The Florida anthropologist

Material Information

The Florida anthropologist
Abbreviated Title:
Fla. anthropol.
Florida Anthropological Society
Place of Publication:
Florida Anthropological Society
Publication Date:
Quarterly[<Mar. 1975- >]
Two no. a year[ FORMER 1948-]
Physical Description:
v. : ill. ; 24 cm.


Subjects / Keywords:
Indians of North America -- Antiquities -- Periodicals -- Florida ( lcsh )
Antiquities -- Periodicals -- Florida ( lcsh )
serial ( sobekcm )
periodical ( marcgt )


Contains papers of the Annual Conference on Historic Site Archeology.
Dates or Sequential Designation:
v. 1- May 1948-

Record Information

Source Institution:
University of Florida
Rights Management:
All applicable rights reserved by the source institution and holding location.
Resource Identifier:
01569447 ( OCLC )
56028409 ( LCCN )
0015-3893 ( ISSN )


This item has the following downloads:

Full Text




JUNE 2006


THE FLORIDA ANTHROPOLOGIST is published by the Florida Anthropological Society, Inc., P.O. Box 6356, Tallahassee, Florida 32314.
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 year. Annual dues are as follows: student $15, individual $30, family $35, institutional $30,
sustaining $40 or more, patron $100 or more, life $500, and benefactor $2500. Foreign subscriptions are an additional $5 U.S. to cover
added postage and handling costs for individual, family, or institutional membership categories. Copies of the journal will only be sent
to members with current paid dues. Please contact the Editor 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 Editor to facilitate acknowledgment in
subsequent issues of the journal (unless anonymity is requested). Submissions of manuscripts should be sent to the Editor. Publications
for review should be submitted to the Book Review Editor. Authors please follow The Florida Anthropologist style guide (49[1], 1996,
pp. 37-43) in preparing manuscripts for submission to the journal and contact the Editor 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 Word Perfect 12.0. All
manuscripts must be submitted in final form on DOS-formatted 3.5 inch diskettes or CDs in Word Perfect, Word, or ASCII 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.


President: Sheila Stewart, 2130 Burlington Avenue North, St. Petersburg, FL 33713 (
First Vice President: David Burns, 15128 Springview St., Tampa, FL 33624 (
Second Vice President: Christine Newman, 504 17th St., St. Augustine, FL 32084 (
Corresponding Secretary: Antoinette Wallace, (
Membership Secretary: Beth Sarney, 5331 Lansdowne Way, Palmetto, Fl 34221 (
Treasurer andRegisteredAgent: Joanne Talley, P.O.Box 788, Hobe Sound, FL 33475 (
Directors at Large: Gloria Fike, 2815 SE St. Lucie Blvd., Stuart, FL 34997 (; Robert J. Austin, P.O. Box
2818, Riverview, FL 33568-2818 (; Tommy Abood, 3857 Indian Trail #403, Destin, FL
Immediate Past President: David Burns, 15128 Springview St., Tampa, FL 33624 (
Newsletter Editor: David Burns, 15128 Springview St., Tampa, FL 33624 (


Editor: Ryan J. Wheeler, P.O. Box 6356, Tallahassee, FL 32314-6356 (
Book Review Editor: Dan Hughes, 2301 8th Ave. North, St. Petersburg, Florida 33713 (
Technical Editor: vacant
EditorialAssistant: George M. Luer, 3222 Old Oak Drive, Sarasota, FL 34239-5019 (
Printer: Durra-Print, 717 South Woodward Ave., Tallahassee, FL 32304
BulkMail: Capital City Mailing, 4013 Woodville Hwy, Tallahassee, FL 32311
Back Issue Sales: Palm Beach Museum of Natural History (


Albert C. Goodyear, Institute of Archaeology and Anthropology, University of South Carolina, Columbia, SC 29208 (
Jerald T. Milanich, Florida Museum of Natural History, University of Florida, Gainesville, FL 32611 (
Jeffrey M. Mitchem, Arkansas Archeological Survey, P.O. Box 241, Parkin, AR 72373 (
Nancy Marie White, Department of Anthropology, University of South Florida, Tampa, FL 33620-8100 (
Robert J. Austin, P.O. Box 2818, Riverview, FL 33568-2818 (

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.





0 A

Volume 59 Number 2 'NCE 19A-1
June 2006

Editor's Page 71
The Ephemeral Cape St. George Shipwreck on the Northern Gulf Coast,
Franklin County, Florida. Nancy Marie White 73
Colorinda and its Place in Northeastern Florida History. Keith H. Ashley 91
Archaeological Testing of Colorinda Shell Middens at the Cedar Point Site (8DU81). Keith H. Ashley 101

Cobb: Stone Tool Traditions in the Contact Era. David K. Thulman 123
MacMahon and Marquardt: The Calusa and Their Legacy:
South Florida People and Their Environments. Michelle J. LeFebvre 124
Missall and Missall: The Seminole Wars: America's Longest Indian Conflict. Justin Martin 126
Pluckhahn: Kolomoki: Settlement, Ceremony, and Status in the Deep South,
A.D. 350 to 750. Greg S. Hendryx 127

Authors 129

Cover: The Cape St. George Shipwreck, July 8 1996 (see article beginning on page 73 for more information).

Published by the
ISSN 0015-3893


This issue begins with an article on the Cape St. George
shipwreckwritten by University of South Florida anthropology
professor Nancy Marie White. This shipwreck, like many
others that have recently been discovered along the Big Bend
and northwestern Florida Gulf coast, was apparently buried in
the past under beach and dune sands. Tropical storms in 1994
uncovered the nineteenth century shipwreck and Dr. White
and her students seized the opportunity to document this
interesting site. Hopefully other researchers will study the
other late nineteenth and early twentieth century shipwrecks
of this area and we hope that more articles on shipwreck and
maritime archaeology are submitted to The FloridaAnthropol-
The second and third articles, by Savannah College of Art
and Design professor Keith Ashley, presents an important
overview of the Colorinda pottery type and archaeological

culture that dates to the ninth century A.D. Dr. Ashley raises
a number of questions that can be explored regarding the
origin of this culture and several avenues for future research,
including a detailed analysis of Colorinda pottery. The third
article is Dr. Ashley's site report on the Cedar Point site
(8DU81), which has a significant Colorinda component.
Together these articles should inform further research on this
culture and raise awareness of those working in northeastern
This issue also includes four book reviews by Dave
Thulman, Michelle J. LeFebvre, Justin Martin, and Greg S.
Hendryx. We hope that all readers find something of interest!

June, 2006


VOL. 59(2)


JUNE 2006

The Florida Anthropologist

are available from the
Palm Beach Museum of Natural History:




Department ofAnthropology, University of South Florida, 4202 E. Fowler Ave. SOC107, Tampa, FL 33620
E-mail: nwhite@cas.usfedu

First discovered in the spring of 1996, the shipwreck on the
cape section of Little St. George Island in Franklin County,
designated 8FR857, was recorded during July 1996 as part of
an archaeological survey in the path of 1994 tropical storms
Alberto and Beryl. The shipwreck is a portion of a large
wooden cargo vessel believed to be American or British, which
apparently wrecked some time around 1870 to 1890. It was
copper-sheathed, with fasteners of copper, iron, Muntz metal,
and wood tunnelss or "tree nails"). It was probably one of
hundreds of vessels participating in the global commerce
involving forest products and other commodities harvested
throughout the nineteenth century in northwest Florida. Like
any shipwreck, it must be understood within the cultural
systems of its time (Gould 1983, Lenihan 1983, Murphy 1983,
Watson 1983). The dynamic environment of the Gulf shore
resulted in greater exposure of the wreck after it was first
recorded, but it was difficult to monitor due to its isolated
location. The sea and sand then reclaimed the material
evidence; the wreck disappeared in a little over two years.
This article expands upon the original report of this shipwreck
(White 1996:70-72) in order to describe the evidence, place the
ship within its historical and socioeconomic context, and show
the behavior of natural site formation processes in this very
dynamic coastal environment.

Cultural and Environmental Setting

A barrier island chain drapes around northwest Florida's
Apalachicola Bay (Figure 1) from west to east like a sparkling
white necklace. The islands are around 4,000 years old,
shaped by the wind, water, and gravity of marine, estuarine,
and fluvial processes. The Apalachicola River brings sand
down to the coast from the Appalachian mountains far into the
interior. St. George Island, the longest one at nearly 48 km,
bends at the western end to create a cape, with a shorter
western arm running northwest-southeast about 7 km and a
longer eastern arm extending some 40 km to the northeast.
Dimensions of barrier islands are often approximate, since
they are constantly changing. One storm can chop off hun-
dreds of meters of land from one end and deposit it on the
other or cut inlets or passes dividing lengthy strips into smaller
pieces. Bob Sikes Cut, named after a long-term U.S. Congres-
sional representative from this region, was dug near where a
natural channel had periodically opened in St. George Island,
and is now formally maintained for navigation with a pair of
jetties. This cut insulates the "developed" (heavily inhabited)

portion of the island from the triangular, uninhabited, 14-km-
long triangle of Cape St. George, also known as Little St.
George Island, the pendant jewel in the necklace.
All these barrier islands except St. Vincent (which is wide
and very close to shore) are thin, far less than 1 km wide in
some washover areas. Composed ofunconsolidated sand, with
dunes sometimes overlying peat deposits, the barrier islands
are the most dynamic part of the watery landscape, changing
yearly, seasonally, sometimes weekly, and often radically, with
both violent storms and also slow erosional processes
(Livingston 1989; Dickinson et al. 1992:7-8; Champion 1996;
Randazzo and Jones 1997; Morton et al. 2004). They are
important buffers for the bay waters and range from a few
hundred meters to many km offshore, separated from the
mainland by Apalachicola Bay and various sounds. Vertical
relief on the barrier islands is primarily due to wind deposits,
with the overall shape of the cape determined by the sea (Gore
The waters of this area are some of the most productive
fisheries in Florida, famous for oysters, shrimping, crabbing,
and finishing. Prehistoric use of the barrier islands is seen in
shell midden ridges lining the bayshores, which were more
sheltered and offered fresh water sources. Seasonal, short-
term seafood harvesting is documented from as early as the
Late Archaic (some 4000 years ago) through protohistoric
times (Donoghue and White 1995; White et al. 1995, 2002;
White 1997, 2005). Historic European and American uses of
the islands have also been mostly ephemeral, until recently
(Owens 1966; Rogers 1986, White et al. 1995; Damour et al.
2001; Meide et al. 2001; Damour 2002; McCarthy 2004;
Horrell 2005). The Spanish and British conducted brief
military activities in these isolated locales, and there were
always people camping to fish and hunt.
The islands' forests became important to harvest for the
shipbuilding and naval stores industries by the early nine-
teenth century and continuing through the mid-twentieth
century. Lighthouses were built to aid navigation. During the
Civil War, brief military activities took place, but afterwards,
cutting the timber, turpentining, and raising cattle became the
main uses for these remote places (along with occasional
smuggling and moonshining). In addition, some wealthy men
bought portions of the islands to use not only as investments
in those resources, but also as seasonal hunting and fishing
retreats, even introducing exotic game on St. Vincent Island.
During World War II a military base covered the isolated
islands. Troops training there considered it a miserable place,


VOL. 59(2)


JUNE 2006


Figure 1. Location of Cape St. George Shipwreck on barrier islai

with burning hot sand, little water, relentless sun and insects,
and bitter winter cold (Bradley andBlair 1983:112; Huntsman
1992; Coles and Gregory 2005:7). Many of these were
African-American soldiers, and the people spending the most
time on the islands for at least a century were poor black
laborers cutting lumber, loading ships, collecting turpentine
until the 1950s. Today much of the island landscape is
covered with houses and commercial establishments mostly for
white middle-class and wealthier seasonal and permanent
inhabitants. What I call the "coppertone culture" has trans-
formed the former isolation and misery into dazzlingly
attractive real estate, despite the high risk of hurricanes and
changing landforms (Kaufman and Pilkey 1979). The perils
of life in the coastal high-hazard zone were seen recently as
two Category-3 storms, Ivan and Dennis, struck the panhan-
dle, including St. George Island, within a 10-month period
(FloridaDepartment of Environmental Protection 2004,2005).
Amid the rapid development, the Cape St. George preserve
is uninhabited, managed by the Apalachicola National
Estuarine Research Reserve (ANERR). It has a research
station and the ruins of an old turpentine camp on the bay side,
and a recently fallen 1852 lighthouse on the Gulf side on the
point of the cape that protrudes southward into the Gulf. With
student crews from the University of South Florida I have
continually monitored site erosion on the bay beaches here,
but seldom was there much of archaeological interest on the
Gulf side until 1996, when the remains of a large shipwreck

suddenly appeared there. Documentation of
its discovery and disappearance is of interest
for both its cultural and natural aspects.

Field Investigations and Description

Exposure and Field Recording

The recording of the Cape St. George
shipwreck was part of the work of a two-
year project entitled "Northwest Florida
Flooded Sites Investigations," supported by
a grant from the Florida Division of Histori-
cal Resources, Department of State, spe-
cially allocated for historic preservation
DOG studies in the wake of the record flooding of
A ISLAND 1994. The project ran from July 1995 to
August 1996. The 500-year record flooding
,fo in the Apalachicola Valley, which was
unusual because it took place not during late
winter-spring annual flood season but dur-
ing the summer, had been caused by two
tropical storms. In early July 1994 Tropical
10 20 Storm Alberto hit the north Gulf Coast,
S-- damaging much of northwest Florida, south
Alabama, and south Georgia. As the waters
receded in early August, I took a small team
nd in northwest of students up the river to see what damage
there was to archaeological sites. On 12
August we went to Cape St. George and,
contrary to our typical practice, did not spend the night on the
island due to a student's injury. This proved to be lucky
timing since that night Tropical Storm Beryl blew in with
more rain and wind, causing even greater destruction and a
few deaths in the region. In 1995 federal funds were given to
the states for the grants to assess damage to historical re-
sources from the storms and flooding.
Our two-year, post-flooding archaeological survey of the
Apalachicola River valley (and parts of the lower
Chattahoochee) has been reported in detail (White 1996).
Meanwhile, in October 1995, between the two field seasons of
this survey, Hurricane Opal battered northwest Florida. It
blew off enough of the Gulf shoreline on neighboring Dog
Island, to the east (see Figure 1), to expose the deeply buried
peat under the white sand (which quickly came back). It may
have begun or accelerated the process of uncovering the Cape
St. George shipwreck, though the wreck was not yet exposed
immediately after this hurricane.
The Cape St. George shipwreck (8FR857) was located on
the Gulf shore of the cape, approximately 500 m west-north-
west of the southernmost tip where the lighthouse stood. It
was exposed right at the shoreline, straddling the sandy beach
and the crashing surf. At the time of discovery, the entire
beach was covered in tangles of seaweed and fallen trees from
the storms mentioned above and countless others. Jimmy
Moses and Pat Millender of the ANERR staff had discovered
the wreck in late winter-early spring 1996. Though they only

2006 VOL. 59(2)



Figure 2. Appearance of shipwreck on 2 July 1996, view facing west. Crew chief Scott Grammar inspects exposed, black
planks at left, and curve of the hull is beginning to be visible at right, partially covered by a tangle of tree roots.

monitor the shoreline sporadically, they were certain that the
wreck was not visible before this, and there had been a specific
check of the beach after Opal. Thus it is unclear whether the
wreck's appearance or exposure was due to longer-lasting
consequences of the tropical storms or that hurricane or to
some other geomorphological processes. The Cape is one of
the most unstable segments of the Apalachicola barrier island
rim (Donoghue and Tanner 1994); evidence from historic
charts shows it has migrated at rates of approximately 8 meters
per year over the past century.
The wreck was first formally recorded by my crew on 2
July 1996, including videotaping and photography. At this
time of initial investigation, less than 25-30 m2 (ca. 275 square
feet) of it was uncovered, consisting of a hull fragment to the
northwest and portions of exposed planks 6 m (20 feet) away
to the southwest (Figure 2). The cluster of planking had a
black pitch coating and a few brass fasteners remaining
(Figure 3). One plank had been blown/washed perhaps 500 m
farther west-northwest along the beach; it was brought back to
the ANERR for curation, along with a sample of the metal
A return visit was made on 8 July with experts from the
state Division of Historical Resources: underwater archaeolo-
gist Roger Smith, museum educator KC Smith, and Bob
Vickery, along with ANERR personnel. Between our initial
investigation and this follow-up, less than one week later, a
typical summer storm had hit and exposed the wreck further.
At this time much more of it was now visible (Figures 4, 5), a
section approximately 13 m (43 feet) square. New tangles of
trees and roots joined what had already been on top the wreck.
More video and photos were obtained, and small samples were
taken to Tallahassee by Smith for curation at the DHR. Some

details of this description are based on notes taken during his

Vessel Description

The wrecked remains were from a large, solidly built
ocean-going cargo ship, British or American, dating to
sometime in the later 1800s. The portion extant was the port
stern quarter of the ship. More planks coated with pine tar
were exposed during our second visit, and one retained a
fragment of copper sheathing (Figures 6, 7), which was not
green but unoxidized copper-colored. Possibly it had been
protected by the sand covering and maybe the salt water, and
was not yet patinated. The planks had brass, copper, and iron
fasteners (pins, spikes, tacks) and tree nails, all ofvarious sizes
(Figures 8-12), with heads ranging from less than 3/4" to at
least 1 3/8" in diameter. Tree nails are wooden spikes, also
called trenails or trunnels, "round, dowel-like hardwood pegs
that are inserted into drilled holes to join wooden elements."
They were used as early as 1100 B.C. in Britain, and continue
to be used in wooden boat and ship joinery today (Gould
2000:99). The tree nails were sometimes set with a wedge
through the middle. Several tree nails that had lost one of the
planks they had once joined were sticking up bare in the surf;
they had also lost their wedges and they looked like a row of
upside-down, old-fashioned solid wood clothespins. Tree nail
holes were mostly 1 3/8" in diameter.
The brass was identified by Smith as a material called
Muntz metal, which was more common in the later nineteenth
century (Flick 1975). None of the wood has been identified as
to species. Some pins had what appeared to be epoxy on top.
Planks had many empty holes and portions where pins and






Figure 3. Closeup of planks exposed amid tree roots on 2 July 1996, showing coating of black pine tar and circular holes
where fasteners once were. These were outer planks whose copper sheathing had disappeared due to either natural processes
or looting.


Figure 4. The wreck on 8 July 1996, view facing west, showing curve of hull amid downed trees and tangles of roots, with
photographer KC Smith.


Figure 5. Appearance of wreck on 8 July 1996, view facing west-southwest; section of hull at left was last to retain small
fragment of copper sheathing, curled up on plank at left edge of photo. The seaward part in the center of the photo is what
was just beginning to become visible in Figure 2.

spikes might have been salvaged or looted, including some
circular cutaway portions (Figure 10) that also might have
been part of the original vessel manufacture. On the west side
of the wreck, 5 iron chain plates (Figure 13) remained. They
were 66" long and bolted to a portion of the chain wale on the
side of the hull. From the chain wale, a construction of broad,
thick timbers or planks projecting horizontally from the side
of the ship, the chain plates and support shrouds (ropes or
cables) extended upward to hold up the mast, in this case
possibly the mizzen mast, according to Smith.
Much of the outer hull planking from the side of the ship
was gone, but the inner planking, known as ceiling, was there,
with many iron fasteners. The ceiling planking kept the cargo
or ballast from damaging the outer hull planks up the to the
water line. Each frame, or rib, of the ship was formed from
two sets of futtocks, the long, curved timbers chamfered
together to produce the rounded framework of the hull
(Figures 5 and 9). Each pair of futtocks was joined together
with iron pins driven laterally. Several wooden patches were
apparent in the planks of the hull (Figures 13, 14). The planks
were put on mostly with copper and Muntz metal fasteners,
with the iron spikes and pins above the water line (because salt
water would hasten rusting) and the tree nails in various
locations. This was altogether an interesting combination of
old and new materials. Commonly, one metal and one wooden
spike alternated per each plank.
The average plank thickness was 4" and width, 7-11". The

ceiling planking averaged 8"-10" thick; the frames had an
average molded height of 8"-9" and sided thickness (width) of
16", so the futtocks preserved had a sided thickness of 8". The
pine tar and copper sheathing were on the lower section of the
hull, with copper sheathing tacks spaced 14" apart (so the
copper sheets were slightly wider than 14"). To the west side
of the wreck there was a metal, almost cleat-shaped object that
Smith identified as a davit, a hook for hanging a lifeboat.
Another interesting item newly exposed after the July storm
was a carved piece of wood clearly done on a lathe, much like
a finial or decorative knob of furniture (Figure 15). Smith
identified it as part of the chain wale.
The ship is estimated to have been about 100 feet long.
With the stem off shore, it was oriented with the bow toward
the shore. The extant portion, the ship's port quarter, was a
little more than the outer hull, frames and other portions.
There was no decking or evidence of cargo. The ship probably
broke up off shore; Smith noted a weak spot in the inner
planking. This segment of the vessel floated in and became
buried in the sand for over a century, only to be uncovered,
either by some new erosional patterns triggered by the tropical
storms, or by Hurricane Opal, or by the same erosional
patterns that have been in force for a century (Donoghue and
Tanner 1994) that just made it as far inland as this ship by
1996. The rest of the vessel and its cargo probably lie out in
the Gulf somewhere, buried in tons of sand.



-s. d


2~.- T~4

lt. *arc_.J ow



Figure 6. Closeup of only hull plank to retain copper sheathing, curled up on the edge; also visible are round heads of copper
and Muntz metal fasteners (nails or spikes). View facing northwest. This plank is the one extending farthest to the left in
Figure 4.

Context and Identity

Dating the Remains

Some details of construction can help characterize or date
the ship, especially the form of brass (copper-zinc alloy)
known as Muntz metal. This was an alloy of 60% copper and
40% zinc introduced in England in the 1830s, becoming
common there in the 1840s, and used in the U.S. beginning
perhaps a decade later. It was as a lighter, stronger, longer-
lasting, and cheaper alternative to pure copper for ship
sheathing and fastening (Flick 1975; Gould 2000:55). Also
known as yellow metal, it had been developed by George
Frederick Muntz, a British inventor and industrialist from
Birmingham (England). This metal is still produced for
various industrial uses.
Metal sheathing of ships' hulls was known as early as
seventeenth-century China, but did not become common
practice until the late eighteenth century. The sheathing
protected ships in warm waters against burrowing molluscs
and crustaceans, marine organisms such as the Teredo
shipworms, that could damage a wooden vessel seriously
enough to sink it. The metal also prevented buildup of marine
growth on the hull; barnacles and other small creatures that
would attach themselves absorbed poison from the metal and
dropped off, leaving the ship's bottom less encumbered.

Sheathed ships were faster and more maneuverable and could
remain in warm waters for a couple years, with less time in the
dock for repairs, while unsheathed vessels might have a half-
year before needing service (Bingeman et al. 2000). By the
end of the nineteenth century, sheathing was replaced by
anti-fouling paints with similar chemical properties.
The fact that the Cape St. George shipwreck vessel was
sheathed in copper, with only Muntz metal fasteners, does not
necessarily indicate it was built earlier in the history of Muntz
metal (while the colors of the yellow brass and copper in the
Cape St. George shipwreck were clearly different, it should be
noted that the two metals are sometimes not visually distin-
guishable from each other, requiring x-ray analysis for
identification [Charlie Pearson, personal communication,
2005]). Copper continued to be used by many, such as the
British royal navy, who had less concern for economy, long
after others had turned to Muntz metal sheathing (Flick
1975:77). The royal navy had copper-sheathed wooden
workboats until nearly the end of the twentieth century
(Bingeman et al. 2000:224). The Muntz metal does place the
shipwreck after 1840. Roger Smith estimated that it dated to
between the 1870s and 1890s.

