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


m --- ----------------



MARCH 2003


* i,-~ '- -~
-. iz`


:63a 7

THE FLORIDAANTHROPOLOGIST is publishedby 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 distributedby the Society during
the 12 months of their membership year. Annual dues are as follows: individual $25, family $35, institutional $25, sustaining $35 or
more, patron $100 or more, and life $500. 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. Thejournal is formatted using Word Perfect for Windows.
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: Phyllis Kolianos,880 Seminole Blvd., Tarpon Springs, FL 34689 (
Second Vice President: Linda Allred, 5460 43rd Ave. N., St. Petersburg, FL 33709 (
Corresponding Secretary: Gordon Davis, 2200 N. Oleander Dr., Avon Park, FL 33825 (
Membership Secretary: Terry Simpson, 9907 High Meadow Ave., Thonotosassa, FL 33592 (
Treasurer and RegisteredAgent: Mary Beth Fitts, 1405 Colonial Dr., Tallahassee, FL 32303 (
Directors atLarge: Gypsy Graves, 4561 SW 24f' Ave., Ft. Lauderdale, FL 33312 (One Year); Dot Moore, P.O. Box 504, New Smyrna
Beach, FL 32170( (Two Years); Gloria Fike, 2815 SE St. Lucie Blvd., Stuart, FL 34997 (Three Years)
Immediate Past President: Jack Thompson, 576 Retreat Drive, Apt. 202, Naples, FL 34110 (
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:. Keith Ashley, P.O. Box 350225, Jacksonville, FL 32225-0225 (
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
Bulk Mail: Capital City Mailing, 4013 Woodville Hwy, Tallahassee, FL 32311
Backlssue Sales: Graves Museum of Archaeology and Natural History, 481 S. Federal Highway, Dania, FL 33004, (954) 925-7770,
FAX (954) 925-7064


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.



Volume 56 Number 1
March 2003


Editor's Page.


New AMS Dates on Orange Fiber-Tempered Pottery from the Middle St. Johns Valley
and Their Implications for Culture History in Northeast Florida. Kenneth E. Sassaman

Testing Partially Submerged Shell Middens in the Apalachicola Estuarine Wetlands,
Franklin County, Florida. Nancy Marie White

The Mystery River Point Site, Englewood, Sarasota County, Florida. Traci Ardren,
Bill Burger, and Brian Keith Sullivan


Patricia Crandon Randell. George M. Luer


Austin, Hoffman, and Ballo: Thinking About Significance: Papers and Proceedings,
Florida Archaeological Council, Inc. Professional Development Workshop, St. Augustine,
Florida. Louis D. Tesar

About the Authors

Cover: Orange Incised and Orange Plain sherds from Tick Island, Florida. From Figure 6 in The Tick Island Site, St. Johns
River, Florida, by Otto L. Jahn and Ripley P. Bullen. Florida Anthropological Society Publication No. 10 (December 1978).

Copyright Notice: Authors retain all copyrights to materials published in this journal, other materials are copyrighted
by the Florida Anthropological Society.

Published by the
ISSN 0015-3893


This issue of The Florida Anthropologist has three varied
articles dealing with the St. Johns River, Northwest Gulf
Coast, and Central Gulf Coast areas.
Ken Sassaman's article presents a very interesting
reconsideration ofRipley Bullen's Orange Period chronology,
which was constructed in the early days of radiocarbon dating.
Sassaman obtained AMS dates on fiber-tempered ceramic
sherds using carbonized soot from sherd exteriors. The dates
indicate that, rather than a unilineal evolution of vessels,
Orange Period ceramic pastes, forms and decorations overlap
considerably and do not represent temporal changes.
Sassaman discusses other explanations for the differences
found in fiber-tempered pottery. Hopefully similar research
will be applied to the Florida Transitional and St. Johns
chronologies as well.
Nancy White's article presents the results of 1993
fieldschool excavations at the Sam's Cutoff and Van Horn
Creek shell mounds in the Apalachicola River delta. Changes
in water table since deposition inundated the lower levels of
the site, requiring White and her students to use large pumps
and well points to dewater excavation units. It was interesting

to read about the logistics ofdewatering the remote sites, since
I worked on a project at Lake Monroe in 1993 that required
similar technology at a more accessible site (see December
1994 issue of The FloridaAnthropologist). It was funto recall
some of the same difficulties that we faced with the pumps and
well points, which have been used only rarely in Florida
excavations. White also presents a useful discussion of the
somewhat enigmatic late Archaic Elliott's Point complex,
which should help others understand this interesting and often
misunderstood culture.
The final article, by Traci Ardren, Bill Burger, and Brian
Sullivan, presents a concise report on their 1985, 1987, and
1989 New College investigations at the Mystery River Point
site in Sarasota County. The report adds to the data on the
Manasota Culture, and compliments other recent FAS
publications, like George Luer's 1999 Martime Archaeology
of Lemon Bay, Florida (Florida Anthropological Society
Publication No. 14).
This issue also contains an obituary of Pat Randell, who
was instrumental in the preservation of part of the Pineland
site in Lee County, and a book review by Louis Tesar.

March, 2003

MARCH 2003



VOL. 56(1)



Department ofAnthropology, University ofFlorida, Gainesville, FL 32611

The St. Johns region of northeastern Florida is home to
some of the oldest pottery in the Southeast, Orange fiber-
tempered pottery, dating to at least 4000 radiocarbon years
before present (rcybp). A cultural-historical sequence for
Orange pottery was established nearly 50 years ago by Ripley
Bullen (1954; also see Bullen 1955, 1972). In this sequence,
Orange pottery was made and used over a period of 1000 years,
first in a plain form, then as a decorated ware. In the final
centuries of its existence, Orange pottery gave way to the
sponge spiculate wares of the St. Johns pottery tradition. This
unilineal sequence-plain fiber-tempered pottery, followed by
decoratedfiber-tempered pottery, followed by sponge-spiculate
pottery-is being challenged by new radiocarbon dates and
petrographic analyses. Here I report the results of seven new
AMS dates on soot that strongly support the inference that all
three major horizons of Bullen's pottery sequence are coeval,
not sequential. The chief implication.of these new findings is
that variation in pottery long believed to be developmental
must now be interpreted as synchronic manifestations of
ethnic, sociopolitical, or functional factors.

The Bullen Sequence

In papers published in 1954 and 1955, Bullen divided the
1000-year-long Orange period into five subperiods. His 1972
chapter inthe monograph series oftheFlorida Anthropological
Society refined this sequence and presented a synthesis of the
data used to construct it. A brief review of Bullen's sequence
and its supporting data follow (also see Milanich 1994:94).
The emphasis here is on sites in the middle St. Johns Valley
(Figure 1).
The Orange 1 subperiod is characterized by the use of flat-
based, rectangular containers that were tempered with fiber,
had plain surfaces, thin walls, and simple, rounded lips.
Excavations at the Bluffton site in Volusia County provided
Bullen (1955) with a sizeable assemblage of Orange 1 pottery.
The 2-foot-thick Viviparus midden containing plain fiber-
tempered pottery was underlain by nearly 30 feet ofpreceramic
shell midden. A shell celt from the base of the plain pottery
stratum returned an uncalibrated (and probably uncorrected1)
radiocarbon age estimate of 3660 1 110 rcybp, which Bullen
(1972:11) suggested may be "too recent." Based on radiocar-
bon dates from the Palmer site on the Gulf Coast (Bullen 1961,
1972:13), Bullen established the onset of the Orange 1

subperiod at 4000 rcybp and its termination at 3650 rcybp. A
two-foot-thick stratum over the Orange 1 stratum at Bluffton
contained incised fiber-tempered sherds, but it was somewhat
disturbed by historic land use and was never dated radiometri-
The subsequent Orange 2 subperiod marked the appearance
of incised decorations that included concentric vertical
diamonds with horizontal lines on vessels of similar form as
the Orange 1 subperiod. Plain pottery persisted in abundance,
and there were occasional uses of the so-called Tick Island
style of incised spirals with background punctations. Exam-
ples of these latter sherds were recovered from the upper
stratum ofBluffion, as well as the namesake Tick Island site
(Aten 1999; Jahn and Bullen 1978). Radiocarbon assays for
Orange 2 were never obtained, although Bullen extrapolated a
date range of 3650-3450 rcybp based on assays from the
preceding and succeeding subperiods.
During Orange 3 times vessels included large, straight-
walled and round-mouthed pots with flat bottoms and fewer
shallow rectangular vessels. Decorations on Orange 3 pots
involve incised straight lines, some set obliquely, and a limited
number ofpunctations or ticks. Three sites with pottery typical
of Orange 3 were noted by Bullen (1972:14): Summer Haven
on the Atlantic Intracoastal Waterway south of St. Augustine
(Bullen and Bullen 1961; Janus Research 1995); Cotton, also
on the Intracoastal Waterway, just north of Daytona Beach
(Griffin and Smith 1954); and South Indian Field, near the
headwaters of the St. Johns River (Ferguson 1951). Marine
shell from Summer Haven was dated at 3330 200 rcybp,
which Bullen (1972:15) suggested "dated the apogee in Florida
fiber-tempered ceramic development." Bullen asserted a time
range of 3450-3250 rcybp for Orange 3.
Accordingto Bullen, pastes with mixtures of sand and fiber
appeared in Orange 4 times. Coiling also appeared, and
decorations were limited to simple incised motifs. Sunday
Bluff on the Oklawaha River (Bullen 1969) provided what
Bullen (1972:17) considered "the first 'pure' Orange 4
subperiod deposits so far reported." Although no radiocarbon
assays were obtained for samples from Sunday Bluff, Bullen
suspected that its pottery assemblage represented the terminal
fiber-tempered phase because decorations and vessel shapes
closely resembled the succeeding St. Johns wares (Bullen
1972:17). He extrapolated the close of the fiber-tempered
pottery period at ca. 3000 rcybp based on radiometric dates


