction of the building
suggests that it may represent the remains of the final cavalry
barracks constructed by the Spaniards during the summer
of 1760. Further work will be undertaken in the future to
determine the exact size of this structure and to confirm that it
is the barracks.
Excavations in other areas of the site produced evidence
of other structures associated with the mission period, as well
as pit features. Features recorded included several smudge
pits filled with burned corn cobs, which were likely used as

a form of pest control. One area of the site produced several
overlapping wall-trench structures of unknown size and
function overlain by a clay capping episode which appears to
date to the mission period as well. Several high status items
were recovered from this are: porcelain, French Brittany
faience, and a straight-razor, along with the largest number of
glass beads recovered so far from any area of the site. More
work will be necessary in this area before any conclusions can
be reached concerning the nature of these structures and this
area of the site.
In addition to the mission period features uncovered,
further work was undertaken to evaluate the area surrounding
a Deptford period post discovered during the 2009 field
season. We hoped to determine if there were any other posts
related to a Deptford structure, but work in this excavation
area proved non-conclusive. The unit was negatively impacted
by the rising water table caused by heavy summer rains;
however, some possibly-related posts were discovered, and
this area will require more work in the future. Along with the
prehistoric Deptford component, the site has also produced
materials dating from the Archaic through the Mississippian

Figure 19. UWF graduate students Tonya Chandler (left) and Rachel DeVan (right; shoeless) work in Block 1.

2010 VOL. 63(3-4)

2010 FIELD SCHOOL SUMMARIES

Figure 20. Feature 10 with charred corn cobs.

periods, though no features have been found relating to these
cultures.
Excavations at the site produced a large volume of
ceramics including prehistoric and mission period Native
American ceramics, as well as Hispanic ceramics such as El
Morro lead-glazed earthenware, Abo Polychrome majolica,
and Puebla Blue on White majolica. Other material recovered
from the site includes pieces of glass bottles, glass trade beads,
lead shot, iron nails and brick fragments. Some pieces are of
particular interest to research on adaptations of the Apalachee
population to the Spanish society in which they were living,
for example, worked olive green bottle glass.
Work at Mission San Joseph de Escambe is producing
valuable information about Apalachee Indians living on the
Spanish colonial frontier during the eighteenth century, and
promises to help to expand our knowledge of how these people
interacted after almost 200 years of Spanish occupation, and
after almost 50 years spent as a refugee population away
from their traditional homeland. Lab analysis is ongoing at
the University of West Florida archaeology lab, and plans are
being formulated to return to Mission San Joseph de Escambe
with another UWF field school during the summer of 2011.

2010 University of West Florida Field School:
Arcadia Village, Santa Rosa County, Florida
by John C. Phillips

As the first large-scale, water-powered commercial lumber
and textile mill complex established in Florida, the antebellum
Arcadia Mill Complex (8SR384) played an important role in
the history of west Florida, the State of Florida, and indeed the
regional southeast (Rucker 1990). Impressive archaeological
features, including a 1,400 foot long, 12-15 ft high earthen
dam, brick and stone building foundations, mill races and
log flumes, and other visible remains of the mill complex
still exist at the site, leaving tangible record of the feats of
engineering and construction that occurred here more than 150
years ago. Those who labored in the mills left behind traces of
community life in ceramics, glass, metal, and building material
in an upland village surrounding the industrial complex. In
combination, the historical and archaeological records provide
a body of evidence that can be used to reconstruct lifeways in
a slave-based industrial community.
Arcadia was established in 1817, and by the late 1820s was
a thriving industrial facility, with an associated community of
mixed ethnicity, including many enslaved African Americans
who built the enterprises and labored in its mills. In the
beginning, lumber was the principal product, and by the late
1830s Arcadia sawed over 900,000 board feet of lumber a year
(Rucker 1990). In addition to a variety of lumber products,
a pail factory (locally termed the "8th wonder of the modern
world"), grist mill, blacksmith shop and other industries also
thrived at Arcadia. A community well provided fresh water.
Roads leading to and from Arcadia connected the mills and
village community with Pensacola, Floridatown, Milton, and
other emerging antebellum communities in north Florida.
Arcadia Railroad Company, operating as one of Florida's
first railroads, was transporting lumber to market in 1838. In
1845 a substantial water-powered cotton textile mill replaced
the sawmills, employing 25-40 highly trained enslaved black
females, adding new members to the diverse community.
Textile production continued for about nine years until the
textile mill burned to the ground and the Arcadia mill complex
was abandoned.
Listed on the National Register of Historic Places in 1988,
Arcadia is now owned by West Florida Historic Preservation
Inc. (WFHPI), a direct support organization of the University
of West Florida. Arcadia and water-powered industry has been
the focus of archaeological research by the University of West
Florida for over 20 years. While our earlier work focused
on the defining the industrial facilities, in the last two years
we have turned our attention to the people who lived and
worked at Arcadia. The research is supported by West Florida
Historic Preservation, Inc. and the University of West Florida
Archaeology Institute.