Cultural Context

Shipwrecks do not occur as isolated incidents in a cultural


2006 VOL. 59(2)


Figure 7. Copper sheathing fragments and tacks displayed at the ANERR education center (scale in inches). Photo by Erik

_3 -~.-
~E 2f- --

Figure 8. Closeup of planking with three different types of fasteners (scale in cm).

vacuum (Lenihan 1983:49), but need to be understood within
the larger global networks of social, economic, and political
interaction. The nineteenth century, especially its last quarter,
saw more shipwrecks in the Gulf than any previous time
periods, probably not only because of increased population and
commerce, but also due to increased international trade

(subsequently, there were even more wrecks in the twentieth
century [Pearson et al. 2002]). Beyond placing it in time,
there is the need to place the ship in historic and anthropologi-
cal context. The wrecking was a historical event but embed-
ded in ongoing natural and cultural processes and conditions
responsible for it (Gould 2000:13). The concept of the




Figure 9. Planks possibly from bottom of hull, with a few
(larger arrows) and tree nail (smaller arrow) fasteners left.

"maritime cultural landscape" includes interpreting the
remnants of shipwrecks but also of harbors, fishing, shipping,
navigation, and everything else pertaining to the use of the sea,
on land as well as underwater (Westerdahl 1992), and under-
standing their relationships within larger economic and
political systems. Archaeologically speaking, the wreck was
not the termination of the ship, since after its intended use, a
part of it transformed and remained to experience the various
slow-to-violent processes of the Gulf, and then to enter a third
"phase" of life as an archaeological site (Steffy 1994:189-190).
The locations and density of shipwrecks along the Gulf of
Mexico are a productof complex historical and natural factors,
rangingfrom imperialism, commerce, warfare, and technolog-
ical change to currents, winds, shoals, reefs, and storms
(Garrison 1998). The Apalachicola Bay and River system had

a long history of maritime culture and played an
important role in international shipping net-
works, especially because the river enabled
transportation well into central Georgia. But it
was always plagued with high transportation
costs due in part to the shallowness of the bay
and passes (Owens 1966). The hazards of navi-
gation increase as land is approached, especially
in narrow channels and shifting shallows around
barrier islands. Even for experienced navigators
who may be familiar with a particular stretch of
coast, barrier islands are treacherous since they
are constantly changing, and many wrecks have
been attributed to these shallow conditions.
However, this may not be what happened to the
Cape St. George shipwreck, given its position on
the farthest tip of land protruding out into the
Gulf. Most likely the cause of its demise lay in
troubles on the open sea.
We do know that mid- to late-nineteenth-
century maritime commerce increasingly in-
volved bulk cargoes of commodities such as
lumber, cotton, or other agricultural products
(Gould 2000:238-239). The Cape St. George
shipwreck likely represents transport of the
important Apalachicola cotton crop or timber or
naval stores, fueling textile, shipping or other
important manufacturing systems of the mid-
Industrial Revolution. The best guess is that it
was transporting lumber from northwest Florida
to the northeast or across the Atlantic, but there
are other possibilities. There was not much
shipping during the Civil War (Rogers and
Willis 1997; Smith et al. 1997:18) because of the
Union blockade of the important cotton port of
Apalachicola, which was of enormous strategic
value for its river access to the interior South.
After the war the region continued producing
Muntz metal cotton, but in smaller quantities. Even before the
Civil War, cotton production was declining and
the area was becoming more important for its
After the war, the traffic across the sea increased, and the
growing forest products industries helped alleviate a postwar
economic slump. Commerce expanded with European
customers, especially Scandinavian ships, for the timber from
both the mainland and the barrier islands (Rogers and Willis
1997:68-69; Burns 2002). Sawmills, lumber settlements, and
turpentine camps and stills multiplied in the region. Ships
involved in transporting lumber even had specialty cargo
hatches; Meide et al. (2001:24-25) show a photo of an 1880
Finnish lumber brig with copper sheathing and a specialized
bow port just above it for easier loading. A new Underwater
Archaeological Preserve dedicated by the State of Florida is
the wreck of the Norwegian iron-hulled barque Lofthus, which
had been built in Britain in 1868 and went down in the
Atlantic with a cargo of northwest Florida lumber en route to


2006 VoL. 59(2)


Figure 10. Closeup of brass/Muntz metal spike head; note circul;
cut into the plank, for unknown function (scale in cm).

Buenos Aires in 1898 (Florida Department of State 2004).
The Cape St. George ship was a deep sea vessel probably
involved in the forest products industries of the second half of
the nineteenth century. Besides lumber, there could have been
tar, pitch, charcoal, gum, turpentine, rosin, pine oil, and many
other derivatives, important as naval stores and in other
manufacturing processes. Florida archaeologists are familiar
with sites associated with these industries, such as sawmills,
turpentine stills, and the ubiquitous sherds of red clay Herty
cups. The remote region of the lower Apalachicola valley and
barrier islands was part of a vast international shipping
network transporting these products overseas (Forney 1985;
Rogers 1986; Bond 1987; Rogers and Willis 1997:83; Butler
1998). Most likely between 1870 and 1890, the ship repre-
sented by our wreck was hauling lumber out to some Atlantic
destination or goods into the region to be sold for lumber,
though it could have been involved also with other cargo
produced in northwest Florida, including cotton, honey,
oranges, other produce, cattle, fish, and even sponges (Rogers
1986; Meide et al. 2001.). If it were coming into port it could
have had a cargo of finished merchant goods, or perhaps salt
from England, fruit from Cuba, wine or other products (Owens


Possible Identities

The identity of the Cape St. George shipwreck is
unknown. Of the wrecks recorded near St. George Island
(Singer 1998; Damour et al. 2001; Meide et al. 2001;
Damour 2002; Horrell 2005), it is too late to be part of the
romantic stories of Monsieur Viaud's Tiger (lost 1766;
Fabel 1990) or William Augustus Bowles's Fox (wrecked
1799), both of which are supposedly located at the east
end of St. George Island, as much as 40 km (25 miles)
away. The George P. Arnay was a sloop bound from St.
Marks to Apalachicola reported from the Dog Island
Station as lost on the north bank of St. George Island on
ar 28 December 1872 (Singer 1998:27); this placement on
the north side of the island makes it an unlikely candidate.
TheMarryMe was a 17-ton schooner built in 1878, going
from Apalachicola to St. George Island and lost on the west
end of the island in a gale on 7 October 1897 (Singer
1998:34), but it was far too small to be the Cape St. George
wreck, which, based on its size, represented well over a 100-
ton ship.
Several shipwrecks are known from around Dog Island. A
terrestrial survey I did in 1995 described a couple of them
easily visible near the shore, and a large scale underwater
archaeology and remote sensing project undertaken from
1999-2001 by Michael Faught and Florida State University
(FSU) students investigated several wrecks and possible
wrecks all around the island's waters. One vessel is likely a
late nineteenth-century fishing schooner, possibly the
Priscilla, located on the Gulf side (Haiduven et al. 1987,
White et al. 1995:26-27; Meide et al. 2001:88-99; Damour
2002; Horrell 2005:113-14). Another wreck, on the bay side,
is likely the Norwegian lumber bark the Vale, one of a group
of 11 ships blown down by an 1899 hurricane (Wright 1990;
Huntsman 1991; White et al. 1995:27-30; Meide et al.
2001:102-131, Appendix H). While some of those other 11
ships may be likely candidates for the Cape St. George wreck,
they seem all to have gone down in the bay, not the Gulf, and

Figure 11. Muntz metal fasteners on display at ANERR education center (scale in inches). Photo by Erik Lovestrand.




Figure 12. Iron spikes on display at ANERR education center (scale in inches). Photo by Erik Lovestrand.

the 1899 date might be slightly too late.
The reporting on the underwater archaeology of the Dog
Island area (now available online, Florida Department of State
2005) is wonderfully thorough in describing shipping of the
times and listing known vessel losses of the area (Meide et al.
2001: Appendix D, Table 22; Horrell's [2005] dissertation also
lists all mid-nineteenth-century Apalachicola vessels, and
those from 1870-1900). From the second half of the nine-
teenth century, the wrecks included barks taking cotton to
Antwerp and Amsterdam, schooners and brigs, ships with
lumber going to Key West and to Central America. Some did
indeed end up sunk by activities of the Civil War. Others went
down for various reasons besides storms; some caught fire,
including one which was struck by lightning, and there was
even one collision. Perhaps a good candidate for the Cape St.
George shipwreck is the George Gilchrest, a 438-ton brig from
New York sailing from Pensacola to Nicaragua with a load of
yellow and pitch pine when it sprang a leak in a storm and
sank some 60 miles south of Cape St. George lighthouse
(Singer 1998:28; Meide et al. 2001:183).
The most likely scenario is that the Cape St. George ship
went down and broke up far out to sea, or hit shoals and broke
up, after which a portion floated in to be buried in the island's
shifting sands. The offshore bathymetry of the Cape St.
George area (Florida Department of Environmental Protection
2002) shows a large and potentially treacherous unmarked
shoal that was clearly a navigation hazard for centuries. The
vessel may have broken up farther southeast and floated
westward, this portion ending up snagged on the projection of
the cape and buried, perhaps by the same storm that destroyed

Discussion and Aftermath

More research is needed on this shipwreck, including
species identification of the wood samples. Roger Smith
continues to research the possible identity of this vessel, but we

may never know its name or circumstances. Similar ship
timbers wash up on many of the country's beaches. Since
Hurricane Floyd hit South Carolina in 1999, for example,
frames and planks with copper sheathing, tree nails, and iron
spikes have been reported scattered the length of Folly Island,
probably from vessels that went down in the Civil War (Harris
2000:25). Suddenly exposed wrecks are certainly common in
Florida. As I was writing this article, Tropical Storm Arlene
had just uncovered another one in spring 2005 along northwest
Florida's coast in Walton County (Civil 2005). The Florida
Museum of Natural History's (2005) website explains the
history of ships and various famous wrecks, and the Florida
Department of State (2004, 2005) has a maritime heritage trail
and other public information on wrecks. In the aftermath of
the tragedies of 2005's hurricanes Dennis, Katrina, Rita, and
others on the northern Gulf, it will be interesting to see what
becomes exposed as the waters recede and the sands and debris
are cleared.

Cultural and Natural Formation Processes

Cape St. George is public land. The fast-traveling news of
the shipwreck meant that looters came to take planks, copper
sheathing, fasteners and other bits, and even during the week
between our field trips. Curiously cut (see Figure 9) or broken
portions of planks may have resulted from removal of brass or
copper pins as souvenirs. There was talk among collectors
about grabbing things from this wreck for months, though the
ANERR staff stressed education, preservation, and monitor-
ing. The island is only accessible by boat, but many people
visit; the research station has been vandalized. In addition,
such a high-energy environment as the crashing Gulf waves
also meant the wreck was not going to last long. Wind, waves,
currents, tides, all would have influenced what happened to the
wrecked ship, but human interference is the most destructive
force for a shipwreck in a terrestrial context; still, preservation
on land is improved if the wreck lies in soft substrates such as

2006 VOL. 59(2)



'5-SL. J- -.~ ~T.

Figure 13. Two of the 5 remaining iron chains extend up from a
the chain wale, amid a tangle of tree branches; between them at I
a small rectangular patch, just below which are the heads of tw
of Muntz metal (left) and iron (right).

sand (Muckleroy 1998:267-70).
Maritime site formation processes usually result in more in
destruction than preservation, especially in the dynamic
shoreline environment. Muckleroy (1978; Delgado
1997:156-57) distinguished "extractingfilters," which remove
materials from the site, and "scrambling devices," which alter
and rearrange materials. The wrecking process itself is an
example of both, as are the Gulf dynamics, which would
include "extraction" when this large portion of the wreck
floated away and ended up on the shore. Remains became
scrambled then buried on the shore. Various storm processes
can continue this back-and-forth action for decades (or hours
or centuries). While the June 2005 storm Arlene exposed that
other panhandle wreck, the July 2005 Hurricane Dennis sent
a storm surge of bay water up to cover the land with sediment.

SSome shipwrecks become centers for coloniza-
S tion of marine life, but the Cape St. George wreck,
.- right on the shore in the pounding waves, became
a place for snagging tree branches and roots ripped
from their once-firm anchor on the dune ridge
back from the water. These snags are testimony to
the continuous loss of shore land as the waterline
advances back toward the forest, partly from still-
rising, post Pleistocene sea levels, and partly from
other factors (since archaeological evidence also
demonstrates extreme erosion of barrier islands
from the bay side in the last few decades, I suspect
there are additional causes, probably relating to
human activity on the planet). Official measure-
ments show global sea level rose by an average 10-
20 cm during the twentieth century, with greater
numbers along particularly vulnerable coasts such
as the Gulf and the Atlantic (U.S. Department of
State 2002:103).
The most important route around the Gulf for
the whole age of sail was the Gulf Loop Current
(Garrison 1998:305), which enters the Gulf
through the Yucatan Channel and the Straits of
Florida. It moves in a great arc around to the
mouth of the Mississippi River, driven by easterly
winds, at 12 to 35 nautical miles per day. Moving
northward and eastward it sends eddies and spi-
rals, termed gyres, in various directions at different
times. One gyre circles into the northeastern Gulf
east of the Mississippi Delta, spinning counter-
clockwise in the winter months and in the opposite
direction in the summer (Gore 1992: 68-72). The
December Loop Current (Gore 1992: Figure 9B)
shows two such gyres approaching closest to land
at the Apalachicola delta and barrier islands. So
the currents in panhandle Florida, running from
west to east to join the Gulf Loop in winter and
piece of
from east to west in summer (Smith et al. 1997:4),
ottoi i would have facilitated navigation toward the
o spes, barrier islands and prominent cape. Only with the
coming of steam-powered vessels were shipping
and therefore shipwreck patterns and routes signif-
icantly changed (Garrison 1998). These currents, coupled
with the effects of storms, help account for many of the
Apalachicola barrier island wrecks. Cape St. George itself is
generally a constructional feature instead of an erosional
product (Schade 1985:123), meaning natural forces deposit
things there.
There is little one can do to preserve a resource such as this
shipwreck, since one cannot stop the sea. The sandy bottom
easily moves around, and single storm surges may have
covered and uncovered the remains repeatedly. It was impor-
tant to record the information on the shipwreck before it was
gone. The site was a piece of history of a type not usually
available unless conditions are just right for exposure. Public
archaeology was served when ANERR educational programs
brought people to learn from the wreck over the months. A




Figure 14. Closeup of hull patch, rectangular wooden fragment at top, with round head of iron spike visible to the right of
the scale and shells; at bottom left is a round head of a wedged tree nail (wedge is vertical central piece of slightly lighter-
colored wood; scale in cm).

small display at the ANERR auditorium in Apalachicola
shows the metal items recovered from the wreck. KC Smith
brought students in the Museum of Florida History summer
camp for two successive years (the "From Dugouts to Dou-
bloons" camp) to learn measuring and recording at this wreck.
Within little more than two years, the wreck had com-
pletely disappeared. In September of 1998 only a small piece
of the northwesternmost portion was visible, a few meters out
in the water. By 2000 it was gone and the beach was smooth
and flat, with no tree tangles, as if nothing had ever been
there. By the following year it changed again; a little beach-
front escarpment appeared, as if chopped vertically by the sea
as it edged closer to the forest.

Further Research

The persistence of barrier islands themselves "in the face
of the immense amount of energy imparted to them from the
sea is one of nature's remarkable idiosyncracies" (Champion
1996:1). More amazing is the persistence of shipwrecks along
barrier island shores. On Dog Island, those wrecks noted
above, which are probably the Vale (lumber ship) and the
Priscilla (fishing boat) have sat on the bay shore and the
higher-energy Gulf shore, respectively, for at least a century
and remained in reasonable shape for study. At the east end
of Cape St. George itself, a recently abandoned, rusting, steel-
hulled fishing boat beached on the Gulf shore in the late 1990s
was nearly covered in sand high up on the shore, with little
more than the upper cabin exposed; by the 2000s it was out in
the water, more buried in sand, with its roof and cabin torn off.
Whether the Cape St. George shipwreck is being preserved
under the sand could be determined by subsurface testing. The
most likely scenario is that it is now underwater and buried in

the bottom sands.
Lately there has been recognition that beached shipwrecks
offer much new knowledge for science and history. They are
sometimes dismissed since they are thought to be disintegrated
and lacking controlled provenience information, as opposed to
offshore, underwater wrecks. They have been called restless
"ghosts, tossing about with each change of wind, tide, and
season" (Bright 1993:91), resting buried then emerging often
to continue their journeys. However, in reality, beached
wrecks are a valuable, underutilized source for archaeological
inquiry that can add a great deal to the shipwreck database
(Russell 2004:369; Bright 1993; Delgado 1997:57-8; Murphy
1983:77). This is especially true for the Cape St. George
wreck since in Florida many beached wrecks consist of
scattered artifacts with little hull structure. Though the Cape
St. George wreck was only a portion of a large ship, with no
evidence of cargo, it offers good potential for further study. In
Delgado's (1997:58) classification system, it would be a
"buoyant hull fracture," in which the vessel washes ashore and
breaks up and the components scatter to be buried and
reexposed at different times.
The largest part of the vessel, as well as items of metal,
ceramic, and other materials that do not float, and probably
much of the cargo unless it was lumber, are doubtless lying out
on the sea floor or buried in sand somewhere else, while the
vessel section recorded floated inland. Some cargo may have
been salvaged by whoever was aware of the wreck. There may
be further historic information on this wreck that could be
unearthed with an intensive search (the many shipwreck
inventories range from the Florida Master Site File to various
federal sources; Garrison 1998:304). Further fieldwork is also
possible using remote sensing such as a magnetometer or side-
scan sonar on the beach and in shallow water (Hudson et al.

2006 VOL. 59(2)



Figure 15. Visible only between waves, and protruding from a
encrusted with sea life is a carved knob, part of the chain wale,
right a bent iron spike.

The Cape St. George shipwreck data may also help us
understand natural processes and relationships with human
action. Since the wreck emerged several months after Hurri-
cane Opal, it is unclear whether this storm began some new
erosional patterns that led to the site's exposure. Opal
uncovered the underlying peat on Dog Island's Gulf shore,
blowing away the deep sand cover and exposing what might
have been a very ancient shoreline or even mainland; the sand
filled back in again within a few months. The presence of peat
and tree stumps on these islands show that they have migrated
northward, over the back-barrier marsh, exposing it to wave
action (Davis 1997:183). This is probably the same set of
processes that operated to expose and then reclaim the Cape
St George shipwreck. On Dog Island, past storms have
exposed peat strata containing fiber-tempered pottery 2000 to
3000 years old.
The Gulf shoreline's normal migration northward toward
the mainland has in recent years seemed excessive. At Cape
St. George, the southwest-facing coast where the shipwreck
was exposed is actually accreting, while the southeast-facing
coast is retreating at about one meter per year (Donoghue and
Tanner 1994). Cape San Bias, at the bottom of St. Joseph
peninsula to the west, is so rapidly changing that it has moved
eastward about a kilometer over the last century (Rupert
1991:10; Randazzo and Jones 1997:167). The action ofjust
one typical summer storm exposed some 10-15 more meters of
our wreck nearly between our two July visits. Yet bayshore
prehistoric sites on this island were not severely affected,

continuing to erode at their usual rate into the water;
nor have those Dog Island wrecks disappeared after
a century.
Further research on the Cape St. George vessel
should also include more on its social and economic
context, both local and non-local. Perhaps it was an
old or poorly constructed ship at greater risk for
disaster. The hull patches may indicate excessive
wear and repair. Iron fasteners were easier and
quicker to attach than wooden ones, allowing for the
possibility of less-skilled workers to construct the
ship's hull (Gould 2000:5), but the combination of
all the different kinds of fasteners may suggest
sophistication of design or just the opposite, using
whatever was handy to keep the vessel going. Burns
(2002:17) has noted how Norwegian prominence in
late nineteenth-century transport of bulk cargo,
especially timber from northwest Florida, relied
upon secondhand sailing vessels near the end of
their working lives, cheap to buy in the new age of
In the end, I hope publication of this information
on the Cape St. George shipwreck will help raise
awareness of the fragility of this type of cultural
plank resource that gives such tantalizing hints of the
and at human past. Stored in the USF archaeology lab are
all project records, including the videotapes that
show the rapid change in the wreck's condition over
the course ofjust a few days. The brief study of this wreck was
like a 10-minute visit to surface-collect at a terrestrial site
(though we spent many hours there). However the difficulty
of access to the site and pounding destruction of the waves that
took away pieces of the wreck even as we watched meant that
very little could be recorded in the time allotted. It was an
ephemeral site on an only slightly less ephemeral landform,
and thus a frustrating, heart-wrenching experience for an
archaeologist trying to salvage as much as possible of the
information before it was gone (now you see it in the sea, now
you don't).


I thank Roger Smith, KC Smith, and the other personnel of the
Florida Division of Historical Resources for taking the time to visit
the shipwreck site and help with technical details of its description,
as well as for reviewing this manuscript and offering useful sugges-
tions. Apalachicola National Estuarine Research Reserve personnel,
especially Jimmy Moses and Pat Millender, first located the wreck
and helped with transportation and other work. Other helpful
ANERR staff were Chip Bailey, Terry Longieliere, and former
director Woody Miley. My 1996 field crew cheerfully took a break
from the dark forest to help record the wreck on the brilliant white
beach; they were crew chief Scott Grammer, field students Kelly
Driscoll, Rebecca (Jacob) Harris, and Nikki Rawlinson, and peren-
nial volunteer Tony White. Others to whom lam enormouslygrateful
for reading this manuscript and offering comments and corrections
are Joe Donoghue, Jeff Duvernay, Lee Edmiston, Mike Faught, Rae
Harper, Chris Horrell, Erik Lovestrand, Roy Ogles, and Charlie
Pearson, as well as an anonymous reviewer and editor Ryan Wheeler.




References Cited

Bingeman, John M., John P. Bethell, Peter Goodwin, and Arthur T.
2000 Copper and Other Sheathing in the Royal Navy. Interna-
tional Journal ofNautical Archaeology 29(2):218- 29,

Bond, Stanley C., Jr.
1987 The Development of the Naval Stores Industry in St. Johns
County, Florida. The Florida Anthropologist 40:187-202.

Bradley, Omar N., and Clay Blair
1983 A General's Life. Simon and Schuster, New York.

Bums, Jason M.
2002 TheLife and Times ofaMerchantSailor. The Archaeology
and History of the Norwegian Ship Catharine. Plenum
Publishers, New York.

Bright, Leslie S.
1993 Beached ShipwreckDynamics. In Underwater Archaeology
Proceedings from the Society for Historical Archaeology
Conference, edited by S. Smith, pp. 91-94. Society for
Historical Archaeology, Tucson.

Butler, Carroll B.
1998 Treasures of the Longleaf Pines. Naval Stores. Second
edition, revised. Tarkel Publishing, Shalimar, Florida.

Champion, Glen
1996 Historic ShorelineMovementRates, Dog Island, Northwest
Florida. Undergraduate honors thesis, Department of
Geology, Florida State University.