MARCH 2003


VOL. 56(1)


Figure 1. Map of the St. Johns River area of northeastern Florida, with locations of sites mentioned in the text.

from the Zabski site on Merritt Island (Atkins and MacMahon
1967), the Tucker site in west Florida (Phelps 1966), and the
Refuge site in Georgia (Williams 1977:329), all with assem-
blages Bullen considered immediately successive to fiber-
tempered pottery. The Zabski site, with St. Johns pottery and
a radiocarbon assay of 2910 80 rcybp, was particularly
influential in Bullen's formulation of the Florida Transitional

Period. He recognized, however, that St. Johns pottery began
to be made and used in northeast Florida before the close ofthe
Orange Period (Bullen 1972:18), during the Orange 4
subperiod, which he dated to 3250-3000 rcybp.
The final Orange subperiod, Orange 5, has had a transient
existence in the culture-historical sequence of northeastern
Florida. In some accounts, Orange 5 is supplantedbyBullen's

2003 VOL. 56(1)



Florida Transitional Period (e.g., Bullen 1972; Milanich and
Fairbanks 1980:156); in others, it stands as the closing
subperiod ofthe Orange sequence (e.g., Bullen 1955; Milanich
1994:94). Either way, the period is generally described as the
time when chalky St. John wares completely replaced fiber-
tempered pottery, and additional ceramic types, as well as
soapstone vessels, appeared (Bullen et al. 1978:22-26). Its
presumed dating is approximately 3000-2500 rcybp. It is
followed in northeastern Florida by the St. Johns sequence,
which continued through to European contact (Milanich
In sum, the Bullen sequence can be simplified as a tripar-
tite scheme consisting of three major horizons: plain fiber
tempered pottery, followed by incised fiber-tempered pottery,
followed by sponge-spiculate St. Johns wares. Although
Bullen recognized overlap amongthese major horizons, as well
as continuity in plain surface treatments throughout, his
formulation was basically unilineal, as were virtually all
cultural-historical sequences developed during the mid-
twentieth century. Apart from the general stratigraphic
superiority of St. Johns wares over Orange wares across
northeastern Florida, no sites encapsulated the entire Orange
sequence and thus Bullen pieced together the chronology from
sites spanning a large portion of Florida. As we have seen,
very few radiometric dates were available to test the strati-
graphic data independently. To be fair, it shouldbe recognized
that Bullen made great strides considering that so many of the
key sites were victims of shell mining and other destructive
forces, and that radiocarbon dating was in a nascent stage of
development. Unfortunately, only a few dates from secure
contexts have been published since Bullen's time. However, in
the last few years several anomalies have risen with new
excavations and laboratory analyses that seriously challenge
Bullen's Orange sequence.

Challenges to the Bullen Sequence

Three major challenges to the Bullen sequence have
surfaced in the past decade. The most significant perhaps is
the 1992 data-recovery project at Summer Haven (8SJ46), one
of the defining sites for Bullen's Orange 3 subperiod. Excava-
tions in advance of a bridge replacement in 1992 by Janus
Research (1995) resulted in the recovery of over 6500 fiber-
tempered sherds. About 1800 of the sherds were classified as
Orange Incised. An analysis of decorations by Mitchell (1993)
revealed a greater variety of design motifs than identified by
Bullen and Bullen (1961), but the assemblage was consistent
with the earlier findings in terms of vessel form and paste.
Deviating significantly from the previous work, however, was
the radiometric dating. Janus Research submittedfive samples
of oyster shell for radiometric dating and received
(uncalibrated) assays ranging from 4000 60 to 3840 60
rcybp after corrections for "C/'2C fractionation (Janus Re-
search 1995:38), age estimates that are four to six centuries
earlier than Bullen's Orange 3 subperiod. All samples were
from contexts with good associations with incised Orange
pottery. The discrepancy between these assays and the 3330

200 rcybp age reported by Bullen may very well be due to
fractionation. Adding 400 years to Bullen's date (3730 200
rcybp) to adjust for fractionation would bring it more in line
with the Janus dates.
The results of ongoing fieldwork near the mouth of the St.
Johns River constitute a second major anomaly in Bullen's
Orange sequence. As summarized by Saunders (n.d.), a
growingbody of evidence suggests thatassemblages dominated
by plain fiber-tempered pottery are coeval with assemblages
containing substantial numbers of incised sherds. Testing at
the 250-m-diameter Rollins Shell Ring has so far produced
over 8000 fiber-tempered sherds, 45 percent of which are
decorated (Saunders n.d.). Four radiocarbon assays on oyster
shell (corrected) and one onbulk carbon returned age estimates
ranging from 3760 60 to 3600 60 rcybp. The nearby
Ribault Clubhouse site (8DU76) produced a similar (corrected)
assay of oyster shell of 3700 70 rcybp, but in association with
plain fiber-tempered pottery (Johnson 2000). The Devil's
Walkingstick site (9CAM177), 31 km north of Rollins, roughly
duplicates the evidence from Ribault Clubhouse with an age
estimate for plain fiber-tempered pottery of 3600 100 rcybp
(Saunders 1985:161).
Finally, recent petrographic analysis of Orange pottery by
Cordell (n.d.) shows that a substantial fraction of the incised
Orange pottery from the middle St. Johns Valley includes
sponge spicules in the paste. In a sample of 50 fiber-tempered
sherds from seven sites in northeastern Florida, Cordell found
that 18 sherds (36 percent) contained sponge spicules in the
paste. The frequency of spiculate paste is especially high
among incised fiber-tempered sherds from Tick Island.
Macroscopic inspection of incised sherds from other middle St.
Johns Valley sites (e.g., Mouth of Silver Glen Run, Mosquito
Hammock) by Jim Mallard (n.d.) bolsters Cordell's findings.
Notably, spiculate pastes are not found in Orange plain wares
from the middle St. Johns Valley (Cordell n.d.), nor are they
typical of any of the fiber-tempered pottery (plain and incised)
from the mouth of the St. Johns River (Saunders n.d.).2
Substantial intraregional variation in paste is thus evident,
little of which, given the radiometric dates reported above and
those reported below, appears to be temporal.

New AMS Dates

Incised Orange sherds from sites in the middle St. Johns
Valley often have soot adhering to exterior surfaces. These
carbon deposits provide good evidence for the use of vessels
directly over fire, a technique of cooking I have argued
appeared only after a long period of indirect-heat cooking
(Sassaman 1993, 2002). Plain Orange sherds are rarely
sooted, thus the Bullen sequence, if it were correct, would tend
to support the idea that pots were first used as containers for
stone boiling, and only later directly over fire.
Apart from inferences about vessel function, soot offers the
opportunity for radiometric dating of sherds. Problems ofpost-
depositional soot formation notwithstanding, carbon deposits
that formed while a vessel was fully intact provide among the
best associations possible between the artifacts we attempt to

Table 1. Data on AMS Assays of Soot Samples from Orange Fiber-Tempered Sherds from Middle St. Johns Valley Sites.

Measured C13/C12 Conventional Intercept Intercept 2-sigma 2-sigma
Lab Number Site-sample* Material C14 Age (BP) Ratio (o/oo) C14 Age (BP) (Cal BC) (Cal BP) (Cal BC) (Cal BP)

Beta 166671 8LA1-12 soot 3690 60 -25.8 3680 60 2040 3990 2210-1900 4160-3850

Beta 166672 8LA1-6 soot 4020 60 -25.2 4020 60 2560 4510 2850-2820 4800-4770
2520 4480 2680-2430 4630-4380
2500 4440

Beta 166673 8LA1-27 soot 4060 40 -24.4 4070 40 2580 4540 2860-2810 4810-4760
2690-2480 4640-4430

Beta 166674 8LA28-2 soot 3600 40 -24.6 3610 40 1950 3900 2120-2100 4060-4050
2040-1880 3990-3830

Beta 166675 8V024-21 soot 3600 40 -23.3 3630 40 1970 3920 2130-2080 4080-4030
2060-1890 4010-3840

Beta 166676 8V024-1 soot 3730 40 -24.4 3740 40 2140 4090 2280-2030 4230-3980

Beta 166677 8V024-252-1 soot 3920 40 -24.1 3930 40 2460 4410 2550-2540 4500-4480
2490-2300 4440-4250