Arcadia Village Project 2009

As the first in a planned series of public archaeology
projects at Arcadia, UWF's 2009 terrestrial field school
initiated a search for the mill workers living area to learn

Figure 21. Tara Giulaino and Meagan Rea in Block 1 at Arcadia Mills.

more about the archaeology of ethnicity, social structure
and community organization of an industrial community
in the antebellum south. The 2009 Arcadia Field School,
led by Field Director Adrianne Sams, with the assistance of
graduate students Jay Bixler, Brian Mabelitini and Melissa
Timo, and a crew of 8-12 undergraduate students, undertook a
program of systematic survey and evaluation, including close
interval shovel testing, soil resistivity, magnetic anomaly
survey, and archaeological test and block excavations. The
2009 archaeological survey and evaluations revealed a very
important mill-related residential area, marked by architectural
and kitchen deposits in the uplands immediately north of the
mill facilities. Following upon the survey successes, late in the
summer we began test and excavations to expose, document
and interpret these residential deposits. The 2009 fieldwork
suggested the presence of at least one residential structure in
the uplands north of the mill complex.

Arcadia Village Project 2010

The 2010 Arcadia Field School continued the
archaeological investigations initiated with the 2009 Arcadia
Field School. The fieldwork was directed by Melissa Timo,
assisted by graduate students Andrew Christiansen, Tara
Giulaino and Kad Henderson, again with and a crew of 8-12
undergraduate students. Anticipating that more residences
would be present, we undertook more focused and intensive
remote sensing and other survey level evaluation while we
continued block excavations. Remote sensing and other
survey data enabled us to identify additional architectural
features, suggesting that several structures are present.
Meanwhile, the block excavations revealed large quantities of
household artifacts, uncovered house foundations, refuse pits,
and dense concentrations of residential artifacts. Among the
artifacts recovered are imported ceramic tableware (pearlware

2010 FIELD SCHOOL SUMMARIES 201

Figure 22. Trey Ropelis excavates Feature 308 at Arcadia

and whiteware), stoneware storage vessels, kitchen utensils,
drinking glasses, bottles and window glass, pharmaceutical
supplies, cut nails, handmade brick, and cut sandstone block
pier supports. The recovered cultural material has proven to be
extremely informative, in terms of social status and community
organization. The artifact assemblage shows limited diversity
and few high status items, suggesting a residential area of
low economic status. Equally important and informative,
the documented architectural features suggest the presence
of a nucleated settlement consisting of several small, closely
spaced houses, in all probability wood framed, vernacular
structures built on brick piers bedded in cut sandstone.
The value of this work artifacts, field documents, and
ongoing analysis will complement and enrich WFHPI's
planned interpretative exhibits. We look forward to our return
to Arcadia with the 2011 field school.

Reference Cited

Rucker, Brian R.
1990 Blackwater and Yellow Pine: The Development
of Santa Rosa County, 1821-1865. Unpublished
dissertation, Florida State University, Tallahassee.

Mills.

2010 Valencia Community College Field School:
Oakland and Windermere, West Orange County, Florida
by Jason Wenzel

In 2010, Professor Jason Wenzel, with assistance from
members of Central Florida Anthropological Society (CFAS)
and student volunteers from the University of Central Florida
(UCF), engaged with Valencia Community College (VCC)
students in archaeological fieldwork at three different historic
sites in west Orange County. During the spring semester, the
team conducted fieldwork at two historic properties in Oakland,
a historic town approximately 10 miles west of Orlando. A
Phase I survey was initiated at the Hartsfield House (80R6269),
which is one of only a handful of surviving homes built by the
Orange Belt Railroad Company in the 1880s. According to
oral history, the house originally served as the site of the first
hospital in Orange County. In 1905, the Brock family, who
operated one of Oakland's early grocery stores, acquired the
property and resided there until 1977, when they sold it to the
Hartsfield family, who still maintain residence. In addition, the
team conducted Phase II testing at the Territo House-Oakland
Hotel site (80R9973), which underwent shovel testing during
the Fall 2009 semester. The Territo family resides in the

uHF FI.oRIuA AN I IIROI'OLOGIST 2010 VOL. 63(3-4)