Civil, Heather
2005 Century-Old Schooner Washes Up in Walton. Northwest
Florida Daily News, Sunday July 3, 2005, p.1, Front

Coles, David J., and David Gregory
2005 Florida World War IIHeritage Trail. A Florida Heritage
Publication, Florida Division of Historical Resources,

Damour, Melanie J.
2002 Looking for HMS Fox: A Model for Applying Barrier
Island Geomorphology to Shipwreck Survey. Unpublished
Master's thesis, Department ofAnthropology, Florida State
University, Tallahassee.

Damour, Melanie J., Christopher E. Horrell, and Michael Faught
2001 Dog IslandShipwreckSurvey, Report ofField Operations:
June 26" through August 4', 2000. Florida State Univer-
sity Program in Underwater Archaeology Research Report
No. 9.

Davis, Richard A., Jr.
1997 The Evolving Coast. Scientific American Library, New

Delgado, James P.
1997 Encyclopedia of Underwater and Maritime Archaeology.
Yale University Press, New Haven, Connecticut.

Dickinson, Matthew, Lawrence W. Huntsman, and Jora M. Young
1992 JeffLewis Wilderness PreserveManagementPlan. Unpub-
lished report produced jointly by the Nature Conservancy,
the Barrier Island Trust, and Florida State University,

Donoghue, Joseph F., and William F. Tanner
1994 Effects of Near-term Sedimentologic Evolution on the
Lifetime of Estuarine Resources. National Oceanic and
Atmospheric Administration Technical Memorandum
NOS-SRD 27. Washington, D.C.

Donoghue, Joseph F., and Nancy Marie White
1995 Late Holocene Sea Level Change and Delta Migration,
Apalachicola River Region, Florida. Journal of Coastal
Research 11(3):651-63.

Fabel, Robin F. A., editor
1990 Shipwreck and Adventures of Monsieur Pierre Viaud.
University of West Florida Press, Pensacola.

Flick, Carlos
1975 Muntz Metal and Ships' Bottoms: The Industrial Career of
G. F. Muntz. Transactions of the Birmingham and
Warwickshire Archaeological Society 87:71-84.

Florida Department of Environmental Protection
2002 Reconnaissance Offshore Sand Search. Bureau of Beaches
and Coastal Systems, Tallahassee. online at http://ross.urs-

2004 Hurrican Ivan, Beach and Dune Erosion and Structural
Damage Assessment and Post-Storm Recovery Plan for the
Panhandle Coast ofFlorida. Bureau ofBeaches and Coastal
Systems. Online at

2005 Hurricane Dennis, Beach and Dune Erosion and Structural
Damage Assessment and Post Storm Recovery Plan for the
Panhandle Coast ofFlorida. Bureau of Beaches and Coastal
Systems. Online at

Florida Department of State
2004 Florida's Underwater Preserves. Office of Cultural and
Historical Programs. Online at http://
dhr.dos. state.,
accessed in September 2005.

2005 Florida State University Program in Underwater Archaeol-
ogy Research Reports. Office of Cultural and Historical
Programs. Online at archaeol-
ogy/ underwater/fsu_pua/, accessed in October 2005.

Florida Museum of Natural History
2005 Down Like Lead: 400 Years of Florida Shipwrecks. Online
at, accessed in September

Fomey, Sandra Jo
1985 The Importance of Sites Related to the Naval Stores
Industry in Florida. TheFloridaAnthropologist38:275-81.

Garrison, Ervan G.


2006 VOL. 59(2)


1998 ADiachronic Study of Some Historical and Natural Factors
Linked to Shipwreck Patterns in the Northern Gulf of
Mexico. InMaritimeArchaeology. AReaderofSubstantive
and Theoretical Contributions, edited by L. Babits and H.
Van Tilburg, pp.303-316. Plenum Press, New York.

Gore, Robert H.
1992 The Gulf of Mexico. A Treasury of Resources in the
American Mediterranean. Pineapple Press, Sarasota,

Gould, Richard A.
1983 Looking Below the Surface: Shipwreck Archaeology as
Anthropology. In Shipwreck Anthropology, edited by R.
Gould, pp. 3-22. University of New Mexico Press, Albu-

2000 Archaeology and the Social History of Ships. Cambridge
University Press, New York.

Haiduven, Richard, KC Smith, D. Muncher, J. Nolin, and A. Nolin
1987 Notes concerning the Dog Island shipwreck #1, 8Fr813. On
file at the Florida Division of Historical Resources, Talla-

Harris, Lynn
2000 Underwater Archaeology: Lowcountry Waterbodies Yield
a Diversity of Artifact Types. Legacy 5 (1):24-25. South
Carolina Institute of Archaeology and Anthropology,
University of South Carolina, Columbia.

Horrell, Christopher Earl
2005 Plying the Waters of Time: Maritime Archaeology and
History on the Florida Gulf Coast. Ph.D. dissertation,
Department of Anthropology, Florida State University
(available online at

Hudson, Jack, Kay G. Hudson, and Harry W. Rhodes III
1998 Magnetic Search and Survey in Shallow Water and Beach
Areas. In Maritime Archaeology. A Reader ofSubstantive
and Theoretical Contributions, edited by L. Babits and H.
Van Tilburg, pp.375-77. Plenum Press, New York.

Huntsman, Larry
1991 Did youKnow...? The BarrierIsland TrustNewsletter 4:3-
5 (October)

1992 Did you Know...? The BarrierIsland TrustNewsletter 6:3-
8 (July)

Kaufman, Wallace, and Orrin Pilkey
1979 The Beaches are Moving. The Drowning ofAmerica's
Shoreline. Anchor Press, Garden City, New York.

Lenihan, Daniel J.
1983 Rethinking Shipwreck Archaeology: A History of Ideas and
Considerations forNewDirections. InShipwreckArchaeol-
ogy, edited by R. Gould, pp. 37-64. University of New
Mexico Press, Albuquerque.

Livingston, Robert J.
1989 Dog Island: A Barrier Island Ecosystem. Center for
Aquatic Research and Resource Management, Florida State

University, Tallahassee.

McCarthy, Kevin M.
2004 Apalachicola Bay. Pineapple Press, Sarasota.

Meide, Charles T., James A. McClean, and Edward Wiser
2001 Dog Island Shipwreck Survey 1999: Report ofHistorical
and Archaeological Investigations. Florida State Univer-
sity Program in Underwater Archaeology Research Report
No. 4.

Morton, Robert A., Tara L. Miller, and Laura J. Moore
2004 NationalAssessment ofShoreline Change: Part 1, Histori-
cal Shoreline Changes and Associated Coastal Land Loss
Along the U.S. Gulf of Mexico. U.S. Geological Survey,
Center for Coastal and Watershed Studies, St. Petersburg,

Muckleroy, Keith
1978 Maritime Archaeology. Cambridge University Press, New

1998 The Archaeology of Shipwrecks. InMaritime Archaeology.
A Reader of Substantive and Theoretical Contributions,
edited by L. Babits and H. Van Tilburg, pp.267-90. Plenum
Press, New York.

Murphy, Lany E.
1983 Shipwrecks as Data Base for Human Behavioral Studies.
In Shipwreck Archaeology, edited by R. Gould, pp. 65-88.
University of New Mexico Press, Albuquerque.

Owens, Harry P.
1966 Apalachicola Before 1861. Ph.D. dissertation, Department
ofHistory, Florida State University. University Microfilms,
Inc., Ann Arbor, Michigan.

Pearson, Charles E., Stephen R. James, Jr., Michael C. Krivor, S.
Dean El Darragi, and Lori Cunningham
2002 Refining and Revising the GulfofMexico Outer Continen-
tal Shelf Region High Probability Model for Historic
Shipwrecks. Revised DraftReport to theMinerals Manage-
ment Service, U. S. Department of the Interior.
Panamerican Consultants, Inc., Memphis, Tennessee, and
Coastal Environments, Inc., Baton Rouge, Louisiana.

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

Rogers, William Warren
1986 Outposts on the Gulf. St. George Island and Apalachicola
from Early Exploration to World War II. University of
West Florida Press, Pensacola.

Rogers, William Warren, and Lee Willis, III
1997 At the Water's Edge. A Pictorial and Narrative History of
Apalachicola and Franklin County. Donning Company
Publishers, Virginia Beach, Virginia.

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




Russell, Matthew A.
2004 Beached Shipwrecks from Channel Islands National Park,
California. Journal ofField Archaeology 29:369-84.

Schade, Carleton J.
1985 Late Holocene Sedimentology of St. George Island,
Florida. Unpublished Master's thesis, Department of
Geology, Florida State University, Tallahassee.

Singer, Steven D.
1998 Shipwrecks of Florida. Second edition. Pineapple Press,
Inc., Sarasota.

Smith, Roger C., James J. Miller, Sean M. Kelly, and Linda G.
1997 An Atlas ofMaritime Florida. University Press of Florida,

Steffy, J. Richard
1994 Wooden Ship Building and the Interpretation of Ship-
wrecks. Texas A&M University Press, College Station.

U.S. Department of State
2002 U.S. Climate Action Report 2002. Washington, D.C.

Watson, Patty Jo
1983 Method and Theory in Shipwreck Archaeology. In Ship-
wreck Anthropology, edited by R. Gould, pp. 23-36.
University of New Mexico Press, Albuquerque.

Westerdahl, Christen
1992 The Maritime Cultural Landscape. International Journal
ofNauticalArchaeology 21(1):5-14.

White, Nancy Marie
1996 Archaeological Investigations of the 1994 RecordFlood
Impacts in the Apalachicola Valley, Northwest Florida.
Report to the Division of Historical Resources, Florida
Department of State, Bureau of Historic Preservation,
Tallahassee. University of South Florida Department of
Anthropology, Tampa.

1997 Barrier Island Archaeology, Northwest Florida. Paper
presented at the annual meeting of the Society for Ameri-
can Archaeology, Nashville, April.

2005 Archaeological Survey of the St. Joseph Bay State Buffer
Preserve, Gulf County, Florida. Report to the Division of
Historical Resources, Florida Department of State, Bureau
of Historic Preservation, Tallahassee. University of South
Florida Department of Anthropology, 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 Division of
Historical Resources, Florida Department of State, Bureau
of Historic Preservation, Tallahassee. University of South
Florida Department of Anthropology, Tampa.

White, Nancy Marie, Keith D. Ryder, Scott M. Grammer, and Karen
L. Mayo
1995 Archaeological Survey of Dog Island, Franklin County,

Florida. Report to the Barrier Island Trust, Tallahassee.
University of South Florida Department of Anthropology,

Wright, Chip
1990 Dog Island Wreck #2. Report on file, Florida Division of
Historical Resources, Tallahassee.


2006 VOL. 59(2)

Florida Anthropological Society Chapters

- 9

1) Archaeological Society of Southern Florida
2495 NW 35th Ave., Miami, FL 33142

2) Gold Coast Anthropological Society
106 West Camino Real, #109, Boca Raton, FL 33432

3) Central Florida Anthropological Society
P.O. Box 947544, Maitland, FL 32794-7544

4) Central Gulf Coast Archaeological Society
P.O. Box 82255, St. Petersburg, FL 33682

5) Indian River Anthropological Society
3705 S. Tropical Trail, Merritt Island, FL 32952

6) Kissimmee Valley Archaeological and Historical Con
195 Huntley Oaks Blvd., Lake Placid, FL 33852

7) Emerald Coast Archaeological Society
333 Persimmon Street, Freeport, FL 32435

8) Panhandle Archaeological Society at Tallahassee
c/o The Tallahassee Trust for Historic Preservation
423 E. Virginia Street, Tallahassee, FL 32301

9) Pensacola Archaeological Society
P.O. Box 13251, Pensacola, FL 32591

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

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

12) Southwest Florida Archaeological Society
P.O. Box 9965, Naples, FL 34101

r,' .^^/

13) Time Sifters Archaeology Society
P.O. Box 25883, Sarasota, FL 34277-2883

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

15) Warm Mineral Springs Archaeological Society
P.O. Box 7797, North Port, FL 34287


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 may be directed to:

The Editor
The Florida Anthropologist
P.O. Box 6356
Tallahassee, FL 32314

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.



Department ofLiberal Arts, Savannah College ofArt and Design, Savannah, GA 31402-3146

Nearly fifty years ago, William Sears first reported the
occurrence in northeastern Florida2 of a distinctive aboriginal
pottery tempered with crushed St. Johns sherds that he labeled
Colorinda. On the basis of stratigraphic evidence he originally
assigned this unique sherd-tempered ware to A.D. 700-1000
(Sears 1957:30), but later amended its temporal range to A.D.
800-1400 (Sears 1959:16). Since Sears's seminal work, very
little has been written about this pottery type or its associated
archaeological culture. As detailed in this paper, Colorinda
was a short-lived and geographically restricted way of life
centered on the lower or northern St. Johns River during the
ninth century A.D. Beginning with the pioneering efforts of
William Sears (1957, 1959) and including the results of recent
excavations at the Cedar Point site (8DU81) (Ashley 2006, this
issue), my intent is to highlight known Colorinda sites and
synthesize archaeological data from these sites to offer the
following overview on the Colorinda archaeological culture
(Figure 1).

Placing Colorinda in Context

Colorinda-as a pottery type and an archaeological
culture-has garnered very little attention outside extreme
northeastern Florida. Even here, it is infrequently mentioned
and rarely discussed, other than in passing. But Colorinda has
not always been so under publicized. In the late 1950s Sears
(1957:25-26) was the first to report its presence and he
formally defined Colorinda pottery. The name comes from
Colorinda Creek, a tidal creek in the vicinity of a series of sites
(8DU58-62) Sears tested in 1955, about 8 km west of the
mouth of the St. Johns River. Today, these sites are located
within the spatial boundaries of the Theodore Roosevelt
Preserve, property within the Timucuan Ecological and
Historic Preserve. Based on his excavation results Sears
(1957:28-31) boasted that Colorinda was one of the four major
ceramic complexes in the lower St. Johns area, with the other
three including the earlier "Deptford" and "Sand-Tempered
Plain" and the later "St. Johns IF' complexes.

Colorinda Series Pottery

Sears (1957:25-26) described Colorinda pottery as a gritty
ware tempered with large, angular St. Johns sherd fragments.
The specific use of crushed St. Johns sherds, as a tempering
agent, is the hallmark of Colorinda pottery. St. Johns pottery
is unique in that its paste contains sponge spicules, which
appear as needle-shaped rods under magnification (Borremans

and Shaak 1986; Cordell and Koski 2003; Rolland and Bond
2003). Microscopic examination of Colorinda sherds clearly
reveals sponge spicules in the tempering material (grog), but
not in the Colorinda paste itself. Recent analysis of Colorinda
sherds recovered from the Cedar Point site (8DU81) indicates
that non-spiculate sherd inclusions occur as well, but far less
frequently than spiculate sherd tempering (Ashley 2006, this
Within the paste of Colorinda pottery, sherd inclu-
sions-typically 2-6 mm in size- range in color from white
to buff to gray, and, as Sears noted, the presence of sherd
inclusions gives the paste a "lumpy and contorted" texture.
Although Sears (1957:25) identified the presence of "a fair
amount of moderately-sized grit" in the paste of Colorinda
pottery, we would classify the amount and size of the quartz
inclusions-with the aid of magnification and the Wentworth
Size Classification- as frequent, fine (.125-.25 mm) to
medium (.25-.50 mm) sized sand; coarse (.5-1.0 mm) grit
inclusions are rare.
In terms of surface treatment Sears only mentions a plain
variety, and for the most part, this appears to be accurate. In
fact, a recurring theme of Colorinda pottery from throughout
the region is that it is almost always undecorated. Russo et al.
(1993:68) reported recovering 3 complicated stamped sherds
"with a Colorinda paste and temper" from Coffee Mound
(8DU7472), while Richter (1993) suggested that a handful of
check stamped and either shell scraped or simple stamped
sherds were among the more than 1200 Colorinda potsherds
from the McCormack site (8DU66). Although the surface of
undecorated Colorinda pots is occasionally well smoothed, the
norm is that surfaces are poorly and haphazardly smoothed,
often revealing sherd tempering (Figure 2). Shell scraping as
a means of surface smoothing is occasionally observed. The
exterior and interior surfaces of sherds are routinely dark gray
to black suggesting that Colorinda vessels were fired in an
oxygen-reduced atmosphere. Specific vessel size and form
data are lacking, but general observations suggest that the
overwhelming majority were medium to large, straight-sided,
open mouth bowls (25-50 cm in diameter) with round bottoms.
Rims are typically straight and rounded, although occasionally
paddle flattened. Sooting is frequently present on vessel
exteriors just below the lip.
It is important to note that in all instances where the
recovery of Colorinda pottery has been reported it always has
been associated with sand-tempered and St. Johns plainwares.
In the past researchers have focused too narrowly on Colorinda
as a single pottery type and have failed to consider it as a




JUNE 2006

VOL. 59(2)


5 Km

Atlantic Ocean

Figure 1. Distribution of Colorinda sites in northeastern Florida.

ceramic complex. Although Colorinda Plain is the featured
type within the complex, it is only part of a diverse assem-
blage. As exemplified by recent excavations at the Cedar
Point site (8DU81) and by Sears's (1957, 1959) earlier
investigations at the McCormack site (8DU66) and Browne
Mound (8DU62), Colorinda Plain, sand-tempered plain, St.
Johns Plain, and St. Johns Check Stamped comprise the
Colorinda ceramic complex. St. Johns Check Stamped,
however, is routinely a minority ware, often making up less
that five percent of the total ceramic assemblage (Ashley 2006,
this issue). Admittedly, on mixed, multicomponent sites it
might only be possible to correlate unequivocally sherd-
tempered Colorinda wares with a Colorinda occupation, since
sand-tempered plainwares and St. Johns series pottery oc-
curred during other periods.

Colorinda Sites

Colorinda Plain pottery, along with sand-tempered and St.
Johns plainwares, prevailed in units dug into middens at three
(8DU59, 8DU62, 8DU66) of the six sites Sears (1957) investi-
gated in 1955 (see Figure 1). His ceramic seriation positioned
Colorinda between an earlier unidentified sand-tempered
complex that included Swift Creek Complicated Stamped and
Weeden Island series wares and a later St. Johns II complex
(Searsl957:30). Based on this analysis and without the aid of
radiocarbon dating, Colorinda was assigned to the "latter part
of the St. Johns Ib" period, ca. A.D. 700-1000. He further
pointed out that Colorinda "is a distinctive type, and appears
to be significant in a specific cultural and temporal context"
(Sears 1957:26).

2006 VOL. 59(2)


Figure 2. Colorinda sherd from the Cedar Point site (8DU81).

Following his initial investigation Sears (1959) returned to
the lower St. Johns region to excavate the Browne Mound
(8DU62), which he had previously tested by digging a 1.5-m
wide trench through the mound (Figure 3). The mound was
1.2 m high and 15.1 m in diameter. His approach this time
was to peel away thin layers of the sand mound in order to
expose large horizontal areas. Excavations revealed a formal-
ized arrangement of burials within an area about 10.6 m in
diameter and enclosed by a 1.5-m wide ring of shell. Accord-
ing to Sears's (1959:6, 9-11) interpretation mound construc-
tion began with the excavation of a central "bathtub-shaped
pit, lined with a thick layer of organic substance." Next as
many as 30 individuals were placed into the pit, with as many
as 13 primary, extended interments laid in "shingle fashion,"
and the pit was covered with a small core mound. A series of
mass bundle burials and a cremation were added to the surface
of the core mound and a few extended interments were laid
along the edge of the core mound. All human remains were
then covered with sand, and the shell ring was laid down along
the mound periphery, partially covering one extended burial.
In conclusion, Sears (1959:10) stated "I see no real alternative
to acceptance of this mound as the product of a single, contin-
uous, short-term ceremony." However, he goes on to suggest
that a short "break in the construction-completion process
appears to have been at the completion of the core mound with

the sand fill over the central mass burial."
In terms of grave goods there were only two items, both
recovered from the central mass grave. These included a
cymbal-shaped copper ornament with a central perforation and
a greenstone celt. Conspicuously absent were ceramic pots,
although Deptford, Swift Creek, Weeden Island, and sand-
tempered plain sherds were scattered throughout mound fill.
Testing of nearby shell middens revealed a sand-tempered
component with a jumbled array of Deptford and Swift Creek
decorated sherds overlain by a Colorinda-dominated shell
midden. Such a sequence corroborated Sears' previously
established ceramic chronology that went from Deptford to
Sand Tempered (including Swift Creek) to Colorinda.
Sears (1959:9) used these midden results in conjunction
with the near absence of Colorinda pottery in mound fill to
"pinpoint mound construction as occurring during the
Colorinda Period." This interpretation, however, is seemingly
lost in the report, as Sears elected to tout the Browne Mound
as a Weeden Island burial mound, presumably in order to
make it more comparable to other Florida mounds alleged to
be contemporaneous. The temporal range of the Colorinda
Period was amended to A.D. 800-1400, which covered a
broader time frame than previously suggested. The Browne
Mound is a unique burial facility in the region, and its charac-
ter differs markedly from the continuous use type mounds of




Figure 3. William Sears in Browne Mound trench, ca. 1955 (ph
courtesy of the Florida Museum of Natural History).

earlier Swift Creek times or later local St. Johns II period
earthworks (Ashley 1998:210-213; 2003:182-208; Thunen and
Ashley 1995).
Despite extensive survey coverage in many areas of the
lower St. Johns region in the ensuing years, Colorinda was
infrequently reported in the archaeological literature. It is
lightly scattered across the Greenfield Peninsula on the south
side of the St. Johns River, immediately west of the
Intracoastal Waterway (Johnson 1996, 1998a, 1998b; Poplin
and Harvey 2000), and peppered on some of the marsh and
barrier islands of northeastern Florida to the north (Hemmings
and Deagan 1973; Russo et al. 1993). A handful of Colorinda
sherds have been recovered from the Grant Mound vicinity
(Hendryxand Smith 2002; Robert Thunen, personal communi-
cation, 2001), but none has been reported for the nearby
Shields Mound area of Mill Cove. Recent and ongoing testing
at the campus of Jacksonville University just south of the
eastward bend of the St. Johns River, north of downtown
Jacksonville, revealed the presence of Colorinda pottery on two

adjacent sites (8DU14683 and 8DU17245) (Greg
Hendryx, personal communications, 2006).
Colorinda pottery may very well be part of the
grog-tempered background noise on some sites that
typically goes unexplained by excavators, but even
in these cases the frequency of grog-tempered
sherds is generally very low.
Of the more than 80 sites investigated within
the Timucuan Ecological and Historic Preserve by
Russo (1992; Russo et al. 1993) in the early 1990s,
8 sites yielded a total of 90 Colorinda sherds. Over
half of this total was recovered from the Coffee
Mound (8DU7472) on Pelotes Island on the north
side of the St. Johns River (see Figure 1). Russo
(1992:116; Russo et al. 1993:67) attempted to
come to grips with the chronological placement of
Colorinda by securing a radiocarbon date on oyster
shell from a midden context associated with
Colorinda pottery at Coffee Mound. The C13/14
corrected date was reported as A.D. 450+60, which
at the time suggested contemporaneity with Dept-
ford, St. Johns I, and Swift Creek within a scram-
bled, local Woodland-period chronology. Appar-
ently frustrated by these results, Russo (1992:109)
tentatively assigned Colorinda to a broad interval,
A.D. 1-800.
A large quantity of Colorinda series pottery
was purportedly recovered from the Turner-McGill
Middens (8DU7495) on Black Hammock Island
(Ellis 1995); a site first sampled by Russo and
colleagues (1993). However, based on the variety
of surface decorations and the dearth of St. Johns
pottery reported for the site, I believe the sherd-
tempered wares actually belong to the San Pedro
)tograph pottery series, which postdates A.D. 1500 (Ashley
and Rolland 1997).
Another reported site that has yielded apprecia-
ble amounts of Colorinda pottery is the Walker
Point site (8NA28) on Amelia Island (see Figure 1). When
excavated in the early 1970s the site was manifested as two
linear shell middens bracketing a conical sand mound
(Hemmings and Deagan 1973:30-53). Although the sand
tumulus had been damaged by past looting activities, an
original height of 3 m and a base diameter of approximately 21
m were estimated. Testing of the earthwork was limited to an
approximate ten percent sample. Intact sections of the mound
revealed a sand mound core with burials and a terminal
episode of construction that involved the deposition of a pink
sand mantle discolored "by disseminated fine hematite parti-
cles" (Hemmings and Deagan 1973:37).
Six individual burials were excavated and all were single
primary interments. A few isolated human bones and teeth
were recovered from undisturbed areas of the mound, includ-
ing the pink sand mantle, whereas an assortment of miscella-
neous human bone was encountered within the disturbed
center of the mound. No whole vessels or grave goods were
discovered, and only 34 potsherds were recovered from mound