* 8LA1 Mouth of Silver Glen Run; 8LA28 Mosquito Hammock; 8V024 Tick Island


date and the actual samples of organic matter submitted for
radiometric assay. Soot dating has proven very successful in
a variety of applications in the Southeast and beyond
(Sassaman 1997, 1999, 2002; Stephenson 1990). With
accelerator mass spectrometry (AMS), only small samples of
carbon are necessary, so even sherds with minute traces of soot
can provide good age estimates.
Numerous sherds of incised Orange pottery curated at the
Florida Museum of Natural History (FLMNH) have preserved
soot, sometimes in abundance. In analyses conducted by
several University of Florida (UF) graduate students in 1999,
over half of the scores of vessel lots included incised sherds
with preserved soot. Among the more numerous are those
from Tick Island (8VO24) and Mouth of Silver Glen Run
(8LA1), both in the middle St. Johns Valley. The presence of
so many sooted sherds from Tick Island is especially fortunate
because the site was largely destroyed by shell-mining opera-
tions in the middle of the twentieth century. As is often the
case, museum collections such as these provide the only
opportunity to learn something new about destroyed sites; the
potential for radiometric dating of soot is a good case in point.
University of Florida grant funds were acquired by the
author in 2001 to examine all FLMNH collections with Orange
pottery from northeast Florida sites and to obtain AMS assays
on soot from a sample of 10-15 sherds. UF graduate student
Jim Mallard was employed to collect these data and to locate
potential samples while conducting his own analyses for a
graduate thesis ontechnofunctional variation in Orange pottery
(Mallard n.d.). The intent of the dating project was to obtain
samples that crosscut the range of technological and spatial
variation in Orange pottery. As it turned out, assemblages
with abundant sherds with soot were confined largely to the
middle St. Johns Valley and restricted to incised specimens.
The two sites noted above, Tick Island and Mouth of Silver
Glen Run, accounted for well over 90 percent of all sooted
sherds. A third site in close proximity to these two, Mosquito
Hammock (8LA28), added a few other sooted examples.
Sherds from seven vessel lots were selected for an initial
batch of AMS samples, three eachfrom Tick Island and Mouth
of Silver Glen Run, and one from Mosquito Hammock. The
results of the AMS assays are given in Table 1. As these data
show, all seven conventional ages fall in the range of roughly
4100 to 3600 rcybp. None has a standard deviation greater
than 60 years and all have 13C/12C ratios close to the standard
for wood charcoal.
Collectively the AMS assays immediately throw the Bullen
sequence into question for all of the sherds sampled would
traditionally be classified as Orange 3 or later on the basis of
surface treatment, and should thus postdate 3450 rcybp. When
compared to other radiometric dates for Orange pottery in the
region, the implications of these findings are far reaching. As
shown in Figure 2, the seven AMS dates from the middle St.
Johns Valley are fully coeval with five dates from two sites
dominated by plain fiber-tempered pottery and traditionally
classified as Orange 1: Blue Spring Midden B (8VO43;
Sassaman n.d.) and Groves' Orange Midden (McGee and
Wheeler 1994; Russo et al. 1992). Both of these sites include

pre-pottery components predating 4200 rcybp, overlain by
strata with plain fiber-tempered pottery (McGee and Wheeler
1994; Sassaman n.d.).
As anticipated by Saunders (n.d.), radiometric dates for
Atlantic Coastal sites likewise discount the possibility of a
horizon of exclusively plain fiber-tempered pottery. This new
sequence compares favorably with middle St. Johns Valley
dates: both involve a roughly 500 radiocarbon-year period
when both plain and decorated fiber-tempered pottery were in
vogue. None of the dates onfiber-temperedpotteryfrom either
subregion post-date 3390 rcybp at two-sigma, and arguably all
could very well pre-date 3600 rcybp. Thus, roughly the latter
half of the Bullen sequence (excepting the Florida Transitional
Period) may be eliminated. Given new evidence for spiculate
paste among fiber-tempered sherds dating 4000-3600 rcybp
(Cordell n.d.), the vacant portion of the Bullen sequence (i.e.,
ca. 3500-3000 rcybp) will most likely be filled with assem-
blages of St. Johns pottery, at least in the middle St. Johns
Valley area. Further consideration of this and other implica-
tions of the new Orange chronology follow in the closing
section of this report.

Implications for Culture-History

If these new chronometric data hold up to additional
scrutiny, then the four major subperiods of Bullen's sequence
(i.e., Orange 1-4) collapse down into one (Orange 1) and we
are forced to explain technofunctional and stylistic variation in
Orange pottery in terms other than-chronological. Indeed, this
apparent synchronic variation is much more than stylistic. Not
only do surface treatments of sherds from middle St. Johns
Valley sites vary, but plain vessels tend to be thin,
nonspiculate, shallow and rectangular, and never used over
fire, while incised vessels tend to be thick, spiculate, tall and
rounded, and routinely used over fire. In so many ways, the
plain and incised assemblages are remarkably different.
On the coast the differences are not so marked. As
Saunders (n.d.) notes, fiber-tempered vessels from the mouth
of the St. Johns River, whether plain or incised, are rarely
sooted and rarely have spiculate pastes. What is striking about
the coastal assemblages is the spatial separation of assemblages
dominated by plain versus incised wares. Saunders (n.d.)
suggests that the former tend to be found at sheet midden and
nonmidden sites, while the latter cluster at shell rings and
other locations of mounded shell and midden. A similar
spatial dichotomy may hold for the middle St. Johns Valley,
where incised pottery dominates at sites such as Tick Island
and Mouth of Silver Glen Run, and plain pottery dominates
sites such as Blue Spring Midden B, Groves' Orange Midden,
and Bluffton. More information on site type and context is
required to verify this dichotomy, but simply knowing that
these different sorts of sites and assemblages are fully contem-
poraneous opens the possibility for a functional difference in
site types, perhaps something akin to Sears' (1973) sacred-
secular dichotomy for Weeden Island culture. In many cases
the close pairing of shell mounds and nonmound sites, or shell
rings and sheet middens, implies a public-domestic arrange




2003 VOL. 56(1)

Corrected Radiocarbon Years BP

4800 4600 4400 4200 4000 3800 3600 3400 3200

i a

Atlantic Coastal -



a a a a a


-- a -
M ia adlae a a a
-aa a
-a- a
a aa a
a a
a a a a a
a a a a a a a a
S a a a a
a a a a a a
a a a a a,
a a a a a
Mida l a at.aJohasa
a a a a a a a a,

a a a a a aa a
a a a a_ a

a aaa;
a a a a a-a a-a-
a a a a a
-- --a a W aa a
a aaa


Rollins (Saunders n.d.)

Rollins (Saunders n.d.)

Summer Haven (Janus Research 1995)

Rollins (Saunders n.d.)

Rollins (Saunders n.d:)

Rollins (Saunders n.d.)

Summer Haven (Janus Research 1995)

Summer Haven (Janus Research 1995)

Summer Haven (Janus Research 1995)

Summer Haven (Janus Research 1995)

Devil's Walkingstick (Saunders 1985)

Ribault Club House (Johnson 2000)

9CM167 (Calvert et al. 1979)

Mosquito Hammock (this report)

Tick Island (this report)

Mouth of Spring Glen Run (this report)

Tick Island (this report)

Tick Island (this report)

Mouth of Spring Glen Run (this report)

Mouth of Spring Glen Run (this report)

Blue Spring Midden B (Sassaman n.d.)

Blue Spring Midden B (Sassaman n.d.)

Groves' Orange Midden (Russo et al. 1992)

Blue Spring Midden B (Sassaman n.d.)

Groves' Orange Midden (McGee and Wheeler 1994)

Groves' Orange Midden (Russo et al. 1992; McGee and Wheeler 1994)
Groves' Orange Midden (McGee and Wheeler 1994)

Blue Spring Midden B (Sassaman n.d.)

Blue Spring Midden B (Sassaman n.d.)

Groves' Orange Midden (Russo et al. 1992; McGee and Wheeler 1994)

pre-pottery 1 2 3 -4 5 St. Johns

Bullen's Orange Sequence (1-5)

-0- Pre-Pottery --- Predominately Orange Plain -- >25 percent Orange Incised -A- Soot on Orange Incised

Figure 2. Radiometric assays (+/- two standard deviations) for pre-pottery assemblages, assemblages with predominantly
Orange Plain sherds (i.e., less than 10 percent incised sherds), assemblages with greater than 25 percent Orange Incised
sherds, and soot from Orange Incised sherds from site in the middle St. Johns Valley and Atlantic Coast subregions of
northeast Florida.



ment, with burial mounds and related ceremonial/feasting
locations located in close proximity to one or more villages.
Aside from a functional dichotomy, the coexistence of sites
dominated by plain and incised wares of vastly different
technology and use might likewise be explained in ethnic
terms. In this regard it is noteworthy to recall that Bullen
(1972:14) believed that coastal sites with incised fiber-tem-
pered pottery reflected the "expansion from the St. Johns River
valley in Orange 3 times." Now that the coastal assemblages
of incised pottery are demonstrably older than Orange 3,
"expansion" from the St. Johns area, or any other locale for
that matter, is not necessary. As Russo (1996) has argued, and
Bullen knew all along (Bullen and Bullen 1976:12, 49-50),
coastal occupations in Florida began well before pottery
appeared. This greater timeframe for localized cultural
developments increases the potential for archaeologically
recognizable interareal differences in material culture. Clearly
the Orange assemblages at the mouth of the St. Johns River
and those of the middle St. Johns Valley differ in technology
and actual function. It remains to be tested whether some of
the more localized variation in pottery may be a manifestation
of short-term movements of regional populations. For in-
stance, it is certainly possible, although perhaps unlikely, that
the apparent coeval assemblages of plain and incised pottery at
closely spaced sites in the middle St. Johns Valley resulted
from serial use of the subregion by riverine and coastal groups
over short time cycles. Radiometric dating will generally be
insufficient to test these sorts of hypotheses. Clay sourcing
analyses and other methods of establishing artifact provenance
will become increasingly necessary, as will more and better
data on seasons of occupation (e.g., Russo et al. 1992).
Finally, the revised Orange chronology leaves quite a gap
in the existing culture-history of the middle St. Johns Valley
area. Given that sponge spiculate paste was employed much
earlier than ever imagined, it stands to reason that the interval
of 3500-3000 rcybp vacated by Orange will be occupied by
early manifestations of the long-lived St. Johns tradition.
Some evidence from farther south lends preliminary support to
this proposition. Three corrected radiometric dates associated
with spiculate-paste sherds at the Joseph Reed Shell Ring
(8MT13) in the East Okeechobee region range from 3425 75
to 3280 60 rcybp (Russo and Heide 2002:69). Early St.
Johns pottery also was recognized by Widmer (1988) from sites
in southwestern Florida, leading Russo andHeide (2002:80) to
suggest some interaction between the two regions.4 Con-
versely, the lack of fiber-tempered pottery at Joseph Reed is
taken as evidence for limited interaction between the East
Okeechobee and St. Johns regions (Russo and Heide 2002:80).
With a revised middle St. Johns Valley chronology that
terminates the use of fiber-tempered pottery at about 3500
rcybp and pushes the onset of spiculate-paste wares back to
4000 rcybp, these suppositions may require rethinking.
The Florida Museum of Natural History holds numerous
collections of St. Johns pottery with sooted surfaces. The next
stage of this research will target some of these assemblages to
see if the 500-year void left by revised Orange chronology can
indeed be filled with the early assemblages ofthe long-lived St.