Figure 23. Project participants excavate a test unit at the
Oakland Hotel site (80R9973).

fonner kitchen house & dining hall (initially built in 1890)
of the Oakland Hotel which was demolished around 1947.
During the summer of 2010, Professor Wenzel and his team
processed artifacts from the Oakland sites in the laboratory of
the Oakland Nature Preserve, Inc. where they will be curated
and eventually put on display at the organization's museum.
The team went back into the field at the start of the Fall 2010
semester, this time working in Windermere, which is located
between Oakland and Orlando. In partnership with the Florida

Public Archaeology Network and the Town of Windermere,
the team conducted a Phase I survey of the site of the historic
Windermere School (80R4151), which was initially built in
1890. The one-room schoolhouse served students of various
ages and enrollment fluctuated between 12 and 23 pupils until
1916, when a new school was constructed. The Armstrong
family moved to the property around 1924 and converted the
former school house into a kitchen after they had built a new
home on the property. In 1995, the family donated the property
to the Town and in 1999 the 1920s addition was demolished.
The Town has recently received a Special Categories Grant
from the State of Florida to restore the interior and exterior
of the structure to the appearance of an 1890s schoolhouse.
In partnership with the East Central Region office of FPAN,
a public archaeology day was held where members of the
community were provided site tours and a public lecture on
prehistoric mounds by Dr. Rachel Wentz. After work was
completed in Windermere, the team returned back to Oakland
to conduct Phase II testing of the Hartsfield House in Oakland.
This is the fifth year that Professor Wenzel has directed
archaeological fieldwork as part of his faculty duties at
VCC. Since this work has commenced, several sites have
been recorded and investigated and the program of work
has provided tangible benefits to area college students, local
communities and the greater region at large. For example, very
few community colleges in Florida have offered their students
any type of archaeological field experience or laboratory-
based learning activities and even the local four-year colleges

Figure 24. Project participants shovel testing the Windermere School site.

THE FLORIDA ANTHROPOLOGIST

2010 VOL. 63(3-4)

2010 FIELD SCHOOL SUMMARIES 203

have provided very little in this way. As a result, the projects
have brought together a community of scholars representing
VCC, UCF, and CFAS by providing them with opportunities
to engage in community-based field archaeology. The team's
participation in events such as the Windermere Public
Archaeology Day and the annual Oakland Heritage Festival
has presented a diverse population of residents from these
towns an opportunity to learn about archaeology and the
community history with which many are unfamiliar.
The projects discussed above have provided greater
Orlando with an applied archaeology program that addresses
issues pertaining to historic preservation and environmental
education as a result of intensive agriculture, population
growth, and urban sprawl that the area has sustained over
the last several decades. Outside of fieldwork, an important
aspect of this work has been supporting the cultural history

and environmental education programs of the Oakland Nature
Preserve, Inc. by examining donated avocational prehistoric
artifact collections, indexing historical documents chronicling
the ecological decline of Lake Apopka, and developing museum
exhibits aimed at addressing the long-term implications of
human interaction with the local environment while increasing
public awareness in the process. Lastly, former VCC student
and project participant Tiffany George has been examining
Oakland's modern tourism by conducting ethnographic
interviews of visitors to the Oakland Nature Preserve and
users of the West Orange Trail. The goal of Ms. George's work
is to synthesize this ethnographic data with archaeological
data from the Oakland Hotel site in order to develop a long
term perspective on changing patterns of Oakland's tourism in
the context of the decline of Lake Apopka and its subsequent
restoration.

Figure 25. Student and volunteer project participants in front of the Winder-
mere School site (80R4151) built in 1890.

S Join the Florida Anthropological Society

Florida Anthropological Society memberships:
Student $15 (with a copy of a current student ID)
Regular and Institutional $30
Family $35
Sustaining $100
Patron $1000
Benefactor $2500 or more
Student membership is open to graduate, undergraduate and high school students. A
photocopy of your student ID should accompany payment
Add $25.00 for foreign addresses
Membership forms also available at www.fasweb.org
The Society publishes journals (The Florida Anthropologist) and newsletters, normally quarterly,
and sponsors an annual meeting hosted by a local chapter.