2006 VOL. 59(2)



fill, which included mostly sand-tempered and St. Johns
plainwares along with a few St. Johns Check Stamped sherds
and one piece of Swift Creek pottery (Hemmings and Deagan
1973:42). No Colorinda sherds were recovered from mound
Tests dug within both shell middens and other nonmound
loci produced mostly sand-tempered plain, St. Johns Plain, St.
Johns Check Stamped, and Colorinda sherds. Based on the
pottery recovered from the mound Hemmings and Deagan
(1973:53) assigned mound construction and use to the early St.
Johns II Period, ca. A.D. 1000-1200. In fact, Walker Point
Mound is most often interpreted as a St. Johns II construction
(Thunen and Ashley 1995:8). However, the presence of
Colorinda pottery, the dominance of St. Johns Plain over St.
Johns Check Stamped, and the complete absence of
cordmarked wares in a submound shell midden argue against
a St. Johns II period of construction. One would expect more
St. Johns Checked Stamped sherds if the mound dated to the
St. Johns II Period. Additionally, that not one cordmarked
sherd was recovered from the site by Hemmings and Deagan
(1973), nor earlier by Bullen and Griffin (1952:42-43, 54)
during their surface reconnaissance of the Walker Point site,
is informative, since grit-tempered cordmarked wares are
always present on local St. Johns II Period sites and sand
tempered cordmarked pottery dominate on later St. Marys II
sites (Ashley 2003; Ashley and Rolland 2002). Their absence
on the Walker Point site indicates that St. Johns II and St.
Marys II components are not present at the site.
In an attempt to date the mound, oyster from mound fill
was submitted for radiocarbon dating (Ashley 2003). The
sample consisted of both valves of a whole oyster shell found
with other "articulated oyster valves, a hard shell clam valve,
a knobbed whelk, and bony elements of at least one black
drum and one searobin" (Hemmings and Deagan 1973:39).
The lack of midden material in mound fill combined with the
position of the animal remains adjacent to a mound interment
led the authors to conclude that it seemed "probable" that the
collection was an intentional "food offering placed at the feet
of Burial 2 at the time of interment" (Hemmings and Deagan
1973:40). The oyster shell yielded a 2-sigma calibrated
radiocarbon date of A.D. 470 to 740 (1-sigma calibrated
radiocarbon date of A.D. 570-680), a solid Atlantic Coastal
Late Swift Creek date (Ashley and Wallis 2006; Stephenson et
al. 2002:337).
The presence of a Swift Creek sand mound atop a shell
midden that contained St. Johns Plain, St. Johns Check
Stamped, and Colorinda pottery runs contrary to the secure
stratigraphic evidence from the McCormack site (8DU66),
discussed below, which demonstrated the presence of Swift
Creek wares below Colorinda-dominated shell middens (Sears
1957:20-22). Though the mound was only partially sampled,
one would expect more than nine Swift Creek sherds-the
total number recovered from the mound and adjacent middens
by Hemmings and Deagan-from a Swift Creek mound site.
Local Swift Creek mounds are well known for containing large
amounts of complicated stamped pottery (Ashley 1995, 1998;
Ashley and Wallis 2006; Wallis 2004). Also the Walker Point

Mound is at least a meter taller than any other reported Swift
Creek mound, and none of the 14 known Swift Creek burial
mounds has displayed a thick and pervasive hematite impreg-
nated sand cap (Ashley 1998:208-213). Thus, could the "food
offering" simply represent shell midden refuse associated with
an earlier Swift Creek occupation of the site, inadvertently
incorporated into mound fill during construction? More
radiometric dates are needed from Walker Point Mound to
determine its precise chronological position, but it could
conceivably date to the Colorinda Period, although its nature
and structure differ from the Browne Mound (8DU62).
The McCormack' site (8DU66) has produced more
Colorinda pottery than any other site to date. Sears (1957:21)
testing of a shell midden at the site yielded mostly sand-
tempered plain and Colorinda Plain pottery, lesser amounts of
St. Johns Plain, and few St. Johns Check Stamped sherds;
Swift Creek and Deptford sherds were recovered from deeper
deposits. In other areas of the site a St. Johns II component
dominated by chalky check stamped wares surmounts a
Colorinda component. In a collection of more than 6000
sherds surface collected from 8DU66 during church construc-
tion activities in 1991, 2500 St. Johns, more than 2000 sand-
tempered, and almost 1300 Colorinda sherds were recovered
(Richter 1993). Johnson (1988:133-137), who earlier shovel
tested the site as part of a regional survey, encountered similar
proportions although far fewer sherds (305 St. Johns, 200
sand-tempered plain, and 62 Colorinda). Salvage excavations
associated with church construction in 1991 resulted in the
identification of a mass secondary burial and a series of
extended face-down burials (Ashley 2003:186-187; Thunen
and Ashley 1995:9). A report of the excavations has not yet
been made available so the cultural affiliation of the burials is
uncertain. However, prone and multiple secondary burials are
known for St. Simons Island, Georgia and dated to the local
Kelvin Phase, a time contemporaneous with Colorinda (Cook
1977:20-22; 1979:73-76).
The most recently reported investigation ofa Colorinda site
took place at the Cedar Point site (8DU81), which was
excavated by a University of North Florida (UNF) archaeologi-
cal field school in the summer of 2003. The site's Colorinda
component consisted of a series of small shell midden heaps.
During fieldwork several middens were sampled and one was
intensively tested. The results of this excavation are detailed
elsewhere (Ashley 2006, this issue), but suffice it to say the
project generated a variety of data in the form of subsistence
and seasonality information, ceramic analysis, and radiometric
assays that have contributed to a better understanding of
Colorinda life along the lower St. Johns River.

Dating Colorinda

Over the past decade significant strides have been made in
refining the late precolumbian chronology of northeastern
Florida. However, certain gaps in the sequence persisted,
including the precise temporal placement of Colorinda
occupations (Ashley 1998:198-201; Russo 1992:116). A
significant contribution of the 2003 UNF Field School at the




Cedar Point site was the recovery of Colorinda pottery from
secure contexts, which provided the opportunity to date
Colorinda occupations via radiometric dating (Ashley 2006,
this issue). Oyster shells taken from a Colorinda shell midden
and soot from a Colorinda Plain sherd were submitted for
radiocarbon and AMS dating, respectively. The 2-sigma
calibrated date ranges for the two samples area A.D. 720-960
and 740-1020, whereas the 1-sigma calibrated date ranges are
A.D. 780-890 and A.D. 800-960, respectively, suggesting that
the Colorinda shell midden there was deposited over brief
period of time, perhaps even a single occupation episode.
Returning to Russo's (1992:110, 116) Colorinda date of
A.D. 450+60 from the Coffee Mound (8DU7472), it is a
corrected date, not a calibrated date. In order to render the
date consistent with two from the Cedar Point site, the assay
was recently calibrated by Beta Analytic, Inc. Its 2- sigma
(A.D. 760-1020) and 1-sigma (A.D. 820-980) date ranges are
consistent with two from the Cedar Point site. While 3
radiocarbon dates from two sites obviously form a small
sample, the tight clustering dovetails nicely with a string of
radiometric dates acquired in the past decade for earlier Late
Swift Creek (A.D. 600-850) and later St. Johns II (A.D. 900-
1250) occupations (Ashley 2005:155; Ashley and Wallis
2006:7). It now appears that Colorinda was a short-lived
manifestation that dated to the ninth century A.D.

Where Did Colorinda Come From and Where Did It Go?

These are difficult questions to answer with the data at
hand, but, by viewing Colorinda in its historical sequence, I
believe some preliminary statements are warranted. In terms
of chronology, Colorinda is sandwiched between Late Swift
Creek and St. Johns II occupations in northeastern Florida.
Most all sites that have yielded Colorinda pottery also have
produced Swift Creek wares, but, in almost all cases in which
stratigraphic evidence is available, Colorinda contexts are
positioned above Swift Creek contexts. It has been observed at
several Late Swift Creek sites that some complicated wares are
tempered with fine grog (Ashley and Wallis 2006; Wallis
2004:278). At the Cedar Point site (8DU81), two shell
impressed or dentate stamped sherds from a single vessel,
similar to those found on other coastal Swift Creek sites, had
grog-tempering and the grog appeared to be finely crushed St.
Johns sherds (Ashley 2006, this issue). Along the Georgia
coast some Late Woodland Kelvin (referred to as waning Swift
Creek in northeastern Florida) pottery contains fine grog
tempering and these wares are thought to date ca. A.D. 800-
1000 (Cook 1977, 1979). Another distinctive characteristic of
waning Swift Creek or Kelvin pottery along the Atlantic coast
is that complicated stamping becomes simplified and sloppy in
appearance (Ashley and Wallis 2006; Cook 1977, 1979;
Hendryx 2004).
Taken together, this information suggests the use of (or
experimentation with) grog tempering by some Late Swift
Creek populations along the Atlantic seaboard. By the ninth
century, grog-tempering dominated within Wilmington and St.
Catherines pottery assemblages along the central and northern

Georgia coasts and within Colorinda assemblages in northeast-
ern Florida, at the same time complicated stamping ceased to
be produced within these regions. Sears (1957:32-33;
1959:17) looked to the sherd-tempering Wilmington ceramic
complex as possible source of "influence" for the emergence of
In all likelihood Late Swift Creek developed into Colorinda
in northeastern Florida, while Late Swift Creek (Kelvin) may
have lingered on a little longer in coastal Georgia before
giving way to grog-tempered Wilmington series pottery. As
stated, available radiometric dates hint at the potential
contemporaneity of Colorinda occupations at the mouth of the
St. Johns River and waning Late Swift Creek activities
immediately to the north during the first half of the ninth

century A.D. In fact, statistically equivalent radiometric dates
from waning Swift Creek contexts at the Honey Dripper site on
Amelia Island in northeastern Florida (Hendryx 2004) and
Colorinda contexts at the Cedar Point site suggest that pottery
change was not uniform and that the shift from Late Swift
Creek to Colorinda may not have been simultaneous among all
settlements or groups.
Colorinda pottery assemblages, however, are not made up
exclusively of sherd-tempered wares. From a temper perspec-
tive, Colorinda assemblages are diverse and consist of sand-
tempered plainwares, sherd-tempered plainwares (Colorinda),
and small amounts of St. Johns pottery; the latter includes
mostly plainwares but minor amounts of check stamping, as
reported herein (also see Ashley 2006, this issue; Russo 1992;
Sears 1957,1959). The ceramic assemblage is unique with the
apparent blending of the long-standing, local tradition of sand
tempering, the sherd tempering of the Georgia coast (e.g.,
Wilmington), and the spiculate tempering (St. Johns) of the St.
Johns River valley to the south.
Compared to earlier Late Swift Creek and subsequent St.
Johns II times, Colorinda is represented archaeologically by a
small number of sites. One probable Colorinda burial mound
(i.e., Browne Mound) and another possible one (i.e., Walker
Point Mound) each appears to have been the scene of either a
single interment episode or a limited number of separate burial
events over a brief time (Hemmings and Deagan 1973:35;
Sears 1959:10-11). This contrasts with the nature of Late
Swift Creek mounds, which were built up incrementally and
repeatedly used for long periods of time (Ashley 1995, 1998).
Also, the presence offace-down interments and single episode,
multiple primary burials in non-mound pits at the McCormack
site and the collective and simultaneous interment of several
extended individuals in a pit at the base of the Browne Mound
further distinguishes Colorinda mortuary practices from that
of Late Swift Creek (Sears 1959:4-7). However, similar types
of burials have been reported on Kelvin sites of the same time
period on St. Simon's Island, Georgia (Cook 1979:73; Holder
1938:8; Martinez 1975:51-57).
As a whole, these data allude to a short period of strife and
uncertainty in northeastern Florida during the ninth century,
a time when some local groups might have moved out of the
area, and those that remained might have undergone a crisis
of community identity. If we broaden our scale of analysis for

2006 VOL. 59(2)


this time period, we see that much of the Southeast was
apparently undergoing a short lull in the long-distance
trafficking of exotic materials during the Late Woodland
period, ca. A.D. 700-900 (Nassaney and Cobb 1991; Cobb and
Nassaney 1995). In northeastern Florida this may be reflected
in the cessation of decorating pots with Swift Creek Compli-
cated Stamped designs and the emergence of Colorinda
pottery; Swift Creek design studies soundly demonstrate
contacts between northeastern Florida groups and those to the
north along the Georgia coast (Ashley 1995, 1998; Ashley and
Wallis 2006). Colorinda populations appear to have been
smaller and perhaps more insular than their Swift Creek
predecessors. However, we should not view them as isolation-
ists since the emergence of certain mortuary ideas (e.g., prone
burials, multiple primary interments) and ceramic attributes
(e.g., grog tempering) may have been the result of extralocal
interaction, although separate cultural identities were main-
tained along the coast of Georgia and Florida.
It is against this backdrop that St. Johns II sites quickly
sprang up across the northeastern Florida landscape sometime
during the tenth century A.D. (Ashley 2003). These new
populations soon were actively involved in long-distance
exchange networks. One could argue for a gradual in situ
development of local St. Johns II out of Late Swift Creek, with
Colorinda reflecting a transitional stage. However, the
rapidity and scale at which St. Johns II populations took over
northeastern Florida in concert with the locational disjuncture
of Colorinda and St. Johns II sites speaks of a settlement shift
along the river; a shift heralded by an influx of St. Johns II
populations from the south to northeastern Florida (Ashley
2003). Additionally, Colorinda wares are conspicuously
absent in early St. Johns II assemblages. Immigrant St. Johns
populations may have absorbed local Colorinda communities,
rendering northeastern Florida St. Johns II a mix of local and
nonlocal populations. Establishing Colorinda as terminal Late
Woodland is a critical step in our understanding of the
emergence and cultural development of the "Mississippian
Period" in northeastern Florida.


Presently, available evidence suggests that the Colorinda
period of northeastern Florida history was brief and occupa-
tions were few and localized. Although it is tempting to argue
for a significant population reduction in northeastern Florida
at this time compared to that of earlier Late Swift Creek times,
we must bear in mind that Colorinda covered a shorter
duration than Late Swift Creek. More investigations are
needed at Colorinda sites to understand in more detail the
ceramic transitions between earlier Late Swift Creek and
Colorinda and between Colorinda and later St. Johns II.
Additionally, relationships between the Colorinda archaeologi-
cal culture and Kelvin and Wilmington along the coast of
Georgia need to be further explored. I recommend a detailed
technological study of Colorinda pottery to document and
quantify its paste constituents. An intriguing question that
needs to be addressed is-are the same clays being used to

manufacture the variously tempered wares (i.e., Colorinda,
sand tempered, St. Johns) within the Colorinda ceramic
complex? Although the temporal position of Colorinda within
the history of northeastern Florida is beginning to come into
focus, many other questions remain unanswered and await
further research.


SI consider native precolumbian history to be part of the overall
history of northeastern Florida. The contribution of the "Colorinda"
natives to the history of northeastern Florida is just as compelling as
those events that surrounded the French Huguenots at Fort Caroline
in the late 1500s, early Americans at Cowford in the 1800s, African
Americans at American Beach in the 1900s, and even the arrival of
the NFL's Jaguars to downtown Jacksonville in the 1990s.

2 Throughout this paper northeastern Florida refers exclusively to
northern St. Johns, Duval, and Nassau counties.

3 Sears named 8DU66 the McCormack site, although the landowner's
surname was actually McCormick. Both versions appear in the
archaeological literature, which has created some confusion. I have
chosen to use McCormack since this is the name given and used by


I appreciate the assistance ofJerry Milanich, Vicki Rolland, Ryan
Wheeler, Ann Cordell, Greg Hendryx, Catherine Runyan, and David
Mynatt with various aspects of this paper. Also thanks to Elise
LeCompte and the Florida Museum of Natural History for allowing
use of the photograph in Figure 3.

References Cited

Ashley, Keith H.
1995 The Dent Mound: A Coastal Woodland Period Burial
Mound Near the Mouth of the St. Johns River. The
Florida Anthropologist 48:13-35.

1998 Swift Creek Traits in Northeastern Florida: Ceramics,
Mounds, and Middens. InA WorldEngraved: Archaeology
of the Swift Creek Culture, edited by M. Williams and D.
T. Elliott, pp. 197-221. University of Alabama Press,

2003 Interaction, Population Movement, and Political Economy:
The Changing Social Landscape of Northeastern Florida
(A.D. 900-1500). Ph.D. dissertation, Department of
Anthropology, University of Florida, Gainesville.

2004 Archaeological Testing at the Cedar Point Site (8DU81):
Results of the 2003 UNF-NPS Summer Field School.
Report on file, Florida Master Site File, Division of
Historical Resources, Tallahassee.

2005 Introducing Shields Mound (8DU12) and the Mill Cove
Complex. The Florida Anthropologist 58:151-173.

2006 Archaeological Testing of Colorinda Shell Middens at the
Cedar Point Site (8DU81). The Florida Anthropologist
59(2):101-122 (this issue).




Ashley, Keith H., and Vicki L. Rolland
1997 Grog-Tempered Pottery in the Mocama Province. The
Florida Anthropologist 50:51-66.

2002 St. Marys Cordmarked Pottery (Formerly Savannah Fine
Cord Marked of Northeastern Florida and Southeastern
Georgia): A Type Description. The FloridaAnthropologist

Ashley, Keith H., and Neill Wallis
2006 Northeastern Florida Swift Creek: Overview and Future
Research Directions. The Florida Anthropologist 59:5-18.

Borremans, Nina Thanz, and Craig D. Shaak
1986 A Preliminary Report on Investigations of Sponge Spicules
in Florida "Chalky" Pottery. Ceramic Notes 3:125-132.
Occasional Publications of the Ceramic Technology
Laboratory, Florida State Museum, Gainesville.

Bullen, Ripley P., and John W. Griffin
1952 An Archaeological Survey of Amelia Island, Florida. The
Florida Anthropologist 5:37-64.

Cobb, Charles R., and Michael S. Nassaney
1995 Interaction and Integration in the Late Woodland Southeast.
InNativeAmerican Interactions: MultiscalarAnalysis and
Interpretations in the Eastern Woodlands, edited by M. S.
Nassaney and K. E. Sassaman, pp. 205-226. University of
Tennessee Press, Knoxville.

Cook, Fred C.
1977 The Lower Georgia Coast as a Cultural Sub-Region. Early
Georgia 5:15-36.

1979 Kelvin: A Late Woodland Phase on the Southern Georgia
Coast. Early Georgia 7(2):65-86.

Cordell, Ann S., and Steven H. Koski
2003 Analysis of a Spiculate Clay from Lake Monroe, Volusia
County, Florida. The Florida Anthropologist 56:113-125.

Ellis, Gary D.
1995 Phase II Investigations at 8DU7495, West Central Black
Hammock Island, Duval County, Florida. Report on file,
Florida Master Site File, Division of Historical Resources,

Hemmings, Thomas, and Kathleen Deagan
1973 Excavations on Amelia Island in Northeast Florida.
Contributions oftheFloridaStateMuseum 18, Gainesville.

Hendryx, Gregory S.
2004 The Honey Dripper Site (8NA910): A Late Swift Creek
Encampment in Northeastern Florida. The Florida
Anthropologist 57:299-310.

Hendryx, Gregory S., and Greg C. Smith
2002 An Intensive Cultural Resource Assessment Survey of
Windswept Point and Site Testing at 8DU5599, Duval
County, Florida. Report on file, Florida Master Site File,
Division of Historical Resources, Tallahassee.

Holder, Preston
1938 Excavations of St. Simons Island and Vicinity, Winter of
1936-1937. Proceedings of the Society for Georgia
Archaeology 1:8-9.

Johnson, Robert E.
1988 An Archeological Reconnaissance Survey of the St. Johns
Bluff Area of Duval County, Florida. Report on file,
Florida Master Site File, Division of Historical Resources,

1996 An Intensive Archeological Survey of Florida Inland
Navigation District Tract DU7, Duval County, Florida.
Report on file, Florida Master Site File, Division of
Historical Resources, Tallahassee.

1998a Phase II Archeological Investigations of Sites 8DU5544
and 8DU5545, Queen's Harbour Yacht and Country Club,
Duval County, Florida. Report on file, Florida Master Site
File, Division of Historical Resources, Tallahassee.

1998b A Phase II Archeological Investigation of Florida Inland
Navigation District Tract DU7, Greenfield Peninsula,
Duval County, Florida. Report on file, Florida Master Site
File, Division of Historical Resources, Tallahassee.

Martinez, Carlos A.
1975 Cultural Sequence on the Central Georgia Coast. Unpub-
lished Master's thesis, Department of Anthropology,
University of Florida, Gainesville.

Nassaney, Michael S., and Charles R. Cobb (editors)
1991 Stability, Transformation, and Variation: The Late Wood-
land Southeast. Plenum Press, New York.

Poplin, Eric C., and Bruce G. Harvey
2000 National Register of Historic Places Evaluation of 8DU78,
US Naval Station Mayport, Duval County, Florida. Report
on file, Florida Master Site File, Division of Historical
Resources, Tallahassee.

Richter, Bob
1993 Surface Collections from 8DU66, The McCormick site.
Ms. on file with author.

Rolland, Vicki L., and Paulette Bond
2003 The Search for Spiculate Clays near Aboriginal Sites in the
Lower St. Johns River Region, Florida. The Florida
Anthropologist 56:91-112.

Russo, Michael
1992 Chronologies and Cultures of the St. Marys Region of
Northeast Florida and Southeast Georgia. The Florida
Anthropologist 45:107-126.

Russo, Michael, Ann S. Cordell, and Donna L. Ruhl
1993 The Timucuan Ecological and Historical Preserve Phase
IIIFinalReport. National Park Service, Southeast Archeo-
logical Center, National Park Service, Tallahassee.

Sears, William H.
1957 Excavations on Lower St. Johns River, Florida. Contribu-
tions of the Florida State Museum 2, Gainesville.