Johns tradition.


Together with existing radiometric and petrographic data,
seven AMS assays on soot from incised fiber-tempered sherds
from middle St. Johns Valley sites provide a basis for revising
Bullen's Orange pottery sequence. Soot samples were taken
from sherds in assemblages that traditionally are classified as
Orange 3, but each returned an assay in the time range of
Orange 1 (4000-3650 rcybp). Definedby Bullen as a period of
exclusively plain fiber-tempered pottery, Orange 1 clearly
involved the manufacture and use of incised, as well as plain
wares, in the middle St. Johns Valley area, and apparently the
Atlantic coastal zone of northeastern Florida. Differences in
the form, paste, function, and spatial distribution of plain and
incised pottery must nowbe understood as synchronic phenom-
ena. Notably, the prevalence of spiculate pastes among incised
Orange sherds from the middle St. Johns Valley, as docu-
mented by Cordell (n.d.), suggests that the St. Johns ceramic
tradition has origins extending back to the very beginning of
pottery making in the region (ca. 4000 rcybp). Moreover, fiber
appears to have been abandoned as an additive to clay as early
as 3500 rcybp inthe middle St. Johns Valley. Additional AMS
assays on St. Johns plainwares from this region are expected
to fill the 500-year void left by this revised Orange sequence.


Funding for the AMS dates reported here was provided by the
College of Liberal Arts and Sciences, University of Florida. Collec-
tions research to locate samples of sooted sherds for AMS dating was
conducted by Jim Mallard. My thanks go to Jerry Milanich and Scott
Mitchell of the Florida Museum of Natural History (FLMNH) for
access to the pottery collections and for permission to sample soot for
dating. FLMNH ceramicist Ann Cordell offered technical assistance
with sampling sherds and shared her unpublished paste data on fiber-
tempered pottery from Florida. I am also grateful for access to other
unpublished work by Ann Cordell, Rebecca Saunders, Mike Russo,
and Greg Heide. Comments on an earlier draft of this paper by
George Luer, Mike Russo, Ann Cordell, Ryan Wheeler, and Nancy
White greatly improved its content and organization.


Presumably none of the radiocarbon assays Bullen obtained from
samples of marine shell were corrected for "'C/2C fractionation,
which, according to today's standards, adds 400 years to conventional
radiocarbon age estimates (see Bowman [1990] for discussion of
corrections for fractionation in radiocarbon dating). Throughout the
discussion of the Bullen sequence, radiocarbon assays are given as
Bullen used them. Most of the discrepancies between the Bullen
sequence and the revised sequence reported here can be resolved by
simply adding 400 years to assays Bullen and his colleagues obtained
from samples of marine shell. However, it is not altogether clear
whether or not corrections for fractionation were made to dates Bullen
2 However, Johnson (2000) found that 11 percent of Orange fiber-
tempered sherds from the Ribault Clubhouse site (8DU76), near the
mouth of the St. Johns River, contained spiculate paste (Michael
Russo, personal communication, 2002).




3 Sherds from an estimated 570 fiber-tempered vessels from Tick
Island (8VO24), Bluffton (8VO22), Silver Glen Springs (8MR123),
and Mouth of Silver Glen Run (8LA1), all in the middle St. Johns
Valley subregion, yielded the following percentages for sooted
exterior surfaces: 67.7 percent for incised fiber-tempered vessels (n
= 266) and 3.0 percent for plain fiber-tempered vessels (n = 231). An
additional 73 incised vessel lots with spiculate fiber-tempered paste
bore traces of soot in 64.4 percent of cases (Sassaman 2002:412,
415). It should be noted that determination of spiculate paste in this
sample was simply macroscopic and not as reliable as those generated
from the microscopic techniques used by Cordell (n.d.). Irrespective
of paste, sooted sherds are vastly more common among incised fiber-
tempered wares than on plain fiber-tempered wares in the middle St.
Johns Valley.
4 Luer (1989:251-253) also reports St. Johns pottery from southwest-
ern Florida and notes that it is more common than Widmer (1988)
suggests. It also is worth mentioning that sites lacking pottery in
southern Florida have occasionally yielded radiocarbon assays from
Orange times, suggesting that local populations without pottery
coexisted for centuries alongside populations with pottery (Ryan
Wheeler, personal communication 2002), as they did in the middle
Savannah region of Georgia and South Carolina from ca. 4200-3800
rcybp (Sassaman 2002).

References Cited

Aten, Lawrence E.
1999 Middle Archaic Ceremonialism at Tick Island, Florida:
Ripley P. Bullen's 1961 Excavation at the Harris Creek
Site. The Florida Anthropologist 52:131-200.

Atkins, Steve, and Jeannie MacMahon
1967 The Zabski Site, Merritt Island, Florida. The Florida
Anthropologist 20(3-4):133-145.

Bowman, Sheridan
1990 Radiocarbon Dating. British Museum, London.

Bullen, Ripley P.
1954 Culture Changes during the Fiber-Tempered Period of
Florida. Southern Indian Studies 6:45-48.

1955 Stratigraphic Tests at Bluffton, Volusia County, Florida.
The Florida Anthropologist 8:1-16.

1961 Radiocarbon Dates for Southeastern Fiber-Tempered
Pottery. American Antiquity 27:104-106.

1969 Excavations atSundayBluff, Florida. Contributions of the
Florida State Museum, Social Sciences, No. 15.

1972 The Orange Period of Peninsular Florida. In Fiber-Tem-
pered Pottery in Southeastern United States and Northern
Colombia: Its Origins, Context, and Significance, edited
by R. P. Bullen and J. B. Stoltman, pp. 9-33. Florida
Anthropological Society Publication 6. Gainesville.

Bullen, Ripley P., and Adelaide K. Bullen
1961 The Summer Haven Site, St. Johns County, Florida. The
Florida Anthropologist 14 (1-2):1-15.

1976 The PalmerSite. Florida Anthropological Society Publica-

tion 8. Gainesville.

Bullen, Ripley P., Walter Askew, Lee M. Feder, and Richard L.
1978 The Canton Street Site, St. Petersburg, Florida, edited by
Jerald T. Milanich. Florida Anthropological Society
Publication 9. Gainesville.

Calvert, P. M., D. S. Introne, and J. J. Stipp
1979 University ofMiamiRadiocarbonDatesXIV. Radiocarbon

Cordell, Ann
n.d. Paste Variability and Possible Manufacturing Origins of
Late Archaic Fiber-Tempered Pottery from Selected Sites
in Peninsular Florida. Draft chapter in Early Pottery:
Technology, Style, and Interaction in theLowerSoutheast,
edited by Rebecca Saunders and Christopher Hays.
University of Alabama Press, Tuscaloosa (in preparation).

Ferguson, Vera Masius
1951 Chronology at SouthIndian Field, Florida. Yale University
Publications in Anthropology, No. 45. New Haven.

Griffin, John W., and Hale G. Smith
1954 The Cotton Site: An Archaeological Site of Early Ceramic
Times in Volusia County, Florida. FloridaState University
Studies, No. 16, pp. 27-60. Tallahassee.

Jahn, Otto L., and Ripley P. Bullen
1978 The Tick Island Site, St. Johns River, Florida. Florida
Anthropological Society Publication 10. Gainesville.

Janus Research
1995 Archaeological Investigations at the Summer Haven Site
(8SJ46), an Orange Period and St. Johns Period Midden
Site in St. Johns County, Florida. Environmental Manage-
ment Office, Florida Department of Transportation, Talla-

Johnson, Robert E.
2000 Archaeological Testing and Monitoring of Site 8DU76 for
the Ribault Clubhouse Foundation Repairs Project Duval
County, Florida. Florida Archaeological Services, Inc.,

Luer, George M.
1989 Notes on the Howard Shell Mound and Calusa Island, Lee
County, Florida. The Florida Anthropologist 42(3):249-

Mallard, Norman J.
n.d. Title pending. Unpublished M.A. thesis, Department of
Anthropology, University of Florida, Gainesville (in

McGee, Ray M., and Ryan J. Wheeler
1994 Stratigraphic Excavations at Groves' Orange Midden, Lake -
Monroe, Volusia County, Florida: Methodology and
Results. The Florida Anthropologist 47(2):333-349.