Name:
Address:
Apt:
City:
State:
ZIP:
Telephone:
E-mail:
FAS Chapter:

SI agree to abide by the Code of Ethics of the Florida Anthropological Society.

MAIL TO:
Florida Anthropological Society
c/o Pat Balanzategui
PO Box 1434
Fort Walton Beach, FL 32549-1434

THE FLORIDA ANTHROPOLOGIST 205

About the Authors:

Alan Brech is an independent researcher of human prehistory and a Brevard County Historical Commissioner. He received
his Masters degree in anthropology from the University of Florida in 2004. His peer-reviewed work has previously ap-
peared in The Florida Anthropologist (March 2007). His other writings have appeared in The Indian River Journal and in
The Florida Today and The Florida Alligator newspapers, and at the 1996 Sundance Film Festival.

J. E Lanham is an archaeologist working in Cultural Heritage Management. He obtained his Bachelor of Arts degree in an-
thropology from the University of Florida in 1997 and is currently pursuing Geomatics studies at the University of Florida.
His peer-reviewed work has previously appeared in The Florida Anthropologist (March 2007), and his non-peer-reviewed
work has appeared in The Indian River Journal. He and Mr. Brech are co-authors of "The Destruction of the Ais Chiefdom
and Other Overlooked Ethnographic and Ethnological Information from Jonathan Dickinson's Shipwreck Journal," which
was presented to the annual meeting of the Florida Historical Society in May 2009.

Nancy Marie White is a professor of anthropology and Registered Professional Archaeologist at the University of South
Florida in Tampa. Her long-term field research centers around the Apalachicola-lower Chattahoochee river valley and St.
Joseph Bay region. Her books include GulfCoast Archaeology and Archaeologyfor Dummies.

FLORIDA ANTHROPOLOGICAL SOCIETY, INC.
POST OFFICE BOX 12563
PENSACOLA, FL 32591-2563

NON-PROFIT
U.S. POSTAGE
PAID
TALLAHASSEE, FL
PERMIT NO. 236

ADDRESS SERVICE REQUESTED

Volume 63 Number 3-4
September December 2010

CONTENTS

ARTICLES

The Location of the Paramount Town of the Ais Indians
and the General Location of the Indians of Santa Lucia.
Alan Brech and J. F. Lanham

Gotier Hammock Mound and Midden
on St. Joseph Bay, Northwest Florida.
Nancy Marie White

2010 FLORIDA FIELD SCHOOL SL\I\IARIES

Cover: Photographs of Basin Bayou Incised jar from the Gotier Hammock site (cat nos. 08-39. -44. -49. -51.-99. -107).
See the White article beginning on page 149 for more information.

Copyright 2010 by the
FLORIDA ANTHROPOLOGICAL SOCIETY, INC.
ISSN 0015-3893

	Front Cover
	Table of Contents
	From the Editors
	The Location of the Paramount Town...
	Gotier Hammock Mound and Midden...
	2010 Florida Field School...
	About the Authors
	Back Cover

MILLISECOND	CLASS.METHOD	MESSAGE
0	sobekcm_page_globals.constructor
0	sobekcm_page_globals.constructor	Application State validated or built
0	sobekcm_database.verify_item_lookup_object
0	sobekcm_page_globals.constructor	Navigation Object created from URI query string
0	sobekcm_database.verify_item_lookup_object
0	sobekcm_page_globals.display_item	Retrieving item or group information
0	sobekcm_page_globals.get_entire_collection_hierarchy	Retrieving hierarchy information
0	sobekcm_assistant.get_entire_collection_hierarchy
0	cached_data_manager.retrieve_item_aggregation
0	cached_data_manager.retrieve_item_aggregation	Found item aggregation on local cache
0	item_aggregation_builder.get_item_aggregation	Found 'all' item aggregation in cache
0	system.web.ui.page.page_load (ufdc.page_load)
0	sobekcm_page_globals.constructor.on_page_load
0	html_echo_mainwriter.add_style_references	Adding style references to HTML
0	html_echo_mainwriter.add_text_to_page	Reading the text from the file and echoing back to the output stream
959	html_echo_mainwriter.add_text_to_page	Finished reading and writing the file