2006 VOL. 59(2)


1959 Two Weeden Island Period Burial Mounds, Florida.
Contributions of the Florida State Museum 5, Gainesville.

Stephenson, Keith, Judith A. Bense, and Frankie Snow
2002 Aspects of Deptford and Swift Creek on the South Atlantic
and Gulf Coastal Plains. In The Woodland Southeast,
edited by D. G. Anderson and R. C. Mainfort, Jr., pp. 318-
351. University of Alabama Press, Tuscaloosa.

Thunen, Robert L., and Keith H. Ashley
1995 Mortuary Behavior along the Lower St. Johns: An Over-
view. The Florida Anthropologist 48:3-12.

Wallis, Neill
2004 Perpetuating Tradition on he Lower St. Johns: Pottery
Technology andFunction at the Mayport Mound (8DU96).
The Florida Anthropologist 57:271-298.

A video on Florida's native peoples

"Shadows and Reflections:
Florida's Lost People"
Produced by the Florida
i Society
Funded by the
Florida Department
7r \ of State

Produced and Directed by Chaos Productions
Executive Producer: Brent Weisman
Written by Marshall Riggan
Artwork by Theodore Morris

C 1998 Florida Anthropological Society and the Florida Department of State

To obtain copies send $23.62 [$18.81 plus $1.31 (sales tax) and $3.50 (S&H) to:
Terry Simpson, 9907 High Meadow Ave., Thonotosassa, FL 33592

SITE (8DU81)


Department ofLiberal Arts, Savannah College ofArt and Design, Savannah, GA 31402-3146

Colorinda is probably one of the least discussed legitimate
pottery types and archaeological cultures in Florida. Despite
the fact that William Sears (1957, 1959) originally defined
Colorinda in the late 1950s on the basis of his seminal
research along the lower St. Johns River, it was not mentioned
in any of the subsequent hallmark overviews of Florida
archaeology (e.g., Milanich 1994; Milanich and Fairbanks
1980). And it is at best awkwardly noted in culture history
sections of some cultural resource management reports.
Among local archaeologists it has been jokingly referred to as
northeastern Florida's dirty little secret. While Sears (1957,
1959) considered Colorinda to be a significant ceramic
complex in northeastern Florida, he also noted that it was
spatially confined to the lower St. Johns region during the
period A.D. 700-1000, which probably explains its exclusion
from many cultural overviews. Elsewhere I have recently
attempted to synthesize variety of archaeological information
on Colorinda to establish its place within the history of
northeastern Florida (Ashley 2006, this issue).
This paper sheds new light on the little-known Late
Woodland Colorinda culture by reporting the results of a
recent University of North Florida (UNF) archaeological field
school at the Cedar Point site (8DU81), near the mouth of the
St. Johns River in northeastern Florida (Ashley 2004).
Because the site lies within the spatial boundaries of the
Timucuan Ecological and Historic Preserve, the 2003 summer
field school was a joint and cooperative endeavor between the
National Park Service (NPS) and UNF. The results of the
project have added a variety of data augmenting our under-
standing of the Colorinda period. Two radiocarbon dates from
the site help anchor Colorinda to the ninth century A.D.;
ceramic evidence improves our knowledge of the Colorinda
pottery series; and the faunal analysis provides the first
reported information on Colorinda subsistence practices.

Site Location and Environmental Setting

The Cedar Point site (8DU81) is located at the southern
end of Black Hammock Island, approximately 6 km (4 miles)
west of the Atlantic Ocean and 18 km (11 miles) northeast of
downtown Jacksonville (Figure 1). It is situated east of the
Cedar Point West site (8DU63) and south of the Cedar Point
North site (8DU64). Together the three comprise a wide-
spread series of adjacent and overlapping shell midden
deposits along the southern tip of the island (a location known
locally as Cedar Point). Although only limited subsurface

testing has been conducted in the past, previous surveys have
demonstrated variable density middens, some of which are
densely packed and mounded in the form of circular to oval
heaps and linear ridges (Jones 1985; Nidy 1980:39-40, 43;
Russo et al. 1993:39-42). Artifacts from the sites indicate
multiple and intermittent occupations that began in the Late
Archaic period, persisted throughout precolumbian times, and
continued into the colonial period and modern times. How-
ever, none of the previous investigators reported the presence
of Colorinda pottery on any of the three Cedar Point sites.
Cedar Point is situated within the lower reaches of the St.
Johns River estuary. The southern and eastern boundaries of
the site are formed by tidal marshes, except in the areas where
Horseshoe Creek, a tributary of the Intracoastal Waterway
(Sisters Creek), breaks through to abut the island edge.
Expansive salt marshes of grasses, rushes, and sedges laced
with a network of tidal creeks lie along the perimeter of Black
Hammock Island. The grassy salt marsh and shallow tidal
creeks serve as nurseries to numerous species of fish and
shellfish and as habitat for an array of waterfowl and other
fauna (Johnson et al. 1974:72-85). In fact, the Florida Game
and Freshwater Fish Commission recorded more than 100
species of estuarine and marine fish in the immediate vicinity
of Cedar Point in 1983. At low tide, extensive mud flats are
exposed along the coastline and lower river margins, facilitat-
ing the capture or collection of a variety of vertebrate and
invertebrate fauna.
The site environment consists of mature growth maritime
hammock on well-drained upland soils, about 5 ft above the
adjacent tidal marshes (Figure 2). In addition to providing
food and medicine, the forest offered native groups variety of
resources such as wood and vines for the construction of
houses, canoes, and other implements; wood and brush for
fuel; and various plant fibers such as Spanish moss, cedar
bark, and palmetto fronds for making twine, cordage, rope,
nets, and basketry. Combined the uplands and marshes would
have provided natives with a wealth of subsistence and other

Field Methodology

The 2003 Summer Field School represents an initial step
in what is hoped to be a long-term archaeological study of the
Cedar Point vicinity (Ashley 2004). In addition to instructing
college students in the basics of archaeological data recovery
and laboratory analysis, the field school was designed to


VOL. 59(2)


JUNE 2006


Atlantic Ocean

Figure 1. Location of Colorinda sites in northeastern Florida.

provide NPS with baseline information on the Cedar Point site
for management purposes. The very limited testing previously
conducted at 8DU81 combined with the exploratory nature of
the field school resulted in the formulation of a series of basic
research questions: who inhabited the site, when was it
occupied, and what was the nature of past occupations.
A testing plan was developed that called for systematic
shovel test sampling followed by unit excavations in areas of
greatest artifact and faunal bone frequency. Fifty-centimeter
(cm) square shovel tests were dug at 25-meter (m) intervals
along the southern periphery of the site and judgmentally
along the unpaved portion of Cedar Point Road (Figure 3).
The latter tests were placed between 7.5-m and 34-m apart in
order to target mounded shell middens. These exploratory
tests were dug systematically in arbitrary 20-cm levels to a
terminal depth of 1-m below surface when possible. Units 1,

2, 5, and 7 were each 1 X 2-m units placed at various locations
across the central part of the site based on shovel test results.
Units 3, 4, 6, 8, and 9 also were 1 X 2-m units, but were
combined with Units 10 and 11 (1-m squares) to form Excava-
tion Block A (Figure 4). Each unit was excavated in arbitrary
10-cm (or less) levels within observable zones. In total, 26
shovel tests, 9 1 X 2-m units, 2 1 X 1-m units, and 2 50-cm
square unit extensions were excavated.
Deposits were dry screened through 6.35 mm (1/4")
hardware mesh in the field; vertical control was maintained by
measuring down from a project-established benchmark with
the aid of an auto-level and stadia rod. A rough estimate of
the total volume of shell from each level was calculated in
liters using plastic measuring buckets, and the variety and
approximate ratio of observed shellfish species was noted prior
to discarding. Within each level, artifacts, vertebrate fauna,

2006 VOL. 59(2)



Figure 2. Cedar Point site, view to the east

and charcoal were collected and placed in separate bags.
Three 50-cm square column samples (CS#1-3) were
removed from separate shell midden loci and excavated in
levels that each corresponded to unit levels. Soil samples also
were taken from selected features and other midden contexts
throughout the site. In the laboratory these samples were
water screened through a nested series of 1/2", 1/4", and 1/16"
mesh. Large and small fractions of each sample were bagged
separately and labeled for future analysis.

Results of Testing

Shovel testing was confined to four loci- Marsh Area,
Eastern Peninsula, Southern Peninsula, and Cedar Point Road
Area-with the latter containing the most substantial refuse
deposits in the form of mounded and non-mounded shell
middens (see Figure 3). As a result, 5 separate middens were
tested along old Cedar Point Road. The northernmost unit
(Unit 5) was dug to sample a low-density Protohistoric San
Marcos/Altamaha culture midden of loose shell, while Units
1, 2, and 7 were excavated in separate areas of mounded shell
midden (see Figure 4); the latter three yielded mostly
Colorinda Period pottery along with other minority types.
Finally, Block A, consisting of 1 X 2-m units (Unit 3,4,6,8,
and 9) and 1-m squares (Unit 10 and 11), was centered on the
summit of a rich Colorinda Period shell midden, with underly-
ing Late Swift Creek Period deposits.

Unit 1

Unit 1 was positioned atop a mounded shell midden that
measured about 6 m in diameter, although probing revealed
that non-mounded (yet dense) accumulations of discarded shell
extended beyond these measurements, particularly to the west.
The shell midden (Zones I and II) was densely packed, with a
maximum depth of 52 cm below surface. In total, approxi-
mately 949 liters of shell were removed from shell midden
levels, with the highest volume (245 liters) taken from Level
2 (10-20 cm below surface). In addition, a light scatter of shell
(21 liters) was recovered from Zone III immediately beneath
the shell midden. Oyster was by far the dominant shellfish
species observed in all levels, with small amounts of quahog
clam and stout tagelus also noted.
Vertebrate fauna was recovered from all excavation levels,
with a total of 88.9 g recovered from the shell midden and 8.9
g taken from contexts below the shell midden. The highest
quantity of bone was 25.0 g from Level 4. Fish was the
dominant vertebrate animal class represented in the shell
midden, and observed taxa included redfish, catfish, mullet,
seatrout, flounder, gobie, and unidentified Sparidae/
Sciaenidae. Small amounts of mammal (including rabbit and
opossum), bird, turtle, and crab (claws) also were recovered.
Of the 64 pottery fragments recovered fromUnit 1, 36 were
larger than 2 cm (Table 1). Sherds from the shell midden
included Colorinda, St. Johns, sand-tempered plain, grog-
tempered plain, grit-tempered check stamped, and Swift Creek




Cedar Point
Road Area

Figure 3. Sampling areas and shovel test locations.

Complicated Stamped. The sub-shell midden sherds consisted
of 4 sand tempered plain and 1 Deptford Check Stamped,
while a single Colorinda pottery fragment was found during
wall cleanup. Based on these results, the shell midden dates
to the Colorinda Period, while sub-shell midden contexts
demonstrated the presence of earlier Swift Creek and Deptford

Unit 2

Unit 2 was located 2.5 m east of old Cedar Point Road and
a meter west of ST#14, atop a distinct oval-shaped shell heap

2006 VoL. 59(2)


® Shovel Test

r] Saltmarsh

that measured 6.5 by 5 m. The unit was placed slightly
southwest of the center of the shell heap. The average depth of
the midden was approximately 60 cm, but it dipped to a
maximum depth of 76 cm. Approximately 989 liters of shell
were recovered from the shell midden (Levels 1-7), while only
3 liters were found below the base of the shell midden. Within
the midden there were two distinct shell volume spikes: the
first in Level 3 (220 liters) and the second in Level 6 (210
liters), These two levels also contained different pottery types
and faunal assemblages. A dramatic decline in shell density
was noted in Level 5, which yielded only 90 liters of shell. A
small amount of burnt shell and bone was noted in Levels 3


Table 1. Pottery totals by unit.

Unit 1 Unit 2 Unit Unit 7A Total
sand tempered plain 8 17 31 3 59
sand tempered check stamped 3 -- 3
sand tempered stamped 2 2
sand tempered dentate 1 1
sand tempered eroded 1 6 7
Colorinda Plain (SSHT) 9 11 3 2 25
Colorinda Plain (SHT) 2 12 4 18
St. Johns Plain 9 3 9 -21
St. Johns Check Stamped 1 2 5 8
St. Johns eroded 1 1 -2
sandy St. Johns 3 -3
Swift Creek Complicated Stamped 1 2 9 -12
Deptford Check Stamped 1 11 -12
grog tempered plain 3 -3
grit tempered check stamped 1 -- 1
Subtotal 36 62 74 5 177

sherd < 2cm 28 28 87 10 153

Grand Total 64 90 161 15 330
SSHT- spiculate sherd tempered
SHT sherd tempered

through 5.
More than a kilogram (kg) of animal bone was recovered
from the shell midden. The largest amount was taken from
Level 3 (335 grams); next was 186 g from Level 1; and then
158 g from Level 6. No bone was recovered from sub-shell
midden contexts. Faunal remains from Levels 1- 4 were
similar and observed taxa include catfish, flounder, mullet, sea
trout, gobie, blue crab, unidentified small and large bird,
unidentified small and large mammal, rabbit, canid (possibly
dog or fox), and turtle. Fish were mostly small sized, but some
large ones were present. The dominant fish species was
catfish. The faunal assemblage from Level 6 contrasted with
that of the previous levels by containing more larger-sized fish
and few catfish remains. Observed species from Level 6
included deer, redfish, grouper, mullet, sea trout, flounder,
gobie, gar, unidentified bird, and unidentified turtle.
Ninety-three pottery fragments were recovered, with the
majority being larger than 2 cm in size (Table 1). Potsherds
from the shell midden included Colorinda Plain, sand tem-
pered, St. Johns (including a hone), Swift Creek Complicated
Stamped, and Deptford Check Stamped. One Deptford Check
Stamped was recovered from the subsoil beneath the shell
midden, and 3 St. Johns Check Stamped were found during
wall profile cleaning. Zones I and II comprise a Colorinda
shell midden, while Zone III were associated with an earlier
Deptford shell midden. Although a few Deptford sherds were
recovered from the Level 1, their presence presumably
represents the inadvertent incorporation of earlier refuse into

later middens by unknown causes (possibly site reoccupation

Unit 7 and 7A

Unit 7 was located about 5 m northeast of old Cedar Point
Road, on top of a mounded shell midden. ST#22, located 30
cm to the west, yielded a large quantity ofbone, prompting the
excavation of Unit 7. Following the completion of Unit 7, a
50-cm square column sample (CS#3) was taken from the
eastern half of the unit's south wall. A possible feature
(Feature 2) was exposed at a depth of 1.90 mbd during column
sample removal. To expose the feature in more detail, a 50-cm
square extension (Unit 7A) was dug west of and contiguous
with CS#3. Once exposed and cross-sectioned, it was deter-
mined that Feature 2 was a taproot.
The densely packed shell midden averaged about 40 cm in
thickness, although it displayed a maximum thickness of 46
cm in the southern part of the unit. In Unit 7, approximately
676 liters of shell were recovered form the shell midden
(Zones I and II); 33 liters was recovered from Zone III; and
Zone IV yielded less than one liter of shell. An additional 92
liters were recovered from the shell midden in Unit 7a, and 7
liters were removed from Zone III of this unit. Oyster was the
dominant molluscan species in all levels, with trace amounts
of quahog clam, stout tagelus, and Atlantic ribbed mussel.
Animal bone was well preserved and abundant within the
shell midden. More than 980 g and 60.8 g were recovered




Cedar Point

Unit 5

SUnit 2

* Unit 7

8 Unit I




Figure 4. Excavation unit and block locations.

from Units 7 and 7A, respectively. Of the more than 1 kg of
combined bone, 516 g were taken from Level 3. There was a
marked drop off in the quantity of recovered bone in Zones III
and IV. Fish was by far the dominant class noted in the faunal
assemblage, with catfish representing the most commonly
observed species. Other fish observed in the assemblage
included mullet, jack, flounder, gar, and unidentified
Sparidae/Sciaenidae. Small- and large-sized fish appear to
have been evenly represented. Mammalian remains consisted
of rabbit, canid, and other unidentified large and small species.
Finally, turtle and crab remains also were recovered.
In all, 75 sherds were recovered from Unit 7 and 5 from

Unit 7A (Table 1). Shell midden sherds from the two units
included sand-tempered, St. Johns, Colorinda, sand St. Johns,
and Swift Creek Complicated Stamped. Of the Swift Creek
pottery, a scalloped rim sherd combined with wide spacing
between complicated stamped lands suggests that some are
Early Swift Creek (A.D. 300-600) as opposed to Lake Swift
Creek (A.D. 600-850); the latter was more common in Block
A. The midden thus appears to represent a mixed Colorinda
and Early Swift Creek refuse deposit. Four pottery fragments,
3 sand-tempered plain and 1 Swift Creek Complicated
Stamped, were recovered from sub-shell midden contexts.

2006 VOL. 59(2)





Figure 5. Units comprising Block A.

BlockA (Units 3, 4, 6, 8, 9, 10, and 11)

Block A began as a 1 X 4-m trench consisting of two
contiguous 1 X 2-m units (Units 3-4). It was soon expanded
to include a total of five 1 X 2-m units (Units 3, 4, 6, 8, 9), two
1-m squares (Units 10, 11), and a 50-cm square (Unit 8A).
The block covered most of the eastern half of a kidney-shaped
shell heap that measured approximately 8.5 by 4.5 m (Figure
5). Probing with a metal rod revealed that non-mounded, yet
consolidated, shell midden extended slightly beyond these
dimensions to the north and east. ST#16 had previously been
excavated into the shell heap 50 cm northwest of Unit 3, while
ST#17 had been dug in an area of minimal shell midden, 10
m to the south of ST#16. Two 50-cm square column samples
(CS#1 and CS#2) were collected from Block A. Block A was
the scene of most intensive and extensive excavation during
the 2003 field school (Figure 6).
Figure 7 depicts the grid-west trench wall of Units 4 and 3
(from south to north), and Figure 8 reveals the grid-north
trench wall of Units 6, 10, and 11 (from west to east). The
distinction between some strata was slight and was not fully
comprehended until after excavation, when each wall profile
was examined intensively. The following provides a general
summary of the stratigraphy of Block A, based on the long
trench profiles.

Zone I was a layer of black (10 YR 2/1) sand containing
duff, root mat, and crushed shell. It was encountered in each
unit, with a depth that ranged from 3 to 12 cm below surface.
Whole shells were more prevalent in the lower part of Zone I
and there was a gradual transition from the bottom of Zone I
into Zone II.
Zone II was a densely packed shell midden within a very
dark brown (10 YR 2/2) to black (10 YR 2/1) sand matrix. It
demonstrated a high ratio of whole and large shell fragments
to broken and crushed shell. This shell midden was encoun-
tered in all excavation units, with tapered edges delineated in
Unit 6 and Unit 4; these corresponded to the western (grid)
and southern (grid) margins of Zone II, respectively. The
main part of this Colorinda shell midden was encountered in
Levels 1 through 5. Oyster shell, pottery, and vertebrate
animal bone were the primary constituents; smaller quantities
of charcoal and calcined shell fragments were scattered
randomly throughout the zone.
Zone IIa was an organically-rich, black (10 YR 2/1) sand
lens very similar to Zone II, but with far less shell. In fact,
shells were loosely consolidated compared to Zone II with a
much higher soil to shell ratio. Zone IIa was encountered in
portions of all units, and sandwiched between Zone II (above)
and Feature 5 (below). It averaged about 10 cm in thickness.
Artifacts recovered from Zone IIa indicated a Colorinda refuse




Figure 6. Block A excavation.

Zone EI was another organically-rich, black (10 YR 2/1)
shell midden restricted to Unit 6 (Levels 9B and 10B). The
tapered edge of Zone II overlaid the tapered end of Zone III, as
revealed in Figure 8. Zone III contrasted with Zone II by
being slightly darker and containing much less shell (i.e.,
lower shell to soil ratio than Zone II). Although some mixing
was evident, Zone II appears to represent an earlier Deptford
Period refuse deposit, as evidenced by the preponderance of
Deptford Simple Stamped from this context.
Zone IV was a dark grayish brown (10 YR 3/2) shell
midden encountered in the grid-southern part of Unit 4. It
appeared as a loosely consolidated shell deposit with a much
higher soil to shell ratio than Zone m. Based on its horizontal
and vertical positioning vis-A-vis other strata, Zone IV appears
to represent soil build-up-and less frequent refuse deposi-

tion-after Zone Ha had already been deposited. However,
artifacts indicated that Zone VI, like Zone II, dated to the
Colorinda period.
Feature 5 (formerly Area A) was a unique shell deposit
encountered within all Block units, save for Units 6 and 10
(see Figure 8). In all areas where it was encountered, Feature
5 was stratigraphically positioned immediately below Zone Ha.
It consisted mostly of large, whole oyster shells with almost no
associated soil matrix. Feature 5 demonstrated the highest
shell to soil ratio of any context encountered during fieldwork
and it was marked by a dearth of animal bone and artifacts. In
fact, only three sherds were recovered from Feature 5 and all
were sand-tempered plain. The lack of soil accumulation
combined with the fact that the feature is thickest near its
center and tapered at its edges suggest it represents a refuse
deposit in which large numbers of oyster shell were heaped


2006 VOL 59(2)



Zone I black (10 YR 2/1) humus 20 cm
Zone II dark brown (10 YR 2/2) to black (10 YR 2/1) shell midden

Zone Ila black (10 YR 2/1) shell midden, with infrequent shell
Zone IV dark grayish brown (10 YR 3/2) shell midden
Zone V black (10 YR 2/1) midden
Zone VI brown (10 YR 3/1) sand, with infrequent shell
Zone VII yellowish brown (10 YR 5/4) fine sand
Feature 5 shell deposit, with little soil
Feature 6 black (10 YR 2.5/1) midden

Figure 7. Block A (Units 4 and 3) trench west wall profile.