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


2003 VOL. 56(1)


Florida, Gainesville.

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

Mitchell, Scott
1993 AnAnalysis of the Ceramics from the Summer Haven Site.
Report of file, Janus Research, St. Petersburg, Florida.

Phelps, David S.
1966 Early and Late Components at the Tucker Site. The
Florida Anthropologist 19(1):11-38.

Russo, Michael
1996 Southeastern Mid-Holocene Coastal Settlements. In
Archaeology of the Mid-Holocene Southeast, edited by
Kenneth E. Sassaman and David G. Anderson, pp.
177-199. University Press of Florida, Gainesville.

Russo, Michael, and Gregory Heide
2002 The Joseph Reed ShellRing. The Florida Anthropologist

Russo, Michael, Barbara A. Purdy, Lee A. Newsom, and Ray M.
1992 AReinterpretation of Late Archaic Adaptations in Central-
East Florida: Groves' Orange Midden (8-Vo-2601).
Southeastern Archaeology 11:95-108.

Sassaman, Kenneth E.
1993 Early Pottery in the Southeast: Tradition and Innovation
in Cooking Technology. University of Alabama Press,

1997 Refining Soapstone Vessel Chronology in the Southeast.
Early Georgia 25(1):1-20.

1999 A Southeastern Perspective on Soapstone Vessel Technol-
ogy in the Northeast. In The Archaeological Northeast,
edited by M. A. Levine, K. E. Sassaman, and M. S.
Nassaney, pp. 75-95. Bergin and Garvey, Westport,

2002 Woodland Ceramic Beginnings. In The Woodland South-
east, edited by D. G. Anderson, and R. C. Mainfort, Jr., pp.
398-420. University of Alabama Press, Tuscaloosa.

n.d. St. Johns Archaeological Field School 2000-2001: Blue
Spring and Hontoon Island State Parks. Technical Report
No. 4, Laboratory of Southeastern Archaeology, Depart-
ment of Anthropology, University of Florida, Gainesville
(in preparation).

Saunders, Rebecca
1985 The Fiber-Tempered Area. InAboriginal Subsistence and
SettlementArchaeology of the KingsBayLocality, Volume
1: The Kings Bay and Devils Walkingstick Sites, edited by
W. H. Adams, pp. 152-167. Reports of Investigations 1,
Department of Anthropology, University of Florida,

n.d. Spatial Variation in Orange CulturePottery Interaction and
Function. Draft chapter in Early Pottery: Technology,

Style, and Interaction in the Lower Southeast, edited by
Rebecca Saunders and Christopher Hays. University of
Alabama Press, Tuscaloosa (in preparation).

Sears, William H.
1973 The Sacred and Secular in Prehistoric Ceramics. In
Variation in Anthropology: Essays in Honor of John
McGregor, edited byD. Lathrop and J. Douglas, pp. 31-42.
Illinois Archaeological Survey, Urbana.

Stephenson, Keith
1990 An AMS Date for Cord Marked Pottery in South-Central
Georgia. LAMARBriefs 15:11-12.

Widmer, Randolph J.
1988 The Evolution of the Calusa: A Nonagricultural Chiefdom
on the Southwest Florida Coast. University of Alabama
Press, Tuscaloosa.

Williams, Stephen
1977 Radiocarbon Dates from the Georgia Coast. In The Waring
Papers: The Collected Works ofAntonio J. Waring, Jr.,
edited by S. Williams, pp. 329-331. Papers ofthe Peabody
Museum of Archaeology and Ethnology, Harvard Univer-
sity, Vol. 58. Cambridge.



Florida Anthropological Society Chapters

9 16 10


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

2) Broward County Archaeological Society 5
181 S. Federal Highway, Dania Beach, FL 33004

3) Central Florida Anthropological Society
'.O. Box 947544, Maitland, FL 32794-7544 4

) Central Gulf Coast Archaeological Society 13 11
'.O. Box 82255, St. Petersburg, FL 33682
15 "^* 5
5) Indian River Anthropological Society
1705 S. Tropical Trail, Merritt Island, FL 32952

,) Kissimmee Valley Archaeological and Historical Conservancy 2
195 Huntley Oaks Blvd., Lake Placid, FL 33852 12

7) Northeast Florida Anthropological Society
1144 Torino Place, Jacksonville, FL 32244

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

9) Pensacola Archaeological Society
P.O. Box 13251, Pensacola, FL 32591 13) Timi
P.O. Box
10) St. Augustine Archaeological Association
P.O. Box 1301, St. Augustine, FL 32085 14) Volu
P.O. Box
11) Southeast Florida Archaeological Society
P.O. Box 2875, Stuart, FL 34995 15) War
P.O. Box
12) Southwest Florida Archaeological Society
P.O. Box 9965, Naples, FL 34101 16) Eme
333 Persi

e Sifters Archaeology Society
25883, Sarasota, FL 34277-2883

sia Anthropological Society
1881, Ormond Beach, FL 32175

m Mineral Springs Archaeological Society
7797, North Port, FL 34287

rald Coast Archaeological Society
mmon Street, Freeport, FL 32435








Department ofAnthropology, University of South Florida, 4202 E. Fowler Ave., SOC 107, Tampa, FL 33620
E-mail: nwhite@chumal.cas.usfedu

The University of South Florida's (USF) 1993 test excava-
tions of submerged shell midden deposits in northwest
Florida's lower Apalachicola River wetlands required complex
equipment and logistics, butprovided new information onLate
Archaic coastal/estuarine subsistence and technology. Van
Horn Creek shell mound also had later Woodland and Fort
Walton components. The single-component Sam's Cutoff
shell mound had a single, shallow burial. Both sites produced
microtools and fiber-tempered pottery, radiocarbon dates, and
information on the Apalachicola's fluvial history.


The Apalachicola River is formed by the confluence of the
Flint and Chattahoochee rivers at the Florida-Georgia border,
and flows 107.4 river miles (ca. 175 km) southward into the
Gulf of Mexico. It is the largest river in Florida in terms of
flow (Livingston 1984), with a great valley rich in archaeolog-
ical resources. The lower delta protrudes into the Gulf (Figure
1), built by continual progradation (accumulation of alluvial
deposits) and shifting of delta lobes as tributary and
distributary streams flow and change. It is today a vast watery
wilderness of tupelo swamp, oak hammocks, stands of pine,
and grassy marshland, throughout which are scattered many
mounded shell middens demonstrating human occupation
from at least Late Archaic through late prehistoric Fort Walton
Previous work in shell mounds here (White 1994a; White
and Estabrook 1994) revealed the presence of submerged
cultural deposits that could not be investigatedby conventional
methods. Testing Late Archaic components at Van Horn
Creek and Sam's Cutoff shell middens in 1993 was ambi-
tiously planned to involve dewatering with a large pump and
well points. One goal was to obtain better dates for
fiber-tempered pottery in this valley; another was to explore
relationships with the Late Archaic Poverty Point adaptations
known from Louisiana (Gibson 2000, Webb 1977) and
elsewhere along the Gulf Coast and in the lower Mississippi
Valley, including the Elliott's Point complex of northwestern
Florida (Lazarus 1958).
Another area of investigation was ancient sea level
fluctuations and the geomorphological history of the lower
river. The 1987 testing at Van Horn Creek (White 1994a), on
the east side of the lower delta, had suggested there was
freshwater Rangia clam harvesting during the laterprehistoric
time periods but utilization of oyster and more saltwater fish

earlier, during the Late Archaic. Sam's Cutoff shell mound,
with only a single Late Archaic component and slightly farther
east, is made predominantly of oyster. If the main channel of
the Apalachicola River was farther west during the earlier
time, conditions would have been more saline around these
two sites, resulting in Late Archaic exploitation of oyster and
more saltwater species. When the river migrated eastward it
would have brought fresh water conditions to the local
environment. All the shell middens on the western side ofthe
delta are composed predominantly of freshwater Rangia
clamshells, even in Late Archaic levels. The hypothesis to be
tested was that these easterly Late Archaic components would
contain more saltwater fauna if fresh water was farther away
in the earlier Holocene.
This article summarizes the investigations and adventure
of the 1993 project (reported in detail in Fradkin 1994; Weill
and White 1994; White 1994b). Fieldwork was conducted
with a crew averaging 12, including USF field school students,
volunteers, and technicians from the Apalachicola National
Estuarine Research Reserve. The field season comprised 6
weeks, from June through mid July. We used (and abused) 3
trucks, 4 boats, 3 pumps, and lots of ingenuity to achieve
partial dewatering of cultural sediments. This also was the
first year that I videotaped all field operations, for future
(research, inspirational, and entertainment) use.

Van Horn Creek Shell Mound

Logistics, Excavations, and Stratigraphy

First investigated was Van Horn Creek shell mound,
8FR744, on the banks of a tiny stream deep in the river swamp
(Figure 1,2). Located during survey in 1983 (White 1987) and
tested briefly in 1987 (White 1994a), this site is a long, narrow
Rangia mound oriented southeast-northwest but not parallel
with the current stream bank. It is 30 m wide, 90 m long, and
rises 1.5 m above the surrounding wetlands. The four units
dug in 1987 demonstrated later components of Fort Walton
and possible Early or Late Woodland age (the latter based only
on the presence of check-stamped and plain ceramics occur-
ring deeper than the Fort Walton materials). Late Archaic
deposits were at the very base of the excavations just before
reaching the water table, and clearly extended well below this
depth (148 cm below surface on the mound summit in Test
Unit 1). The upper cultural components were mixed, doubtless
continually disturbed by repeated prehistoric occupations, as


VOL. 56(1)


MARCH 2003


Figure 1. The lower Apalachicola valley, showing Van Horn Creek (8FR744) and Sam's Cutoff (8FR754) shell mounds, and
other mounded shell middens in the estuarine/river swamp area.

well as pothunters (incredibly, in this extremely remote
location). This resulted in recovery of some fiber-tempered
pottery and microtools in upper levels of the site. Thus our
work in 1993 also sought to find less-disturbed Late Archaic
Van Horn Creek is not easily accessible. After crossing the
wide Apalachicola near its mouth we had to go into increas-
ingly smaller streams until the creek mouth itself was reached,
then it was 45 minutes of navigating 2 km up this creek, in
many places barely wider than our 14-foot boat. The daily
tides seriously affected navigation and excavation. If the tide
was in, the creek had more water and was easier to tackle
(though we had to duck for overhanging branches) but the
water table was higher in the units. At low tide the digging
was easier but we often had to push our boats through a few
cm of mud and water, fending off alligators that fled and logs
that attacked!
Relocation of old test units brought additional problems
worth describing. We had backfilled well in 1987 and
pounded the southwestern corer stake of each of the four

units into the ground. With the heavy ground cover it was
possible to relocate only TU 1, apparent because of a rectangu-
lar depression, and even there the southwestern corer was not
to be found; the wooden stake rotted away in 6 years. We
finally found the southwestern corer of TU 3, recognizable
only as a perfect rectangle (the size of the former stake) of
orange flagging in the ground. In 1993 we pounded iron
rebar into the southwestern corners of the units.
Two tests were excavated in 1993, numbered sequentially
after the 1987 units. Test Unit 5 was 3 x 3 m (dug in 1-meter
squares for better control) and TU 6 was 1 x 2 m. As we
excavated and the water began to seep in, we began setting up
the well point system to pump the units dry (Figure 3). The
intent was to surround the units with points all connected to a
single system and pump the water out, isolating the block of
the unit within which excavation could continue downward
into dry deposits. Equipment was rented from pump company
that also provided a dewatering advisor for a day. Hauling in
the huge pipes, pumps, and other equipment was tricky. The
double-diaphragm mud hog diesel pump, weighing 465 kg


2003 VOL. 56(1)


N -


1.a -

8Fr744 o 5 10
contour interval 20 cm
I test units

edge of visible
~s., swampy areas

Van Horn Creek

Figure 2. Van Horn Creek shell mound, showing 1987 and 1993 excavations.

(1023 lbs.), was towed up the narrow creek on a motorless, 18-
foot boat using another, 20-foot boat, a process that took 4
hours one way. Jetting in well points around the units was
impossible through the clam and oyster shell matrix with only
waterpower from the 4-hp jet pump. Instead, several workers
had to grasp the metal jet pipe and physically pound it into the
ground for 20 minutes for each well point, then quickly yank
it out and stick in the pvc pipe to be connected to the rest of
the system. Even hauling in 100-lb bags of clean white sand
to fill the hole surrounding each well point to avoid pumping
air from the spaces between the shells, and getting plenty of
groundwater rushing out the outflow hose of the big pump, did
not allow dewatering down to the depth of the points. There-
fore the bottom of the cultural deposits was not reached,
though we learned it was at about 3 m depth when we reached
soft sand under the shell during installation of the well points.
After setting up 7 well points, five on the bigger unit and
two on the smaller, we were able to pump with only modest
success. Excavation was done in 15-cm arbitrary levels since
there was no discernible cultural or natural stratification, just

varying proportions ofRangia and oyster shell. Nine-liter (30
x 30 x 10 cm) soil samples from the southwestern corner of all
levels were saved for flotation and one-liter samples for
permanent storage. All other soils were waterscreened using
1/8" mesh; a smaller 3-hp pump was set up at a
waterscreening station on the creek bank. I use the word
"soil" advisedly, since there was little sand and mostly solid
shell throughout.
Test Unit 5. This unit was placed on the highest area of
the summit, with its southeastern corner 72 cm west of 1987's
TU 1, which had produced the best evidence of later freshwa-
ter environments and earlier deposits dominated by oyster and
more saltwater fish species (Walker 1994), suggesting a more
saline earlier local environment. In all deposits there were at
least a few of both kinds of shell, however, and in TU 5 we
encountered mixing of the components, with Late Archaic
fiber-tempered ceramics and microtools present in the upper
levels of both units, as well as in the deepest, and presumably
undisturbed levels. One source of the mixing was a 50-cm-
diameter pit feature filled with Rangia clamshells originating



.Table 1. Cultural materials recovered from Van Horn Creek shell mound, 1993, by counts and weights (grams).

CERAMICS L1 L2 L3 L4 L5 L6 L7 L8 L9 L10 L11 TOTALS L1 L2 L3 L4 L5 L6 L7 L8 TOTALS
Pensacola Inc 2 (29) 2 (29)
Cool Branch Inc 6 (62) 6(62) 5 (39) 5 (39) 1(31)
Marsh Island Inc 2 (10) 2 (10) 2(86)
Pt. Washington Inc 12 (108) 4 (51) 16 (159)
Lake Jackson 7 (75) 7 (75) 5(9) 5(9) 2 (12)
FtWalton Inc 56 (287) 16 (50) 72 (337) 4 (52)
indet punctate 22 (122) 1(5) 1 (1) 24 (128) 5 (14) 5 (14) 2 (28)
indet incised 72 (275) 9 (23) 2 (3) 83 (301 5 (13) 5(13) 4 (31)
shell-t pl 20(64) 1 (1) 3(3) 24 (68) 1 (7) 2 (13) 3 (20) 15(75)
limestone-t pi 192 (630) 9 (81) 13 (22) 214 (733) 3 (10) 3 (10) 18(71)
check-st 88 (519) 56(299) 28(195) 10(10) 182 (1023) 85 (334) 5(14) 90 (348) 32(199)
limestone-t ch-st 2 (10) 2 (10
indet stamped 7(19) 5 (25) 1 (1) 13(45) 11(34) 11(34) 3(19)
grit-t pl 592(2225) 40(152) 25(55) 1(1) 658 (2433) 27(170) 2(5) 29 (175) 19(167)
grog-t pl 84(221) 1(1) 8(23) 3(2) 96 (247) 16(58) 8 (31) 24 (89) 18(186)
sand-tpl 575(1011) 332(512) 63(55) 17(17) 7(11) 994(1606) 184(447) 49(153) 2(1) 236 (602) 81(329)
fiber-t 9 (48) 19(83) 18(195) 9 (31) 4(22) 2(1) 61(380) 3(17) 1(3) 2(8) 1(3) 6 (17) 13(48) 1 (1)
clay ball frag 1(13) 1(13) 2 (94) 1(20) 1(15) 2(55) 6 (184)
clay lump 28(52) 24(44) 17(12) 2(9) 1 (1) 1(1) 4(2) 1(15) 21 (17) 43(64) 134(217) 5(10) 6(10) 1 (1) 5(9) 3(4) 2(2) 52(60) 74(96) 6(16)
TOTAL 1774(5757) 518 (1340) 178(564) 42(17) 9(13) 1(1) 4(2) 1(15) 25(39) 45(65) 2589(7866) 357(1172) 75(323) 4(29) 5(9) 5(22) 1(1) 2(2) 62(78) 511 (1691) 208(1203)
LITHIC ** __
microtool 1(4) 2 (3) 3(5) 1 (1) 1(1) 8 (14 2(2) 1(1) 1(1) 4(4) 2(2)
blade/frags 2(2) 2 (5) 1 (1) 5 (8) 3 (3) 2 (2) 5(5) 2(6)
tool frag 2(6) 2 (6) 2(72) 2(72) 1(249)
core/frags 1(26) 3 (66) 1 (129) 2(191) 1(194) 8 (606) 4 (246) 2 (50) 1(49) 7(345) 7(542)
block shatter 3 (11) 3(1) 6 (12) 3(9) 4(12) 4(4) 11(25) 4(77)
primary decort 3 (9) 1(8) 4 (16) 8(33) 1 (1) 2(8) 3(10) 6(19) 9(83)
2ndarydecort flake 3(11) 9(72) 8(28) 1(45) 2(2) 23 (158) 2(4) 1(1) 1(2) 1(1) 1(1) 1(1) 2(2) 9(12) 12(63)
2ndary flake 2(2) 10(13) 8(10) 4(10) 1(1) 1 (1) 1(1) 9(4) 1(2) 37(44) 5(2) 2(11) 2(4) 2(2) 16(13) 27(32) 9(9)
quartzite frag 1 (6) 1(6)
sandst/limest rock 1(2) 1(2) 6(17) 2(4) 8(10) 1(5) k 1(1) 2(1) 20(66) 2(36)
TOTAL LITHICS 7(40) 26(115) 24(229) 19(251) 3(240) 4(2) 1(1) 1(1) 1(1) 12(7) 1(2) 99(889) 26(284) 13(80) 11(16) 2(6) 2(2) 5(12) 31(152) 138(1619) 48(1067)
WORKED SHELL 2(202) 2(74) 1 1 (.3) 2 (74) 1 (13) 8 (363 1(135) 2(35) 7(327) 4(157)
TU 5: Levels 1-4 = 3 x3x .15 m; Levels 5-8 = 1 x 3 x .15 m; Level 10 = 1 x x .15 m; Level 11 1 x x .13 m; TU 6 levels all = 1 x 2 x .15 m; MIXED= mixed levels, wall fall, both units, and site surface.
primary decortication flake= over 50% cortex; secondary decortication flake = under 50% cortex; secondary flake = no cortex, secondary lithic reduction product; sandstone, limestone may be naturally