Zone I- black (10 YR 2/1) humus 0
Zone II dark brown (10 YR 2/2) to black (10 YR 2/1) shell midden
Zone Ila black (10 YR 2/1) sand, with infrequent shell
Zone III black (lo YR 2/1) shell midden
Zone VII yellowish brown (10 YR 5/4) fine sand
Feature 5 shell deposit, with little soil

Figure 8. Block A (Units 6, 10 and 11) trench north wall profile.


directly on a former ground surface (atop earlier refuse
deposits) over a short interval of time. Shell from Feature 5
yielded a 2-sigma calibrated radiocarbon age range of A.D.
440 to 710, suggesting it represents the byproduct of oyster
harvesting, meat extraction, and shell disposal during the local
Late Swift Creek Period.
Feature 6 (formerly Area C) was a shallow, intrusive pit
dug into the upper part of Feature 5, within the northwestern
(grid) corner of Unit 3 (Level 5-6), extending into the unit's
north and west walls. It consisted of a rich, black (10 YR
2.5/1) midden with loosely consolidated shell and small
pockets of ash. There was a high proportion of calcined shells
along with abundant charcoal flecking and common chunks of
charcoal; many shells were burned to a gray-blue color. The
exposed segment of Feature 6 measured 72 by 48 cm, with a
maximum thickness of 23 cm. The quantities of charcoal and
calcined shell suggest that this area was exposed to fire,
perhaps representing a campfire or oyster-roasting pit.
Previously deposited shells were likely calcined or suffered
heat-induced breakage due to exposure to heat emanating from
the roasting or fire pit dug into the shell midden. Large
quantities of oysters could have been roasted, the meat
extracted, and the emptied shells returned to the basin along
with other refuse. Eventually the feature was covered by the
deposition of a more consolidated shell midden (Zone II).
Three pottery fragments were recovered from Feature 6 that
included 1 Colorinda Plain and 2 sand tempered plain sherds.
A small amount of animal bone was recovered, some of which
was partially burned.
Zone Vwas an organically-rich, black (10 YR 2/1) midden
encountered in the grid-southern part of Unit 4. Its lower
portion (Level 7) was designated Area B, although inspection
of the unit wall profile indicated that it was actually first
encountered in Level 5. Zone V was slightly darker than Zone
IV, but contained less shell. Pottery from this zone was more
mixed than in Zones II and IV, yielding both Colorinda and
Swift Creek wares. This appears to be the result of incidental
artifact admixture of earlier and later refuse. Zone V appears
to be a Late Swift Creek refuse deposit, whereas Zone IV is a
more recent Colorinda midden.
Zone VI was a brown (10 YR 3/1) sand matrix with little
shell encountered at the shell midden-subsoil interface,
primarily in Unit 3. Its color apparently derived from the
leaching of organic materials within the overlying refuse
deposits. No artifacts and only a few pieces of bone were
recovered from this leachate zone. A slight dip or low spot in
Zone VI was designated Area D. This section of Zone VI
yielded 5 diminutive sherds and a handful of faunal remains.
Zone VII was a yellowish brown (10 YR 5/4) fine sand
subsoil that contained rare shell inclusions. It was encoun-
tered in all units below the shell midden. For the most part,
few artifacts were recovered from Zone VII, although portions
of an Orange fiber tempered bowl were recovered in Unit 8
near base of Level 9. Interestingly, this was the only area
within Block A where Orange pottery was recovered.
Two very distinct stains were encountered in Level 11 of
Unit 6. Feature 3 was a roughly circular stain of very dark

grayish brown (10 YR 3/2) soil exposed at a depth of 2.32
mbd. It was 8 cm in diameter and 4 cm deep. In profile the
stain displayed inward sloping sides and a round bottom; its
interior core was darker with charcoal flecking. Feature 4 was
located 23 cm to the grid-southeast. In plan it was identical to
Feature 3, with a diameter of 9 cm. Feature 4 was shallow, a
little more than 2 cm in depth, with an indistinct profile.
While these features may represent possible postholes, they
also may reflect the work of burrowing animals.
Pottery totals for each unit are presented in Table 2 (see
Ashley 2004 for detailed artifact provenience information) and
suggest evidence of four occupational components. The
primary occupation, represented by Colorinda, St. Johns, and
sand-tempered wares in Zones I, II, IIa, IV and Feature 6 took
place during the Colorinda Period. Earlier Woodland period
Deptford (Zone III) and Late Swift Creek (Zone V, Feature 5)
occupations were encountered beneath or along the margins of
the Colorinda refuse deposits. Finally, a Late Archaic Orange
Period component was represented by the remains of what is
probably a single vessel recovered from beneath the shell
midden in Unit 8.
Table 3 presents the results of Block A excavations by
level. For the most part, Levels 1-6 represent the heart of the
Colorinda shell midden (Zone II). There was mixing of
Colorinda and Swift Creek sherds in Levels 4 and 5, but this
was mostly restricted to Unit 4 along the outer margin of the
Colorinda midden. UnmistakablyDeptford and Orange pottery
were situated below the Colorinda and Swift Creek deposits.
Other artifacts from Block A included 7 sherd hones, 4
nondecortication flakes, 1 secondary flake, and 11 sandstone
fragments. Four whelk hammers, 1 whelk punch or gouge, 1
quahog clam scraper or spoke shave, and 2 disk beads com-
prise the shell artifacts. Finally, bone artifacts included a
fid/awl fragment, 5 pin fragments, 2 polished pieces of bone,
and 3 modified bone fragments.
In sum, Zone II and Feature 5 were clearly the most
consolidated and most pervasive deposits within Block A. The
two were easily distinguished from one another. Zone I was
manifested as a dark organic shell midden, while Feature 5
was a densely packed amalgam of mostly whole oyster shells
with very little soil. Temporally Zone II dates to the Colorinda
Period (ca. A.D. 850-900), whereas Feature 5 dates to the Late
Swift Creek Period (ca. A.D. 500-850). Zone V was an
organically-rich, black midden encountered in the grid-
southern part of Unit 4, and artifacts suggest it represents a
Late Swift Creek refuse deposit (although mixing with
Colorinda artifacts was evident).

Artifact Analysis


Pottery was by far the most common artifact category
recovered from the Cedar Point site. Two-thousand, one-
hundred sixteen fragments of prehistoric pottery were recov-
ered from shovel testing, unit excavation, and surface recon-
naissance (Table 4). Of these, 920 were sherds "larger than 2




Table 2. Pottery totals by unit, Block A.

U-3 U-4 U-6 U-8 U-8A U-9* U-10 U-11 Total Percent
sand tempered 15 35 4 73 3 14 36 26 206 35.1
sand tempered check 3 1 4 0.7
sand tempered simple 3 3 0.5
sand tempered shell 1 4 5 0.8
Colorinda (SSHT) 19 25 16 32 20 12 31 155 26.4
Colorinda (SHT) 14 13 2 7 6 1 1 44 7.5
St. Johns Plain 11 3 7 6 3 1 1 8 40 6.8
St. Johns Check St 1 1 3 2 2 6 15 2.6
St. Johns Stamped 1 1 0.2
St. Johns eroded 2 3 1 6 1.0
sandy St. Johns 1 1 1 1 1 4 0.7
Swift Creek 3 16 5 12 1 16 1 54 9.2
Deptford 16 16 2.7
Orange 18 18 3.1
grog tempered plain 2 5 7 1.2
grog tempered shell 1 1 2 0.3
grit tempered 6 1 7 1.2
Subtotal 73 105 52 162 8 45 68 74 587 100.0

sherd < 2cm 93 154 130 142 10 61 43 55 688

Grand Total 166 259 182 304 18 106 111 129 1275
* Level 4 some sherds lost in the field
SSHT- spiculate sherd tempered
SHT sherd tempered

cm" and subjected to detailed typological analysis. As a result,
8 formal pottery series and 4 general temper categories were
identified. The former included Orange, Deptford, Swift
Creek, St. Johns, Colorinda, Ocmulgee, San Marcos/
Altamaha, and Irene series wares, types spanning the Late
Archaic through Mission periods. The general temper
categories consisted of sand-tempered, grog-tempered, grit-
tempered, and sandy St. Johns.
At Cedar Point, as elsewhere in northeastern Florida,
Colorinda pottery contains large grog inclusions that represent
broken and crushed potsherds, as indicated by their flat-sides
(Figure 9). Most Colorinda specimens included broken
spiculate-tempered (St. Johns) sherds (abbreviated SSHT in
Tables), while some appeared to contain non-spiculate-
tempered (e.g., sand-tempered) sherds (abbreviated SHT in
tables) (see Table 4). The majority of Colorinda sherds from
Cedar Point were undecorated or plain, although a few eroded
or obliterated sherds may have been stamped or incised. In
addition, a small number of grog-tempered sherds (n=18)
contained fine-sized pieces of fired clay. Sixteen plain and 2
shell impressed sherds were recovered. The grog inclusions
were much smaller than the larger pieces of crushed pottery
present in Colorinda wares. Some of the specimens in this

category (i.e., grog-tempered), particularly those from Block
A, may represent variants of the Colorinda pottery type.
The presence of crushed St. Johns sherds in Colorinda
pottery indicates that the two types were being manufactured
concurrently, particularly in light of the fact that there is no
evidence of earlier St. Johns I refuse deposits in northeastern
Florida from which Colorinda potters could have scavenged St.
Johns sherds to use as temper. Sand-tempered plainwares
appear to be the numerically dominant type in Colorinda
assemblages followed by Colorinda Plain, St. Johns Plain, and
small amounts of St. Johns Check Stamped. To date, only a
few simple stamped, check stamped, and complicated stamped
designs have been recognized on Colorinda paste (Ashley
2006, this issue; Richter 1993; Russo et al. 1993:68).
Of the 265 Colorinda sherds recovered from the Cedar
Point site, the overwhelming majority are undecorated and
tempered with crushed St. Johns sherds. As much as 30
percent of the Colorinda ware fragments, however, were
tempered with non-spiculate sherds, indicating that Colorinda
potters usedboth spiculate (St. Johns) and non-spiculate (sand-
tempered) sherds as tempering agents. Although approxima-
tions, evidence from the Colorinda shell midden in Block A
suggests that a little more than 40 percent of the pottery in


2006 VOL. 59(2)


Figure 9. Colorinda sherd from Block A.

Colorinda assemblages was sand tempered plain; a little less
than 40 percent was Colorinda Plain; and less than fifteen
percent was St. Johns wares (plain outnumbered check
stamped 3:1).

Sherd Hones

Eight of the recovered pottery sherds were used as hones,
and all appear to have been recovered from Colorinda con-
texts. The sandy nature of these modified sherds provided an
abrasive surface upon which to sharpen, smooth, or polish
softer material such as wood and bone. Use in this fashion
resulted in the formation of a distinctive U- or V-shaped wear
pattern or groove across the sherd surface. Several sherds
displayed multiple grooves going in different directions. The
hone collection consists of one St. Johns Plain, 3 Colorinda
Plain, and 4 sand-tempered plain sherds. Although they did
not mend, 3 of the sand tempered sherd hones were recovered
from the same unit and appear to be part of one large broken

Lithic Artifacts

Modified stone from Colorinda contexts included 6 chert
flakes, 11 amorphous chunks of sandstone, and a waterworn

quartz pebble. All but one flake lacked cortex on their dorsal
surface, and the majority were heat treated. The sandstone
fragments, all from Unit 11 Level 5, represent unmodified
masses or clumps of concreted sand. Exteriors range from a
light yellow brown to a mottled reddish brown or ferruginous
color. A few specimens had small flecks of oyster shell
attached to them. These concretions might have been part of
mudflat strata to which live oysters had been attached. If this
was the case, then they may simply have been attached to
collected oysters and discarded along with shells after meat
extraction. Because northeastern Florida lacks naturally
occurring stone outcroppings, it is not surprising that few
lithic artifacts were recovered.

Shell Artifacts

The massive quantities of shells strewn and piled along the
estuarine waters of the lower St. Johns River attest to the
importance of shellfish as a food source among the natives of
northeastern Florida. In addition, the hard shell of certain
species, such as quahog clam and whelk, provided a raw
material for tool and ornament manufacture. This is particu-
larly true for coastal northeastern Florida where the absence of
local lithic sources of any kind renders shell a locally available
substitute in tool production. In the field, all whelk shells (and



Table 3. Block A pottery totals by level (excludes surface finds, wall cleanup, sherds < 2 cm, and features).

1 2 3 4* 5 5A 5-6C 6 7 7 7B 8 8A 8D 9 9b 10 10b Total %
sand tempered 4 11 32 52 52 2 2 26 3 3 3 7 1 2 200 35.9
sand tempered 1 1 1 4 0.7
check stamped
sand tempered 3 3 0.5
simple stamped
sand tempered 1 4 5 0.9
shell impressed
Colorinda (SSHT) 12 7 17 38 61 9 6 1 2 153 27.5
Colorinda (SHT) 8 2 4 4 13 1 1 3 3 1 40 7.2
St. Johns Plain 6 6 2 7 9 3 1 34 6.1
St. Johns Check St 4 2 3 2 1 3 15 2.7
St. Johns stamped 1 1 0.2
St. Johns eroded 3 1 4 0.7
sandy St. Johns 2 1 1 4 0.7
Swift Creek 5 10 18 4 12 5 54 9.7
Deptford 1 1 14 16 2.9
Orange 1 11 2 14 2.5
grog tempered 2 1 3 1 7 1.3
grog tempered 1 1 2 0.4
shell impressed
grit tempered 1 1 0.2
Total 34 32 72 118 156 2 3 50 19 6 5 12 1 26 14 2 5 557 100.1
*Unit 9, Level 4 some sherds lost in the field
A=Area A SSHT- spiculate sherd tempered
B=Area B SHT sherd tempered
C= Area C
D=Area D
b=Zone III


Table 4. Pottery totals.

Shovel Tests Units Total Count (%) Total Weight grams (%)
sand tempered 29 297 326 (35.4) 3598.8 (32.3)
Colorinda (SSHT) 3 180 183 (19.9) 3236.5 (29.0)
Colorinda (SHT) 19 63 82 (8.9) 1187.0 (10.6)
St. Johns 31 102 133 (14.5) 1133.7 (10.2)
sandy St. Johns 1 11 12 (1.3) 222.3 (2.0)
Swift Creek 2 66 68 (7.4) 664.4 (6.0)
Deptford 6 28 34 (3.7) 285.6 (2.6)
San Marcos 9 20 29 (3.2) 148.7 (1.3)
Orange 3 18 21 (2.3) 284.7 (2.6)
grog tempered 7 11 18(2.0) 190.5 (1.7)
grit tempered 4 8 12(1.3) 188.7 (1.7)
Ocmulgee 1 0 1 (0.1) 5.2 (0.05)
Irene 1 0 1 (0.1) 5.6 (0.05)
Subtotal 116 804 920 (100.1) 11151.7 (100.1)

sherd < 2cm 230 968 1199 (56.6) 1505.1 (11.9)
Grand Total 346 1772 2119 12656.8
SSHT- spiculate sherd tempered
SHT sherd tempered

fragments thereof) were collected for laboratory analysis.
Quahog clam and other robust shell species were examined for
evidence of wear, breakage, or utilization. Those demonstrat-
ing such evidence were retained for laboratory analysis.
Most shell artifacts from Cedar Point were made of whelk
(Table 5), a marine gastropod that inhabits shallow waters and
sand flats exposed at low tides. The only identified species
from the Cedar Point site was knobbed whelk (Busycon
carica), a dextral or right-handed type of whelk. Several
whelks in the assemblage display signs of forceful impact in
the form of chipping, shattering, and spelling to their apex,
beak, or along their sides (or any combination of the three).
These battered-edged whelk tools are generally referred to as
hammers or pounders (Marquardt 1992:199-205). Disk beads
were made from rounded and biconically drilled cut sections
of whorl. Finally, a few whelk fragments displayed signs of
being intentionally modified or worked.
The only non-whelk shell artifact was a single modified
valve of quahog clam (Mercenaria spp.). In the field it was
noted that the shell fragment had two notches along its outer
edge and a break to the body that may have been cut. After
closer laboratory examination the "cut" section and one notch
were interpreted as recent breaks, although the other notched
areas appear to have greater antiquity. The edge of the shell
may have been used to shave or scrap a softer material such as
bark or wood.

Bone Artifacts

Bone artifacts were all fragmentary pieces divided into four
categories: awl/fid, pin, polished bone, and modified bone, all

of which were recovered from Block A. The awl/fid category
consists of a single specimen of split mammalian long bone
with a tapered, rounded tip and smoothing (use wear) along its
modified shaft. This artifact may have been used in weaving
or to work textiles or nets. Five small pin fragments were
recovered and each displayed polished, rounded, or nearly
rounded, shafts. Two polished bone fragments exhibited
localized areas) of surface polish or smoothing, probably due
to use wear. Modified bone included 3 pieces that have been
used or worked in some way to alter their natural form;
butchering and postdepositional damage are not included in
this category. Because both the size of individual bone
artifacts and sample itself were small, little can be said about
the bone artifacts from Cedar Point.

Faunal Analysis

Sampled shell middens from the Cedar Point site yielded
a well-preserved assemblage of faunal material that consisted
of animal bone and estuarine shell. As expected on the basis
of site location, shellfish and fish species dominate the
assemblage. While fine-mesh samples were collected, the
faunal analysis presented here consists of vertebrate faunal
remains greater than 1/4", retrieved from three different
contexts within the Colorinda shell midden in Block A.
Tremendous quantities ofbone were observed in the fine mesh,
but these samples await future analysis.
Standard zooarchaeological methods were used during
laboratory analysis (see Ashley 2004 for details regarding
faunal analysis and results). Based on identifiable skeletal
elements specimens were identified to the class, order, genus,



Table 5. Shell artifacts.

Species Description Context Count Weight (g)
Knobbed Whelk Hammer; most of outer whorl removed; Unit 3, L-4 1 280.8
columella shortened and end heavily battered

Whelk (Busycon spp) Drilled disk bead; bead diameter is 8.0 mm Unit 3, L-4 1 0.5
and hole diameter is 3.2 mm
Whelk (Busycon spp) Drilled disk bead; bead diameter is 6.9 mm Unit 3, wall 1 0.2
and hole diameter is 2.5 mm cleanup
Quahog Clam Scraper or spokeshave; notch along edge; Unit 6, L-6 1 10.2
also clean notch and body breakage suggests
postdepositional damage as well
Knobbed Whelk Possible punch or gouge; cut and modified Unit 11, L-5 1 2.7
outer whorl fragment;
Knobbed Whelk Hammer; outer whorl at one time had a Unit 11, surface 1 251.9
hafting whole; later much of whorl was
removed including one side of the hafting
hole; columella shortened and end heavily
Knobbed Whelk Extraction hole, slight beak and lower outer Surface 1 492.8
whorl damage _

or species level whenever possible. In addition to bone count
and weight quantification, Minimum Number of Individuals
(MNI) calculations were made for the three faunal samples.
MNI estimates do not determine precisely how many individu-
als are in the collection, but merely indicate the least or
minimum number of individuals of a species needed to account
for bones in the collection (Casteel 1977; Grayson 1979).

Vertebrate Fauna

The Cedar Point site provides the first information on
Colorinda subsistence. The Colorinda occupants of the Cedar
Point site exploited an array of fauna, including bony fish,
cartilaginous fish, mammal, bird, and reptile. Table 6
provides a breakdown of the bone count, bone weight, and
MNI for these taxa by sample. Combined the three samples
contained 7908 skeletal elements that weighed 1.49 kg and
included at least 217 individuals (Table 7). While one sample
(Unit 9, Level 4) included far more bone than the other two
(Unit 8 Level 3 and Unit 10 Level 5), the three were remark-
ably similar in composition.
Fish were undoubtedly the most important subsistence
resource to the Colorinda inhabitants of the Cedar Point site.
Combining the three samples, fish constitute more than 90%
of the counted elements, more than 75% of the bone weight,
and more than 80% of the MNI (see Table 6). The most
abundant fish were catfish of the order Siluriformes. Identi-
fied species were gafftopsail catfish (Bagre marinus) and
hardhead catfish (Ariusfelis), both of which are sea catfishes
of the family Ariidae. With respect to MNI, there were 86
Ariidae, 8 hardhead catfishes, and 2 gafftopsail catfishes.

Modem fisheries data indicate that hardhead catfishes are
typically larger and more abundant in Atlantic coastal waters
than gafftopsail catfishes. Combined catfish made up 44% of
the total MNI, 49% of the total fish bone count, and 60% of
the total fish bone weight. The general trend observed among
the catfish bones and otoliths throughout the Colorinda shell
midden in Block A was that while a few very large catfish
were represented, the majority was in the small to medium
Sea catfishes can be present within bays and estuaries on
a year-round basis, but typically move to shallow ocean waters
during cold weather (November to February) and return
inshore in the spring (McClane 1974; Reitz 1988). On
average, catfishes are most common in the lower reaches of
estuaries during the months of June through October. Water
temperatures and salinity levels are the primary factors
dictating their seasonal distributions. Catfishes are opportu-
nistic feeders over mud and submerged sand flats, and depth
preferences appear related to water temperatures and bottom
composition (McClane 1974). Both species can be caught with
baited hooks, while the use of nets would require fishing
technology adapted to deeper estuarine waters (Quitmyer
From the standpoint of bone count and weight, the next
most prevalent fish in the samples was flounder (Bothidae).
However, flounder was represented by only 5 MNI. Flounder
are bottom-dwelling fish found in bays and shallow estuarine
waters throughout the year, depending on species (Reitz
1988:144). Many of the flounder in the sample were small
sized, suggesting they were caught with the aid of nets, traps,
or weirs.

2006 VOL. 59(2)



Table 6. Faunal analysis results.

Unit 8 Level 3 =Unit 9 Level 4 Unit 10 Level 5 ~ on-
Count Wt (g) MNI Count Wt (g) MNI Count Wt (g) MNI MNI

Mammalia, large 4 21.8 1 1
Mammalia, medium-large 9 11.4 2 14 7.7 1 3 0.7 3
Mammalia, small-medium 34 9.3 2 0.2
Mammalia, unidentified 1 5.0 -
Odocoileus virginianus 1 0.8 1 -- 1
Didelphis virginiana 5 7.8 1 1 0.1 1 -- 2
Geomyidae 2 0.3 2 5 0.6 3 1 0.1 6
Rodentia 3 0.5 1 0.1 1 0.4 1 1
Sciurus spp.2 0.5 1 6 1.6 2 1 3
Sylvilagus spp. 3 1.2 1 6 4.0 1 2
Aves large 3 6.6 2 2
Aves medium 17 5.0 2 2
Aves -- 6 1.7
Anatidae, large 9 7.3 2* 2 1.4 1 3
Anatidae, small-medium -7 1.2 1* 1
Testudines 13 3.6 25 11.3 1 6 1.4 1 2
(f Terrapene carolina) 17 16.9 2 2 0.9 1 3
(cf Gopheruspolyphemus) 3 11.5 1 45 28.9 1 2
Kinosternidae 73 16.8 3 65 12.4 1 6 1.3 1 5
Cheloniidae 1 9.2 1 1
Serpentes 1 0.1 1 1
Chondrichthyes 3 0.3 1 0.1 1 1 0.1 1 2
Rajiformes 3 0.8 1 4 2.8 1 -- 2
Lamniformes 1 9.0 1 1
Osteichthyes 500 61.1 5 1726 193.6 712 83.8 1 6
Lepisosteus spp. 3 0.1 1 9 18.6 2 13 1.1 1 4
Brevoortia spp.- 1 0.1 1 1
Amia calva 1 0.1 1 -- 1
Clupeidae 4 0.1 1 7 0.3 4 11 0.2 1 6
Siluriformes 415 71.8 16 1647 290.7 51 633 108.2 19 86
Ariusfelis 71 11.1 5 3 0.2 3 38 5.3 8
Bagre marinus 102 23.3 2 329 95.7 275 68.1 2
Gobiidae 58 2.6 2 24 1.2 1 55 2.1 2 5
Caranxspp.- 6 18.6 2 2
Sparidae/Sciaenidae 12 1 3 55 11.9 2 21 2.7 2 7
Sciaenops ocellatus 1 0.3 1 1
Archosargus 10 2.8 2 2
probatocephalus _______ _________ _____ ________
Cynoscion spp. 27 2.2 3 19 3.0 8 45 9.3 4 15
Cynoscion nebulosus 3 1.4 2 2
ynoscion realis 1 0.3 1 1
Micropogonias undulaues 1 0.2 1 1 0.3 1 2
Pogonias cromis 6 2.7 1 2 0.7 1 2
Mugil spp. 24 1.0 1 114 5.8 11 16 0.6 1 13
Bothidae 34 3.1 2 126 12.9 2 28 3.6 1 5
UID vertebrata 132 24.5 144 37.4 25 4.5 -
Total 1580 322.5 63 4424 785.1 112 1904 307.3 42 217

Mullet (Mugil spp.) was nearly as common as flounder
with regard to bone count and weight, but was higher in MNI
(MNI=13). Adult mullet are typically found in the estuary
throughout the year, although most migrate to offshore ocean
waters when temperatures drop to below 450 F (Reitz

1988:144). Mullet spawn offshore and return to the estuary in
schools between May and August. Many of the mullet in the
sample can be described as fingerling size. Schools ofjuvenile
mullet are often found feeding in tidal creeks and shallow
nearshore waters in the late spring through late summer (May-




Table 7. Faunal analysis totals (three samples combined).