Figure 3. Test Unit 5 at Van Horn Creek shell mound, view facing southwest, showing beginning of excavation of Level 4 and
well points in place to begin dewatering.

in the topsoil, that was only distinguishable near its bottom in
Floor 3 (45 cm deep) and in the unit wall. Within the matrix
of solid shell coated in dark yellowish brown (10YR4/4) slimy
clay, the pit deposit consisted of the same solid shell but coated
in black (10YR2/2), more sandy clay.
To meet field deadlines, excavation of TU 5 continued
down into the undisturbed Late Archaic deposits only in the
three western 1-m squares. Only faunal remains and very few
lithic artifacts but no ceramics except for clay lumps were
recovered between 75 cm and 135 cm deep (Levels 6-9). Then
in Levels 10 andl 1 were more stone tools and fiber-tempered
pottery (Table 1). After excavation of Level 9, time and dry
deposits were running out. Despite having curtailed work at
TU 6 and moved the other two well points to surround TU 5,
it was still not possible to excavate more than Levels 10 and
most of 11 in the single 1-m southwestern square of TU 5
before the water seeped in even during pumping.
In retrospect, it seems we were still pumping lots of air
from the spaces between the shells. In addition, there was a
small amount of clay content in some of the upper strata (the
dark and greasy coating on the shells, perhaps the residue of
human activity or other organic decay) which may have been
holding in a perched local water level, so that while we were
pumping out water below this, some remained trapped above
the pumping zone, seeping into the unit. At the time nobody,

including the dewatering professional, realized that perhaps
the pump should have been kept running 24 hours, not just a
morning, before digging deeper, to give the needed time for it
to dewater properly. Nor was it realized that raising or
shortening some of the well points could have been done to
drain the water that may have been perched above the dewater-
ing zone, as has been done successfully elsewhere in Florida
(McGee and Wheeler 1994:338)
The final depth of excavation was 163 cm below the
surface. This is only 15 cm or one level deeper than was
achieved in 1987 without pumping. From the jet-
ting/excavation of the well point holes we knew the midden
was at least 3 m thick, so about half the depth of it was
sampled in TU 5. A core taken in 1987 off the southern end
of the mound had shown clayey soil, mixed with crushed shell,
continued to a depth of 2 m under the mound edge. It is
entirely possible that deeper materials were leftby even earlier
cultural groups than those represented by the Late Archaic
Test Unit 6. This unit was placed 15 m south-southeast
and downslope from TU 5. It was taken to a final depth of 117
cm below surface when water seeped in despite pumping, and
it was decided to end excavations there and move its well
points to TU 5. Cultural stratification was expectably com-
pressed (on the slope) in comparison with that at TU 5, so that

19 -




2003 VOL. 56(1)

Table 2. Fauna recovered from Van Horn Creek and Sam's Cutoff shell mounds, 1987-93.

Odocoileus virginianus white-tailed deer X X X
Sylvilagus spp. rabbit X x
Sigmodon hispidus hispid cotton rat X
Neofiber alleni round-tailed muskrat X
Rodentia rodents X X
Cricetidae mice X
Procyon lotor raccoon X
Anatidae duck X
Chen caerulescens snow goose X
Kinosternidae musk/mud turtle X X X
Terrapene carolina box turtle X
Pseudemys floidana Florida cooter turtle X
Emydidae water turtle X X
Pseudemys scripta pond slider turtle X
Trionyx ferox softshell turtle XX_
Crotalus spp. rattlesnake X
Colubridae colubrid snakes X
Serpentes snakes X X
Alligator mississippiensis alligator X X X
Carcharhinidae requiem sharks X
Lepisosteus spp. gar fish X X X
Brevoortia sp. menhaden X
Eops saurus ladyfish X X
Amia calva bowfin X
Aius felis hardhead catfish X X X
Bagre marinus gafftopsail catfish X X X
Leiostomus xanthurus spot X
Lepomis sp. sunfish X
Caranx hippos crevalle jack X X X
Cynoscion nebulosus spotted seatrout X
Archosargus sheepshead X X X
Lutjanus sp. snapper X
Micropogonias undulatus Atlantic croaker X X X
Pogonias cromis black drum X X
Sciaenops ocellatus red drum X X X
Mugil sp. mullet X _.X
Sparidae/Sciaenidae porgy/drum X X
Osteichthyes bony fish X X_ _X
Perciformes perch-likes X
Pristis sp. sawfish X

the Late Archaic deposits were reached by the beginning of
Level 4. In addition there was possibly a bit more mixing; in
Level 8, 2 tiny sand-tempered sherds occurred among the solid
oyster andfiber-tempered pottery (these sherd crumbs may just
be broken off bits of fiber-tempered vessels too small to show
the distinctive fiber lines). A flotation sample of 9 liters was
taken from the inundated zone below (and labeled Level 9). It
contained no artifacts, only faunal remains; this may be a

result of the small sample size. Lithic artifacts occurred
throughout the deposits, but Levels 3-6 had very few artifacts,
similar to the situation in TU 5.

Materials Recovered

Table 1 lists all artifacts recovered in 1993, also giving soil
volume of the level. The mixing of Late Archaic deposits



Table 2, continued. Fauna recovered from Van Horn Creek and Sam's Cutoff shell mounds, 1987-93.

Balanidae barnacle (M) X X
Phalium granulatum Scotch bonnet shell X
Ischadium recurvum hooked mussel (M) X X
Crassostrea virginica oyster (M) X X X
Rangia cuneata marsh clam (F) X X X
Polymesoda caroliniana Carolina marsh clam (F) X X
Mytilidae mussels (M) X X
Mercenaria sp.* quahog clam (M) X
Melongena corona crown conch (M) X
Macrocallista nimbosa sunray venus clam (M) X X
Neritina reclivata olive nerite (F, M) X X X
Busycon contrarium* lightning whelk (M) X X X
Pleuroploca gigantea* Florida horse conch (M) X
Pholadidae piddocks (M) X
Martesia striata striate piddock (M) X X
Columbellidae dovesnails (M) X
Ellobiidae melampus (M) X
Oligyra orbiculata globular drop (T) X X X
Hydrobiidae dusky cavesnail (F) X X
Nassaius acutus sharp nassa (M) X X
Odostomia impressa impressed odostome (M) X X
Detracia sp. melampus (M) X X
Gastrocopta pellucida slim snaggletooth (T) X X
Strobilops aeneus bronze pinecone (T) X
Pupillidae snaggletooth (T) X
Euglandina rosea rosy wolfsnail (T) X X X
Physella sp. physa (F) X X
Melampus bidentatus eastern melampus (M) X
Helicodiscusinermis oldfield coil (T) X X
Helicodiscus parallelus compound coil (T) X X
Euconulus chersinus ild hive (T) X
Glyphyalinia indentata glyph (T) X X X
Nesovitrea dalliana depressed glass (T) X X
Stfiatura meridionalis striate (T) X
Hawaiia miniscula minute gem (T) X X
Zonitoides arboreus quick gloss (T) X X
Polygyra cereolus southern flatcoil (T) X X
Polygyra pustula grooved liptooth (T) X
Pomacea paludosa Florida apple snail (F) X X

(M) = marine; F= freshwater; T= terrestrial

above Level 6 in TU5 is apparent. By Level 6, the volume
excavated was 3 times less than when excavations began
because the unit size was reduced from 3 x 3 to 1 x 3 m, and
for Level 10 it was reduced to 1 x 1 m, 9 times less. Level 11
was further reduced because only 13 of the 15 cm could be
excavated before the water poured in. For TU 6, similar
mixing is apparent above Level 4. It is interesting to compare

* very few specimens; probably all artifacts or debitage

the materials from the 1993 work, which opened 11 m2, with
those of the 1987 work, which opened 7 n2. In 19871 did not
have access to waterscreen equipment, so all excavated
materials were sorted on a large wooden board. Waterscreen-
ing itself, as well as use of the 1/8" mesh, produced, for
example, 4.6 times more ceramic specimens (from sherds to
clay lumps; 2.4 times more by weight) in 1993 than were


Figure 4. Clay balls or Poverty Point-type objects from Van Horn Creek shell mound: top left) wedge from TUSECen, L2;
right) two grooved fragments from TU6, LS; bottom) ball from TU5SW, LS.