Count % Wt (g) % MNI %
Mammal 104 1.3 144.3 9.7 19 8.8
Bird 44 0.6 23.2 1.6 8 3.7
Reptile 257 3.2 114.3 7.7 14 6.5
Cartilaginous 13 0.2 13.1 0.9 5 2.3
Bony fish 7189 91.0 1123.8 75.7 171 78.8
TTnidntified 301 3.8 66.4 4.5 -
Total 7908 100.1 1485.1 100.1 217 100.1

September) (Lee et al. 1984:218). Netting is a productive way
to capture this swiftly-moving, schooling fish.
Drums and sea trout (Sciaenidae) were next in frequency.
Although limited in bone count and weight, sea trout
(Cynoscion spp.) were represented by a relatively high MNI of
15. In addition, 2 spotted sea trout and 1 gray sea trout
weakfishh) were identified. Other Sciaendiae represented in
the samples include black drum, redfish or red drum, and
Atlantic croaker. Although slight variations occur depending
on species, these fishes are found throughout the year in the
estuary (Reitz 1988:144).
The family Sparidae consists ofporgies, represented in the
assemblage by sheepshead (MNI=2). They are common in
shallow coastal waters and along banks near shellfish colonies.
Their teeth are similar to some of the drums (family
Sciaenidae), and tooth fragments are often difficult to identify
to the species level. Seven MNI were assigned to a general
Sparidae/Sciaenidae category. Clupidae (herring and shads)
are shallow water fishes represented by 7 MNI, including 1
menhaden. Some species are present in the estuary year-
round, while others move out at different times to spawn.
Gobies (Gobidae) are small to tiny fishes represented by a MNI
of 5. The specific habits and movement pattern varies with
species, although almost all are bottom dwellers. Although
fine mesh faunal samples were not analyzed during this study,
a quick scan of several samples revealed the presence of large
numbers of gobie vertebrae along with those of killifish and
small herring. Undoubtedly these fishes were caught with fine
mesh equipment, such as nets, seines, or weirs.
The largest fish in the analyzed collection was crevalle
jack. Individuals of this species can reach lengths of up to 5
feet (1.5 m) and weigh as much as 20 pounds (9 kg). Six bone
fragments representing two individuals were recovered. These
jacks can be found in deep ocean waters, estuaries, and
riverine settings depending on life stage of the fish. Juveniles
occur in shallow brackish waters, while adults can range from
shallow inshore waters to deep offshore waters. They often
travel in schools, although larger individuals may be found
moving alone.
Many of the fishes discussed above use estuaries as
nurseries. The young are found in shallow estuarine waters,

whereas the larger fish will at times move to deeper waters.
Thus, species composition within the estuary can vary season-
ally, and there is also fluctuation in the size of individuals of
each species and their location within the estuary. Salinity
levels also have a major impact on distributions compounding
variation patterns. In general, the total number of fishes is
much higher in the summer and fall than in the winter and
spring (Reitz and Quitmyer 1988:100-102).
Cartilaginous fish remains from Cedar Point fall into the
class Chondrichthyes, which includes sharks, rays, and skates.
In all, 13 skeletal elements representing 5 MNI were recov-
ered. Five elements and 2 MNI were unidentified
Chondrichthyes, 7 elements and 2 MNI were identified as
Rajiformes (order that includes skates and rays), and 1 was
identified as Lamniformes (order of shark). Certain sharks
move in to the estuaries during the wanner months to take
advantage of its role as a fish nursery. Rays are common
throughout the year, but peak numbers occur during the
summer. Small to medium sized individuals could have been
taken in sturdy nets or weirs (Quitmyer 1985).
Mammalian remains were limited and included 104
skeletal elements that weighed 74.3 g and represented 9 MNI.
Mammals accounted for 1.3% of the total bone count, 5.0% of
total bone weight, and 8.8% of the total MNI. Sixty-seven
elements (64%), weighing 56.1 g (39%), were unidentifiable,
but were separated on the basis of relative size (e.g., large,
medium to large, small to medium). Of the remainder, one
bone was identified as deer (MNI=1); 6 were identified as
opossum (MNI=2); 8 were identified as pocket gopher
(MNI=6); 5 were identified as rodents (MNI=1); 8 were
identified as squirrel (MNI=3); and 9 were identified as rabbit
(MNI=2). These animals could have been hunted or trapped
throughout the year within close proximity to the Cedar Point
Save for a single snake vertebra, turtles were the only
reptilian remains recognized within the Cedar Point samples.
Some turtle remains were only identified to the broader order
Testudines, while others were classified to either the family
Kinosternidae (musk and mud turtles) or family Cheloniidae
(sea turtles). Gopher tortoise (Gopheruspolyphemus) andbox
turtle (Terrepene carolina) were the only species recognized.


2006 VOL. 59(2)



Table 8. Boonea impressa size data.

P 4 45=45 0 | 6 | 76
[TOTAL] 34 100.0

The most commonly occurring skeletal part was shell (cara-
pace and plastron). A breakdown by MNI is as follows: musk
and mud turtles (n=5), box turtle (n=3), turtles (n=2), gopher
tortoise (n=2), and sea turtle (n=1). Turtles constituted 6% of
the total MNI. These slow moving animals were probably
opportunistically collected from land and water settings by the
native inhabitants of the Cedar Point site. Sea turtles (e.g.,
green, loggerhead, leatherback) typically inhabit the offshore
surf zone, but are found along the beaches during the nesting
season (April through August) (Larson 1980:129-132).

Bird remains consisted of 44 bones that weighed 23.2 g.
Most of the avifaunal remains were identified simply as Aves
(birds), although several elements belonging to the Family
Anatidae (ducks, geese, swans) were recognized. MNI counts
included 2 large birds and 2 small birds, along with 3 large
Anatidae and 1 small to medium Anatidae. Birds were
presumably hunted and netted when the visited the tidal flats
and marshes as well as freshwater sources.

Comment on Shellfish

Although no formal attempts were made to quantify
shellfish ratios, American oyster (Crassostrea virginica) was
clearly the most abundant shellfish species exploited. Field
estimates suggest that oysters easily constituted more than
90% of the shellfish species within the Block A. Oysters are
sessile creatures that form beds on solid strata of rock or shell
in shallow brackish to marine waters. During precolumbian
times oysters would have been readily available off the
southern end of Black Hammock Island. Other species
observed in the midden and locally available included quahog
clam (Mercenaria spp.), Atlantic ribbed mussel (Geukensia
demissa), stout tagelus (Tagelus plebeius), marsh periwinkle
(Littorina irrorata), eastern mud nassa (/lyanassa obsolete),
knobbed whelk (Busycon carica), and giant Atlantic cockle
(Dinocardium robustum).

Inferring Season of Oyster Collection
Retrodicting season of aboriginal resource exploitation is

critical to interpreting coastal settlement and subsistence
patterns. Recent attempts along these lines have employed

metric analysis of certain biological data. Specifically, the
annual growth pattern of certain shellfish species has proven
auseful measure of season of exploitation. Because oysters are
often the dominant shellfish species on Atlantic coastal sites,
archaeologists have attempted to develop means by which to
determine season of oyster collection. As a result, archaeolo-
gists have focused their attention on impressed odostomes or
Boonea impressa (Russo 1991), a small gastropod that are
commensal with oysters. Measurement of the size of im-
pressed odostomes at the time of death can be used to infer
season(s) of the year when oysters from the same archaeologi-
cal context as the recovered odostomes were harvested (Russo
1991, Russo et al. 1993).

Modal size class data for the CS #3 sample are presented
in Table 8 and depicted graphically in Figure 10. While small
the archaeological sample is adequate and represented by a
truncated modal peak across the 3.0 mm, 3.5 mm, and 4.0 mm
size categories. Together these three sizes account for over 60
percent of the odostomes in the sample. Size intervals 4.5 mm
and 5.0 mm contain 6 and 5 odostomes, respectively. Finally,
size intervals 2.5 mm and 5.5 mm are each represented by a
single odostome. Thus, these data indicate that the majority of
odostomes in the sample were in the 3.0 - 4.0 mm size range
at the time of death.

Comparing the distribution plot for CS#3 with the master
monthly modal size plot developed by Russo (1991:210) for
northeastern Florida, a summer to autumn season of oyster
collection (July through November) is suggested, perhaps with
a peak from mid-July to September. This season of occupation
is corroborated by fish composition and size data (e.g.,
fingerling mullet, catfish quantities and sizes), which also
indicates a summer occupation. We must keep in mind,
however, that the subsistence information merely tell us what
a specific group of Colorinda people were exploiting and
eating at the Cedar Point during the summer. More data are
needed from other sites representing different seasons of the
year in order to reconstruct adequately the settlement-subsis-
tence cycle of Colorinda groups in northeastern Florida.

Radiometric Dating Results

Although pottery types provide solid relative dating
information, radiometric age estimates were needed to pin-
point the temporal placement of the site’s occupational
components. Four samples from separate midden contexts
within Block A were submitted to Beta Analytic, Inc. for
analysis. Oyster shells were taken from two midden contexts
and submitted for radiocarbon dating, while soot from a
Colorinda Plain sherd and a Swift Creek Complicated
Stamped sherd were submitted for Accelerator Mass Spectrom-
etry (AMS) assay. All four samples were calibrated, and none
has a standard deviation greater than 60 years (Table 9).

The calibrated 2-sigma dates for the two Colorinda
contexts are A.D. 720-960 (AMS on soot) and A.D. 740-1020
(radiocarbon on shell). Statistically, these dates are very
similar and suggest that the organic shell midden in Block A
was deposited over brief period of time, perhaps even a single






2 2.5 3 3.5 4 4.5 5 5.5 6 6.5
Spring Summer Fall Winter

Figure 10. Boonea impressa modal size distributions, CS#1, L-4.


Table 9. Radiometric dates from Block A, Cedar Point site.

Measured C12 Conventional Calibrated 1 Calibrated
C14 age ratio C-14 Age Sigma (AD) 2 Sigma
Beta # Material Context BP) (o/oo) (BP) with intercept (AD)
Feature 5
181303 Oyster U-8A, L-6 1460 +70 -3.6 1810 +70 540 (610) 670 440 710
182335 Soot U-8,L-3 1250 + 40 -25.4 1240 + 40 705 (775) 855 680 885
180189 Soot U-10, L-5 1190 + 40 -24.7 1190 + 40 780 (870) 890 720 960
180188 Oyster U-9, L-2 1190 + 60 -5.2 1520 + 60 800 (890) 960 740 1020

occupation episode. The other two radiometric dates appear to
be associated with the local Late Swift Creek Period, ca. A.D.
500-850 (Ashley and Wallis 2006). Beta-182335 was a
sample of soot from a Swift Creek Complicated Stamped sherd
that yielded a calibrated two-sigma date range of A.D. 680-
885. This date is statistically identical to a Swift Creek date
from the nearby Dent Mound (8DU68) on Pelotes Island to the
west. The sherd, however, was recovered from the Colorinda
midden (Block A, Zones II), so it might help date this occupa-
tion. Alternatively, the sherd may originally have been
deposited within the earlier Swift Creek midden (Block A,
Zone V) located along the margins of, and partially overlap-
ping, the Colorinda midden and was displaced and mixed
during subsequent site reoccupation.
Beta-181303, the oldest radiometric date, was associated
with Feature 5, a zone of shell with very little soil, bone, and
pottery situated beneath the Colorinda midden. The ceramic
fragments were all sand tempered plain sherds less than 2 cm
in size. While no Swift Creek sherds were recovered from
Feature 5, a2-sigma calibrated date of A.D. 540-670 combined
with its stratigraphic positioning beneath the Colorinda
midden suggest the shell-dominated deposit was laid down
during Late Swift Creek times. The Feature 5 date is similar
to another 2-sigma calibrated radiometric date from the Dent
Mound (A.D. 550-660). Although open to interpretation, the
dates suggest seventh and/or eighth century Swift Creek
occupations, and early ninth century Colorinda habitation of
the Cedar Point site.


The 2003 UNF-NPS Summer Field School afforded the
opportunity to explore the southern portion of the Cedar Point
site in some detail for the first time. Shovel testing and unit
excavation provided a wealth of data on the site's ninth-
century Colorinda inhabitants, who appear to have occupied
this section of the site during the summer, at which time they
targeted small fish for subsistence, although some catfish were
quite large. Catfish was the dominant catch, with smaller
amounts of mullet, flounder, and sea trout taken from the
shallow estuarine waters. Additional information is needed
from other Colorinda sites to reconstruct the settlement-

subsistence cycle and to determine how the Cedar Point site
articulated with contemporaneous sites in the region.


I would like to thank Buzz Thunen and John Whitehurst for
helping with various facets of the 2003 UNF Field School. John and
Superintendent Barbara Goodman were instrumental in getting the
project off the ground, while Ann Lewellan of NPS helped to
catalogue and inventory the field school materials. Vicki Rolland
deserves special recognition for analyzing the faunal collection.
Keith Stephenson funded one of the AMS assays. The UNF field
school participants deserve a great deal of credit for their admirable
service and dedication in the field and laboratory, they include Stacey
Bailey, Jessica Cleary, PamDipietro, AutumnDubois, James Gordon,
Rebeca Gorman, Cindy Kenard, FeliciaKilgore, Stefan Krause, Kara
Larsen, Kim Lidikay, Lisa Parks, Keith Pricer, Chelsea Quinn, Kevin
Salazer, Matt Schmitt, and Jamie Spruell. I also greatly appreciate
David Myantt's help with the graphics. Finally, thanks to Ryan
Wheeler, Ann Cordell, Vicki Rolland, and the anonymous reviewers
for their helpful comments.

References Cited

Ashley, Keith H.
2004 Archaeological Testing at the Cedar Point Site (8DU81):
Results of the 2003 UNF-NPS Summer Field School.
Report on file, Florida Master Site File Division ofHistori-
cal Resources, Tallahassee.

2006 Colorinda and its Place in Northeastern Florida History.
The Florida Anthropologist 59(2):91-99 (this issue).

Ashley, Keith H., and Neill Wallis
2006 Northeastern Florida Swift Creek: Overview and Future
Research Directions. The Florida Anthropologist 59:5-18.

Casteel, R.
1977 Characterization ofFaunal Assemblages andthe Minimum
Number of Individuals determined from Paired Elements:
Continuing Problems in Archaeology. Journal ofArchaeo-
logical Science 4:125-134.

Grayson, Donald K.
1979 On the Quantification of Vertebrate Archaeofaunas.
Advances in Archaeological Method and Theory, Vol. 2,
edited by M. Schiffer, pp. 199-237. Academic Press,





Johnson, A. Sydney, Hilburm Hillestad, SherylF. Shanholtzer, and
G. Frederick Shanholtzer
1974 An Ecological Survey of the Coastal Region of Georgia.
National Park Service Scientific Monograph Series,
Number 3.

Jones, William
1985 A Report on the Cedar Point Ruins, Black Hammock
Island, Duval County, Florida. Ms. on file, Jacksonville
Main Library, Jacksonville.

Larson, Lewis H., Jr.
1980 Aboriginal Subsistence Technology on the Southeastern
CoastalPlainDuringtheLatePrehistoricPeriod. Univer-
sity of Florida Press, Gainesville.

Lee, C., I. Quitmyer, C. Espenshade, and R. Johnson
1984 Estuarine Adaptations During the Late Prehistoric Period:
Archaeology of Two Shell Midden Sites on the St. Johns
River. Office of Cultural and Archaeological Research,
Report oflnvestigations No. 5, University of West Florida,

Marquardt, William H.
1992 Shell Artifacts from Caloosahatchee Area. In Culture and
Environment in the Domain ofthe Calusa, edited by W. H.
Marquardt, pp.191-228. Monograph 1, Institution of
Archaeology and Paleoenvironmental Studies, University
of Florida, Gainesville.

McClane, A.J. (editor)
1974 McClane's Field Guide to Saltwater Fishes of North
America. Holt, Rinehart, and Winston, New York.

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

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

Quitmyer, Irvy R.
1985 Aboriginal Subsistence Activities in the Kings Bay Local-
ity. InAboriginal Subsistence and SettlementArchaeology
of the Kings Bay Locality, Vol. 2, edited by W.H. Adams,
pp. 73-91. University ofFlorida, Department ofAnthropol-
ogy, Reports of Investigations No.2, Gainesville.

Reitz, Elizabeth J.
1988 Evidence for Coastal Adaptations in Georgia and South
Carolina. Archaeology of Eastern North America

Reitz, Elizabeth J., and Irvy Quitmyer
1988 Faunal Remains from Two Coastal Georgia Swift Creek
Sites. Southeastern Archaeology 7:95-108.

Richter, Bob
1993 Surface Collections from 8DU66, The McCormick site.
Ms. on file with author.

Russo, Michael
1991 A Method for the Measurement of Season and Duration of
Oyster Collection: Two Case Studies from the Prehistoric
South-east U.S. Coast. Journal ofArchaeological Science

Russo, Michael, Ann S. Cordell, and Donna L. Ruhl
1993 The Timucuan Ecological and Historical Preserve Phase
IIIFinalReport. National Park Service, Southeast Archeo-
logical Center, National Park Service, Tallahassee.

Sears, William H.
1957 Excavations on Lower St. Johns River, Florida. Contribu-
tions of the Florida State Museum 2, Gainesville.

1959 Two Weeden Island Period Burial Mounds, Florida.
Contributions of the Florida State Museum 5, Gainesville.

2006 VOIL. 59(2)



Stone Tool Traditions in the Contact Era. Edited by Charles
R Cobb. 2003. University of Alabama Press, Tuscaloosa.
214 pages, 43 b&w illustrations, 15 tables, bibliography,
index. $34.95 (paper).

Department of Anthropology, Florida State University,
Tallahassee, FL 32304

Most lithic analyses have at their core the assumption that
prehistoric people made choices about raw material and tool
design based upon the efficient use of energy. The efficiency
paradigm, however, begs the question: why people did not
immediately abandon stone in favor of metal, once introduced,
as the preferred raw material? Stone Tool Traditions in the
Contact Era offers an alternative paradigm by putting stone
tools and their eventual replacement by metal tools into a
cultural context. The case studies in this volume illustrate that
Native Americans' raw material decisions were influenced as
much by their responses to European contact as by the needs
for energy efficiency.
This edited volume of nine relatively short case studies from
North America and Hawai'i examine the process of replace-
ment of stone tools with metal tools. In these chapters we can
glimpse the variety and commonalities of responses between
different indigenous peoples as they negotiated European
influences. In Chapter 1, Charles Cobb introduces and
reviews recent ideas about the nature of interactions between
Native Americans and the early Europeans. The general
chronology of that interaction gives structure to the organiza-
tion of the case studies, which progress from initial, limited
contact with European metal tools, to frequent contact with
Europeans and more metal, and finally, to incorporation into
mercantile and capitalist economic systems. The traditional
view that stone tools were rapidly replaced with technologi-
cally superior European metal tools as soon as those became
available has been all but debunked. These chapters show that
the transition was not immediate and in some cases never
complete, and that Native Americans' attitudes towards newly
introduced metal tools and traditional stone tools changed
along with their social and economic relationships with
The first three case studies concern some of Native
Americans'earliest exposures to metal tools. In Chapter 2,
Charles Cobb andDino Ruggiero discuss the mortuary context
of iron artifacts at the King Site in northwestern Georgia, a
Mississippian village occupied by the Coosa in the mid-
sixteenth century. Although only eight iron objects were
recovered from five of More than 250 burials, most were
interred with flint knappers, which may indicate that iron was

incorporated by these individuals as a raw material option. In
Chapter 3, which is one of the best chapters in the book,
George Odell presents a thorough analysis of the use of metal
and stone tools at the Lasley Vore site, south of Tulsa,
Oklahoma. This site was occupied for no more than 30 years
after European goods were available in quantity. Odell
believes that the process of enculturation, at least initially, was
limited to the adoption of a new raw material (i.e., metal) into
traditional stone tool forms. Chapter 4 is a review of a
preliminary reassessment by Jay Johnson of National Park
Service collections of several Chickasaw settlements in
northeastern Mississippi. Johnson hopes to create a fine-
grained chronology of these early eighteenth century sites to
examine the process and rate of stone tool replacement, and
how these coincide with changes in social and economic
organization within and between villages that could be related
to rise of the deer-skin trade with Europeans.
In Chapter 5, Michael Carmody examines the Oneida
Iroquois Cameron site in central New York, occupied from ca.
A.D. 1590-1600. Carmody emphasizes the need to think about
stone tools in a mixed techno-symbolic framework rather than
strictly as an issue of technological efficiency. But, like
Odell's work at Lasley Vore, Carmody shows that a functional
analysis cannot be adequately done without a thorough use-
wear analysis. In Chapter 6, Michael Nassaney and Michael
Volmar show that social and gender roles are reflected in
changes in the traditional forms of stone tools used by the
Narragansetts of southern New England. Although European
trade goods in the seventeenth century were not immediately
adopted unless they were deemed acceptable substitutes for
traditional tools, not all acceptable substitutes were adopted.
Traditional stone pestles and smoking pipes were retained, but
in modified form, to reproduce new social and gender roles in
the face of European-induced disease, land pressures, and
commodity markets. The authors attribute these modifications
to an effort by men to retain the ritual use of tobacco and
women to assert their new roles in the changing economy. In
Chapter 8, Mark Wagner also examines changes in tobacco
smoking traditions among the Potawatomi of Illinois. After
contact, the Potawatomi started using European clay pipes but
continued to make and use stone pipes. The apparent retention
of traditional tobacco smoking with stone pipes may have been
related to a concurrent revitalization movement, in which the
Potawatomi were admonished by Shawnee Prophet and other
nativists to reject all European goods.
In Chapter 7, James Bayman looks at the replacement of
stone adzes for canoe-making with iron adzes in Hawai'i. The
rate of replacement was determined not just by chiefly control
over access or a population's proximity to the seaports, but
also by the different quality of the final tool product. Although


VOL. 59(2)


JUNE 2006


more efficient, metal tools produced a different "style" of
canoe that was initially less desirable. Rather than simply a
matter of efficiency, the replacement of traditional stone adzes
resulted from a confluence of changing canoe style, greater
availability of metal tools, and the eventual loss of stone adze
A site that holds promise for unraveling the complex social,
economic, and gender relationships of Native Americans and
Europeans in the early nineteenth century is discussed in
Chapter 9 by Stephen Silliman. He examines the use of stone
and metal tools at Rancho Petaluma, a large Mexican-Califor-
nian ranch in Northern California, where Native Americans
resident. Although iron and other metals must have been
plentiful, Native Americans, presumably men, continued to use
obsidian and chert for utilitarian purposes, while women,
presumably, used metal sewing implements. At another
interesting site, Mark Cassell explores in Chapter 10 Native
Alaskans' continued use of chert scrapers into the late nine-
teenth century in the Alaska fur trade. It may not be surpris-
ing that chert scrapers were better at processing skins than
metal tools, but their patterns of discard indicates that the
same tool was conceptualized differently by Native Americans
in commercial and domestic contexts.
In the final chapter, Douglas Bamforth summarizes some of
the problems and promises of research into Contact-period
transformations. As Bamford states, although the results are
clear, the process of transformation from stone to metal is
complex and idiosyncratic. While the studies in this volume
illustrate that Native Americans were not passive recipients of
European goods, the data are open to several alternative
I see two themes developed in these studies. First, all lithic
analysis should include an effort to understand the function of
all the tools in an assemblage, both stone and metal. Only
then can one assess whether the replacement of stone by metal
is best explained with an efficiency paradigm. Second, after
the basic lithic analysis is done, the tools must be placed in a
broader social context to understand why they were unmodi-
fied, modified, or replaced. In addition, I think another
promising avenue of explanation touched on in several of the
studies but not thoroughly explored is the use of traditional
stone tools in the maintenance of existing exchange relation-
ships among Native Americans, in which the social value of
the item exchanged is more important than its utilitarian
This book, which is a laudable effort to expand the breadth
of archaeological information we can squeeze from small
pieces of stone, shows both the possibilities and frustrations of
further work in the area of material culture change. The
brevity of these chapters is both a boon and a disappointment;
the salient points are not obscured by too much data, but
several of the studies, such as the Rancho Petaluma study, lack
sufficient data to justify the conclusions. Despite this short-
coming, the book is a stimulating collection that should spur
further research and alternative interpretations in other
instances of material culture change. In the end, we see that
efficiency is still a viable and powerful explanation for raw

material and tool choice, but that the richness of stone tool
analysis will only be realized when it is placed in broader
social context.

The Calusa and Their Legacy: South Florida People and Their
Environments. Darcie A. MacMahon and William H.
Marquardt. 2004. University Press of Florida, Gainesville.
240 pages. $39.95 (hardcover).

Department of Anthropology, University of Florida,
Turlington Hall, Rm 1112, PO 117305 Gainesville, FL 32611-

In The Calusa and Their Legacy: South Florida People and
Their Environments, Darcie MacMahon and William
Marquardt provide an in-depth, multiscalar account of the
intimate and sustainable relationship the ancient Calusa
shared with their estuarine environment, one that is accessible
and relevant to both public and academic audiences. Inspired
by the award-winning museum exhibit, the Hall of South
Florida People and Environment at the Florida Museum of
Natural History in Gainesville, this book explores the dynam-
ics and intricacies of the relationships humans have shared
with the south Florida coastal and estuarine environments
throughout history. Three main thematic issues emerge in the
book through which an anthropological perspective of human
estuarine resource exploitation and conservation is presented:
the ecology of south Florida and the Calusa domain, the
culture and legacy of the Calusa, and contemporary south
Florida peoples and estuarine exploitation. The Calusa and
Their Legacy educates readers about the rich prehistoric,
historic, and contemporary cultures of south Florida, while at
the same time invoking critical consideration of how precious
and vital south Florida's estuarine environments were in the
past and continue to be today to all Floridians.
Throughout the text MacMahon and Marquardt carefully
explain how they have gathered the information presented in
the book, namely through archaeology and historical research.
The practice and methods of archaeology are presented in
order to orient the reader with the methods archaeologists use
to gather data, and the supplementary tools and types of
analyses they use to interpret it (e.g., zooarchaeology).
Archaeological terms (e.g., midden) are well defined, and
original historical references are included in the text, provid-
ing clear examples of how archaeology and history are
complimentary in understanding south Florida's past. The
authors also present modern Native American testimonies
about traditional cultural practices as a cross cultural basis for
comparison with the archaeological and historical records.
Chapter 12 is dedicated to detailed discussions of archaeologi-
cal artifacts and the important archaeological sites of Key
Marco, Pineland, and Fort Center, which MacMahon and
Marquardt argue all have contemporaneous Calusa compo-
nents. In addition, several artistic renditions of Calusa activi-
ties are included in order to provide a visual guide to the
reader as well as numerous high quality photographs of


2005 VOL. 59(2)


estuarine wildlife and plants, artifacts, and contemporary
Seminole and Miccosukee peoples living in Florida.
The ecology of south Florida and the Calusa domain is
discussed in Chapters one through eight. Chapter one intro-
duces the Calusa of south Florida and the methods and data
archaeologists use to understand Calusa culture and life. More
specifically, chapters two through seven present an overview
of both the micro- and macro-life forms and ecological
processes that compose the estuarine environments on which
the Calusa culture thrived. Chapter eight synthesizes several
sources of environmental degradation threatening south
Florida's estuaries today and calls for an "environmental
ethic" promoting respect for the estuaries and the urgent need
to protect their delicate ecological systems that directly impact
the quality of life of all Florida residents.
The culture and legacy of the Calusa are explored in
Chapters nine through 12. MacMahon and Marquardt present
a comprehensive discussion of Florida's 6,000 year old Gulf
coast fishing heritage through tying together archaeological,
historical, and ethnographic data. Archaeological evidence
suggests that the culture and people identified as the Calusa
lived on the southwestern Florida coast beginning about 1,500
years ago and expanded their domain to encompass the entire
southern half ofFlorida. The authors explain that archaeologi-
cal evidence of fishing peoples that preceded the Calusa may
have been Calusa ancestors, but that due to limited evidence
such claims are merely speculative. Therefore, the archaeologi-
cal remains and European historical documentation of the
Calusa provide the most extensive and concrete evidence of
Florida's fishing heritage.
Chapter nine specifically discusses fishing technologies used
by the Calusa, illuminating the dietary importance of the
estuary to the subsistence requirements of Calusa life. The
Calusa relied most heavily on estuarine food sources and
practiced small scale horticulture. They utilized an array of
ecologically sensitive procurement technologies, with an
emphasis on various sized nets made of a range of plant fibers.
In addition, the bones and teeth of animal food resources were
utilized as tools, such as shark teeth plant graters.
In Chapters 10, 11, and 12 MacMahon and Marquardt
demonstrate the importance of the estuary in the development
of Calusa politics and ideology. The legitimization of Calusa
political power and leadership was directly tied to the bounty
of the land and water sources. The estuarine environment of
the Calusa domain was embedded with ideological and ritual
significance. The very animals that the Calusa captured, ate,
and shaped artifacts out of represented deceased ancestors. The
Calusa believed that when a person died their soul entered an
animal and when that animal was killed the soul of the
deceased entered into a lesser animal and the cycle was
repeated until the soul no longer existed. MacMahon and
Marquardt carefully explain that the physical and ideological
realms of Calusa life were inseparable, and that in order to
understand Calusa culture all parts-including political
leadership, economy (e.g., fishing), and spirituality-must be
considered concomitantly.
Chapters 13 through 17 discuss contemporary south Florida

peoples and estuarine resource exploitation. By the start of the
1700s the Calusa and their way of life finally came to an end
at the hands of European warfare, disease, and forced aban-
donment of the estuary. However, fishing along Florida's coast
and human-estuarine relationships continued to be an integral
part of life. During following centuries new fishing groups and
cultures formed, including the combinations of remnant
Calusa and their descendents, Spanish-Cuban fishermen, as
well as other Native American fishing cultures located in
Florida (namely Seminole and Miccosukee). Although the
cultural composition of Florida's fishing heritage changed
through time, the fishing technology did not significantly
change until the mid 1900s with the onset of increased large-
scale commercial fishing and accompanying technological
The future of Florida's estuaries, coastlines, and resource
exploitation face several challenges as Florida's population
continues to grow exponentially and coastline development
increases, both causing irreversible damage to estuarine
systems. For generations, the Calusa were able to exploit the
south Florida estuarine and coastal environments in a man-
aged and ecologically sustainable fashion, never resulting in
the overexploitation or depletion of resources. Through the
study of the Calusa, MacMahon and Marquardt argue that
contemporary Floridians can gain an improved perspective of
just how fundamental the environment is to sustaining our way
of life; hopefully resulting in more ecologically responsible
behavior and land development.
This book eloquently demonstrates the rich cultural diversity
that has been characteristic of south Florida peoples through-
out (pre)history. At the heart of this diversity are Florida's
unique estuarine and coastal environments. MacMahon and
Marquardt are able to bridge hundreds of years of resource
exploitation and the cultural importance of human-environ-
ment relationships on both micro- and macrolevels of ecology
and human cultural complexity, demonstratingthe inextricable
connection between human beings and their environments.
Moreover, they tie the archaeology of south Florida into bigger
issues of human-environment relationships faced by people
throughout the world today, such as conservation and land
development, thus demonstrating how academic archaeologi-
cal interests can have relevance to contemporary life and reach
popular audiences. The Calusa and Their Legacy is a notewor-
thy addendum to an already famous museum exhibit, showing
its worth as an educational tool to readers of all different levels
of interest.


The Seminole Wars:America'sLongestIndian Conflict. John
Missall and Mary Lou Missal. 2004. University Press of
Florida, Gainesville. xv+280 pages, figures, tables, biblio.,
index. $29.95 (paper)

Department of Sociology and Anthropology, University of
North Florida, 4567 St. Johns BluffRoad, South
Jacksonville, Florida 32224-2659

Anthropologists have long appreciated the product of the
historian as a tool to better understand the peoples we are
called to study. Of interest to the Florida anthropologist is our
state's Native American population. Descendants of the
Lower Creeks, the Florida Seminole are today enjoying
something of a cultural renaissance. It is not without irony
that this revitalization is, in part, financed by profits the
Seminoles collect from the "white man's" losses at the tribe's
gaming casinos. John and Mary Lou Missall's book The
Seminole Wars: America's Longest Indian Conflict traces the
United States government's attempts to relocate the tribe from
north-central Florida to the Indian Territory during the early
nineteenth century. Although some of the Seminoles acqui-
esced to government demands and moved to Oklahoma, those
who remained were pushed into hiding among the swamps of
the Florida Everglades by the tide of American troops that
swept down the peninsula. Their descendants have adopted
the refuge of the desolate swampland and are beginning to
thrive there even today.
The book is structured by the events of the three Seminole
wars and begins with the origins of the conflict. As a Spanish
possession, Florida and the Seminoles who resided within its
borders were a welcome refuge for black slaves hoping to
escape the oppression of their Southern masters. It was under
the pretext of returning runaways to their American owners
that led, then Major General, Andrew Jackson to order the
invasion of Florida and, in fact, the First Seminole War had
little to do with the Seminoles. Instead, it was a successful
attempt to put down a runaway slave insurrection at a Florida
fort that had little Seminole involvement. Later, the objectives
of U.S. involvement in Florida were expanded to recapture
runaway slaves and rid Florida of its indigenous population.
Tensions between the two increased as some ofFlorida's white
settlers filed reports of cattle theft by the Seminole, who
themselves believed the ranchers had stolen cattle from them.
In attempts to regain "stolen" cattle, casualties were inflicted
on both sides. The American press branded the Seminoles as
savages for the brutal murder of white settlers. No such outlet
existed for the Seminole whose losses included not only men
but also their land.
The actual inducement for the wars was the country's
growing need for land due to an expanding eastern population.
There was also a desire to wrest Florida's attractive ranch land
from Spanish control and close the door to a southern land
base for any would-be foreign invaders. U.S. troops dutifully
heeded the call of yet another of our nation's early wars, only
to be met with an opponent whose tactics and determination

tested their mettle at every turn. Having forgotten the zeal
with which their forefathers had defended their land and the
right to rule over it freely, the U.S. Army attempted to take on
their native opponents with standardized battle formations.
The Seminoles, whose weapons and munitions came by way of
rogue British and French opportunists, took to the trees and
marshes with which they were so familiar and engaged the
U.S. troops using ambush and "guerilla" tactics, handily
defeating the Americans at nearly every encounter. However,
the Seminoles had their own problems: they lacked a coherent
strategy, and sometimes even a viable leader; they were
constantly short on supplies; and they had to tend their fields
in the summer seasons to provide for their families. The U.S.
troops were so ill-equipped to fight in the harsh Florida
summers that an informal truce was required by both sides
each summer.
The three Seminole wars correctly could be classed as a
single, albeit lengthy, conflict. They were prosecuted to their
final, conciliatory peace partly because of the government's
desire for conclusion to Jackson's Indian Removal efforts and
partly to preserve the battered honor of the nation's military.
The book well expounds on the government's political and
military actions to achieve Jackson's goals, tracing troop
movements, major battles, and the efforts at the negotiating
table. The authors provide biographical sketches of the leaders
of the American troops, including such notables in American
history as Generals Duncan Clinch, Winfield Scott, Zachary
Taylor, and Thomas Jesup, among others.
While the authors provide a rather detailed accounting of
U. S activities, the book adds almost no information that would
elucidate the Seminoles' organization, political ideology, or
strategies for resistance beyond the accounts of their ad hoc
leaders' efforts at the negotiating table. Thus, while not
unfairly so, the book becomes decidedly imbalanced. As the
authors themselves note, they did not set out to tell the
Seminole side of the war, citing the lack of documentary
evidence and their opinion that the Seminole should tell their
own story (p. xviii). While citing only written and verifiable
accounts may be in keeping with the craft of the historian,
anthropologists are also interested in how a pre-literate,
apparently loosely organized tribe of warriors could mount so
effective a resistance against a highly trained, experienced,
and well-equipped U.S. Army under the leadership of the
most capable men of its day. Additionally, no attempt is made
to explain the Seminoles' current condition or the impact these
wars have had on the tribe.
Of concern to the critical reader is the authors' frequent use
of unsupported generalizations. Phrases meant to capture the
American or Seminole ethos such as: "[General] Gaines, like
most Americans, harbored a special hatred for the British." (p.
33). At the time of his arrival in Florida, Gaines was recover-
ing from near fatal wounds he had received from the British at
the Battle of Erie in 1814. Gaines may well have hated the
British, but the authors' claim that most Americans shared his
opinion is without warrant and suggests a laxity of scholarly
rigor. Comments of this type are prevalent, extraneous to the
narrative, and give the critical reader reason to pause.


2005 VOL. 59(2)


The authors state their goal for the book in the Preface: to
tell, in a single volume, the events leading up to the Seminole
wars, the events of their progress, and of their conclusion. As
such, they admit to adding little new information to the
historical understanding of the wars. Indeed, they encourage
readers with a particular interest to consult the volumes in
their bibliography. For its intended, general audience, the
book well introduces the bloody conflict that embroiled
Florida's early modern history. It is regretful that the authors
confine their effort by omitting the Seminoles' story. For the
anthropologist, the book makes clear the need for that story to
be told.

Kolomoki: Settlement, Ceremony, and Status in the Deep
South, A.D. 350 to 750. Thomas J. Pluckhahn. 2003. Univer-
sity of Alabama Press, Tuscaloosa. xv+264 pages, figures,
tables, notes, biblio., index. $70.00 (cloth), $34.95 (paper)

Environmental Services, Inc., 7220Financial Way, Suite 100,
Jacksonville, FL 32256
E-mail: ghendryx@esinc. cc

The Kolomoki site has long been an enigma of southeastern
archaeology. Nestled in the Lower Chattahoochee River
Valley in southwestern Georgia, this site contains at least nine
earthen mounds and a ceramic complex dominated by Middle
and Late Woodland Swift Creek and Weeden Island pottery.
Yet, despite the occurrence of Woodland ceramics, William
Sears' work from the 1940s and 1950s resulted in a Mississip-
pian classification of the site. Influenced by the cultural
historical paradigm of the time, Sears was more apt to question
the temporal placement of Swift Creek and Weeden Island
pottery than to accept that a site with platform mounds and
elaborate mortuary accompaniments could date to the Wood-
land period. Sears' argument was accepted by many, yet
others, such as Gordon Willey, Joseph Caldwell, John Goggin,
and Stephen Williams, were unconvinced. Regardless, major
opposition was limited in published form, largely due to the
fact that the archaeologists of the time were a less populace
and tightly-knit group able to discuss the chronology in person
without publicly belittling the work of a reputable colleague
(Knight and Schnell 2004:8-9). Thus, through time, the site
generally became accepted as Woodland period, and even
Sears recanted his Mississippian period interpretation in his
1992 publication Mea Culpa.
Resolution to Kolomoki's temporal classification is but the
first step in understanding its role in the development of the
southeastern cultural sequence. In this book, Thomas
Pluckhahn synthesizes past research and presents data from
his recent excavations to provide a framework for viewing the
rise and fall of this very important site. The book is divided
into seven chapters. The first lays the groundwork for a half
century of site misconception, within which Pluckhahn offers
a "middle range" classification for the site. On the surface, the
presumed labor investments needed for the construction of the
earthworks, coupled with the presence of exotic funerary

items, point to a hierarchical chiefdom society, a level of
sociopolitical complexity seemingly incongruous with the
traditional classification of Woodland societies as egalitarian
and tribal. Pluckhahn's middle range interpretation is further
contingent upon a better understanding of the role of ceremony
and trade in Woodland societies, a topic also addressed in the
opening chapter.
In the second chapter Pluckhahn establishes the site's
chronology and temporal placement. He sequenced the site
into four 100-year phases from A.D. 350 to 750. The phases
were labeled Kolomoki I-IV and were recognized largely on
the basis of ceramic occurrence and frequency, with some
assistance from radiocarbon dating. This discussion is
thorough and contains useful tables that synthesize the ceramic
attributes that define the phases. Pluckhahn's chronology is
further supported through comparison with the nearby,
stratified Fairchild's Landing and Hare's Landing sites. This
chapter also offers an explanation for the location of this
paramount settlement from an environmental and cultural
perspective. Specifically, the site is relatively isolated, yet
somewhat equidistant between two contemporaneous site
clusters, including one to the north and one to the south.
Moreover, these two site clusters are approximately 70 km
from Kolomoki, and this great distance suggests that
Kolomoki did not function as an administrative center, but
perhaps as a location for gathering and ceremony, which begs
the question of whether the site was occupied year-round or on
a seasonal basis.
In the third chapter, Pluckhahn provides a chronological
narrative of prior work at the site, and utilizes the earlier
information to identify the sequence of mound construction.
Interestingly, many of the earlier accounts also describe
additional mounds and an earthen enclosure no longer evident
at the site. The mound synthesis is thorough and supported by
photographs of excavations, including one of an amazing
ceramic cache in Mound D.
In the fourth chapter, Pluckhahn reports the results of
systematic shovel testing he implemented to define site limits
and activity areas. Testing results revealed spatial discontinu-
ities that prompted Pluckhahn to define twelve activity areas
based on artifact ubiquity and density that enabled an analysis
of the spatial distribution of the components.
The fifth chapter presents the results of geophysical testing
and limited excavation in off-mound areas. The testing
program was aimed at strengthening Pluckhahn's established
site chronology and providing data on the internal structure of
the site. Ground penetrating radar was employed to identify
feature concentrations and limit the amount of necessary
fieldwork for the sake of site conservation. Testing results
from each area are synthesized and many inferences are made.
Most notably, Pluckhahn recognized that the site had sustained
year-round and intensive habitation, based on feature type and
horticultural evidence, which refutes the site's interpretation
as being one of seasonal use.
Chapter 6 details the results of block excavations. In an
attempt to expose and interpret a domestic structure,
Pluckhahn opened three block units in a portion of the site he


termed "the Northwest Area." The bulk of this work focused
on Block Unit A, where a semi-subterranean house, with a
projecting entrance and interior fire pit, was revealed, thus
providing one of the best examples of Middle Woodland
architecture in the southeastern United States.
The seventh and final chapter offers a holistic synthesis of
the information. This chapter alone is worth the price of the
book. The author does a great job presenting the information
and building a case for his chronology and interpretation of the
site's function and role. The chapter is arranged chronologi-
cally and provides an in-depth discussion of each of the four
subperiods. Pluckhahn revisits the topic of mound construc-
tion, whereby he identifies construction periods, estimates
labor investments for mound construction, and offers func-
tional interpretations. He evaluates intrasite settlement shifts
and provides population estimates for each of the periods.
This chapter also considers intersite comparisons, drawing
from such sites as McKeithen and Bernath Place in Florida.
Numerous other topics are also discussed, such as the organi-
zation of ceremony, funerary variability, social stratification,
and the site's general role as it evolved through time.
Pluckhahn offers numerous caveats and recognizes areas
where the evidence is significantly more tenuous, yet presents
a fine model that lays the groundwork for refinement through
further work at the site.
In sum, Pluckhahn provides a comprehensive synthesis of
the site. His research is thorough and inclusive, and the book
is well organized in its presentation. The book is recom-
mended for a wide audience; in addition to the site overview,
it touches on a variety of topics, including Woodland settle-
ment, studies in middle range theory, and ceramic seriation
and classification, to name a few. It is highly illustrative in
terms of tabular presentation and informative site maps;
although ceramic artifact illustrations would have compli-
mented the earlier chapters. The book is available through the
University of AlabamaPress and provides a major contribution
to southeastern prehistoric archaeology. Kolomoki is long
overdue for a synthesis of this magnitude, and Thomas
Pluckhahn should be commended for his efforts. Whereas
much of the book contains an enormous amount of data
necessary for building Pluckhahn's case, many casual readers
may want to skip to the final chapter, and digest the supportive
information of the earlier chapters on an as-needed basis.

References Cited

Knight, Vernon James, and Frank T. Schnell
2004 Silence over Kolomoki: A Curious Episode in the
History of Southeastern Archaeology. Southeastern
Archaeology 23(1):1-11.

Sears, William H.
1992 Mea Culpa. Southeastern Archaeology 11(1):66-71.


2005 VOL. 59(2)


About the Authors:

Keith H. Ashley is an Assistant Professor of Liberal Arts (Anthropology) at Savannah College of Art and Design. His current
research centers on the native history of northeastern Florida, with an emphasis on political economy and historical process.

Greg Hendryx is a senior archaeologist with Environmental Services, Inc. in Jacksonville, Florida. He has worked as an
archaeologist throughout the southeastern United States since 1990 and his current research interests lie in coastal Georgia
and northeastern Florida.

Michelle J. LeFebvre is a graduate student in the Department of Anthropology at the University of Florida studying
zooarchaeology with a focus on prehistoric human-environment relationships in the Caribbean. She is currently the project
zooarchaeologist for the Carriacou Archaeological Field Project.

Justin Martin is a senior anthropology student at the University of North Florida. His current research is interested in modern
place-making and structural models for Victor Turner's communitas using graph theory and sociograms.

David Thulman is a doctoral candidate in Anthropology at Florida State University, where he is working on Paleoindian
social structures in Florida. He is also an attorney for the Florida Department of Environmental Protection.

Nancy White is a professor of anthropology at the University of South Florida, Tampa, who has been investigating the
prehistoric and historic past of northwest Florida's Apalachicola Valley region for many years.





Editor's Page

The Ephemeral Cape St. George Shipwreck on the Northern Gulf Coast,
Franklin County, Florida. Nancy Marie White

Colorinda and its Place in Northeastern Florida History. Keith H. Ashley

Archaeological Testing of Colorinda Shell Middens at the Cedar Point Site (8DU81). Keith H. Ashley


Cobb: Stone Tool Traditions in the Contact Era. David K. Thulman

MacMahon and Marquardt: The Calusa and Their Legacy:
South Florida People and Their Environments. Michelle J. LeFebvre

Missall and Missall: The Seminole Wars: America's Longest Indian Conflict. Justin Martin

Pluckhahn: Kolomoki: Settlement, Ceremony, and Status in the Deep South,
A.D. 350 to 750. Greg S. Hendryx

Copyright 2006 by the
ISSN 0015-3893


Full Text