obtained in 1987, even though only 1.6 times more area was
excavated (White 1994a:59).
Ceramics. During 1993 a total of 2885 sherds (nearly 9
kg) was recovered from Van Horn Creek shell mound (Tables
1,2). Types discussed here are the standard ones (Willey 1949,
White 1982). The Fort Walton component is characterized by
Fort Walton Incised, Lake Jackson (Plain and Incised, with
ticked and notched rims), with occasional sherds ofthe slightly
less common Marsh Island Incised and Cool Branch Incised.
There is a small amount of shell-tempered plain (1.2% by
count, 1.3% by weight) and 2 sherds of shell-tempered.
Pensacola Incised, consistent with typical Fort Walton assem-
blages in this valley. Limestone-tempered plain, also part of
typical Fort Walton here, occurs in higher percentages than I
have seen before (10% by count, 14.4% by weight). There are
several examples of Point Washington Incised or Marsh Island
Incised (types with no punctations, as compared with Fort
Walton Incised) made on limestone-tempered paste. The
greater frequency of grit tempering in the plain ceramics of the
upper levels is consistent with Fort Walton assemblages also;
there is often a mixture of the grit with some grog and/or shell
or limestone.
No diagnostic Deptford or Late Woodland sherds were
recovered. Check-stamped pottery still occurred deeper than
the diagnostic Fort Walton types. Much of the check-stamped

was grog-tempered, and a few sherds were limestone-tem-
pered, casting doubt upon the existence of an Early Woodland
component since limestone is so far associated with Fort
Walton in this valley.
The fiber-tempered pottery was plain-surfaced except for
a single sherd from TU6, disturbed Level 1, with some
irregular impressions that appear to be a production flaw.
Additional bits of Archaic ceramic evidence (Figures 4 and 5)
were the amorphous, blocky clay chunks and a few fragments
of typical Poverty Point type clay balls; all but one of the latter
are grooved along the sides. Earlier I have labeled the chunks
as daub fragments (White 1994a:73-74). This is possible, and
occasionally they have stick or cane impressions that would
confirm this function. However, many of these are hypothe-
sized to be for cooking during the Late Archaic, just as the
clay balls are. Experimental studies suggest they were used
either for dry roasting (Hunter 1975), steaming, or boiling
(McGee 1995, Wheeler and McGee 1994). Other, very tiny
clay lumps of more rounded, irregular form (and unknown
function) were found in consistent numbers throughout the
entire midden, concentrating in the Late Archaic deposits.
Both the rounded and blocky clay fragments are included on
the tables as clay lumps.
Lithic materials. Even from the earlier and mixed levels,
lithic specimens are mostly from a microtool industry (Figure

2003 VOL. 56(l)



F' :
(.~ 4~~
.i .'

Figure 5. Complete assemblage of large to small clay lumps from Van Horn Creek shell mound, TU5SW, L 5 (a 15-cm level
within the 1-x-l-m southwestern sub-unit).

6)., Test Unit 5 produced 8 microtools and 8 cores/core
fragments; TU 6 had 6 microtools and 14 cores. The micro-
tools are all long, narrow side scrapers with intensive damage
from use wear on one or both sides and/or the tip. The other
end was not utilized, and often retains cortex from the original
nodule. None of these tools has an expanding base like typical
Jaketown perforators found at other shell mounds in the region
(White and Estabrook 1994); perhaps their function was
different from that of the perforators. The cores show signs of
being worked until nearly exhausted, with flake (blade) scars
running from both ends. Other tools recovered are a few
unifacial andbifacial items or fragments, and small, long, thin
blades that clearly came off the typical microcore and were not
yet retouched or utilized as tools. Some blade fragments had
broken on hinge fractures and may have been manufacture
rejects; some cores also demonstrate hinge fractures.
The remainder of the lithic assemblage consists of primary
and secondary debitage, a small amount of it thermally altered,
that is consistent with core production. There is so little of it
that it suggests bringing prized and relatively rare (on the
coast) lithic raw material to the site in the form of worked--
down nodules. All the material is chert that was probably
obtained at least 150 km (80 river miles) to the north; there is
no agatized coral from beach rock. The whole assemblage is
consistent with the Late Archaic cultural affiliation, and with

the long-observed fact that Fort Walton components in this
valley manifest a striking absence of lithic remains (this of
course leaves no stone tool industry to the hypothesized
Woodland people who possibly lived here too, so it is easier to
interpret if instead they are just the very earliest Fort Walton
folks). Other rock recovered included a quartzite fragment
possibly for grinding, and sandstone and limestone pieces that
may have been natural inclusions.
fWorkedshell. This category, comprising 16 specimens in
1993, includes Busycon sinistrum (lightning whelk)
columellae and cut fragments similar to those obtained in
1987. and 3 tiny disc beads, from Levels 4 and 5 of TU5 and
L 8 of TU6. Another item is a quahog (Mercenaria) clam
fragment, rare in northwestern Florida sites. Combined with
the 11 specimens found in 1987, this assemblage indicates a
minor shellworking industry. However, there are too few
items to support the idea that the microtools were for working
shell. Furthermore, it is not certain which components) the
worked shell might be associated with; it is certainly well
distributed vertically. The cut Busycon from TU5,.L10 is
clearly Late Archaic, and probably the disc bead from TU6 L8,
since it is associated mostly with fiber-tempered pottery, a
microtool and core, and only the two tiny sand-tempered sherd
crumbs. Probably people from all time periods made shell
tools, though in far smaller numbers than in southern Florida,





Figure 6. Microliths from Van Horn Creek shell mound. Proveniences: l-r, top) TU6 L5 (tiniest one); TU5WCen L7; TU5SW
L5 (broken midsection); surface, SW area of site, TU6 L1; middle) TU5 NW L2; TU5WCen L3; TU5SW L3; TU5Cen L1;
bottom) TUSSW L10 (utilized blade); TU5NCen L3.

where stone is less available. Other than the beads, all the
shell artifacts appear to be expedient tools. Further research on
shelltool industries of northwestern Florida is underway in the
USF archaeology lab.
Zooarchaeological Remains. All excavation levels
produced faunal remains. Small samples, recovered by both
waterscreening and flotation, from Test Unit 6, Levels 1, 3, 5,
7, and 9, were analyzed (Fradkin 1994). A summary of fauna
present is given for both the earlier and later components in
Table 2 (including also the species identified during the 1987
work), and Fradkin's excellent detail work for the analyzed
proveniences is given in Appendices A-E. Most of the fauna

are fish and turtles, in addition to the shellfish. Fradkin was
impressed by the numbers and diversity of species of small
gastropods present in all levels and went to great pains to
identify the 13 species of terrestrial snails and 4 freshwater
and 3 marine snails. Explaining their presence in the midden
is difficult, and their stratigraphic frequency distributions are
hard to interpret. Terrestrial snails may have been commen-
sals, that is, they either slithered in or were obtained by people
in the process of procuring something else; similar to our ideas
about barnacles and other such species, they are not expected
to have been eaten. Some of the mussels might have been
clinging to oyster shells. Many of these types of tiny shelled


2003 VOL. 56(1)




unident bivalve [ | freshwater clam BM oyster

Figure 7. Oyster and clam percentages by weight by level at Van Horn Creek shell mound, Test Unit 6, suggesting changes

through time in use of fresh and saltwater environments,

creatures might not be efficient to obtain or process, unless
they were thrown into the cooking pot just to add a dash more
protein (or flavor?). The snails may be important environmen-
tal indicators. Further research on these species is clearly
needed. . ,

Fradkin identified 70 taxa of animals from the Van Horn
Creek samples, adding 40 taxa to those identified by Karen
Walker in 1987 (Walker 1994), and including many turtle
species; fish such as bowfin, hardhead and gafftopsail catfish,
and jack crevalle; muskrat; rattlesnake; snow goose; and the

many snails. The goose is interesting because 1) it is in a Late
~ Archaic context; 2) it is present in winter in both freshwater
and salt marsh environments, giving one of the few seasonal
indicators known from area shell middens; and 3) birds are
generally rare (for unknown reasons) in area shell midden
faunal assemblages. Future work should involve calculation
of comparative meat weights for the different species and
reconstruction of their estimated proportions of the diet. For
now, based on minimum numbers of individual animals in the
" representative samples from flotation, shellfish make up over
95% of the assemblage, with the next-largest component being
saltwater fish, then turtles.

Concerning the hypothesis that more saltwater species were
used earlier in time, the data are more complex than the1987
results. Comparison of oyster, a more saltwater species, with
freshwater clams in the representative samples obtained by
flotation from TU6 is useful (Figure 7). In Level 1, Rangia
predominates, at 55% by weight, compared with 17% oyster
(with another 24% of unidentified bivalve shell fragments).
Level 1 also had only terrestrial snails in the flotation sample,
and one freshwater snail in the waterscreened deposits. All
this evidence suggests more use of freshwater environments
during the Fort Walton occupation. In Level 3 the sample was
97.5% oyster by weight; in Level 5, 98%; in Level 7, 99%.
These levels had only fiber-tempered pottery, if any, support-

‘ing the idea that the Late Archaic deposits represent species

drawn from more saltwater environments. . However other
fauna from these levels are from both fresh and saltwater, and
the marine, freshwater, and terrestrial snails also occur in no
patterned frequencies, though terrestrial species predominate.

Below all this, the Level 9 sample (which had no artifacts)
was only 52% by weight oyster-compared with 37% clam
(Rangia and a few Polymesoda, another freshwater species).
By minimum numbers of individuals in Level 9 there were 51
oyster, 137 Rangia, and 11 Polymesoda in 9 liters of soil.
Furthermore, it is easier to determine oyster than clam in small
fragments, which means that the 11% of unidentified’bivalves
in the Level 9 sample are probably also clamshell fragments
(see Figure 7). This Level 9 sample was the one taken below
the water table from unexcavated deposits just before closing
the unit, so it was less controlled and possibly less representa-
tive. But if it is a good sample, it may mean that earlier Late
Archaic people still utilized more saltwater environments, but
also fresh water zones nearly half the time. If the unstated
assumption here is that people collect mostly from their local
surroundings, the faunal sequence might be explained by
fluctuating sea levels earlier in the Archaic.

One type of animal specimen that has been commonly
recovered from Apalachicola delta shell middens is the
pneumatized (puffed-up) fish bone, usually from a jack or
other species. These have also been called hyperostoses or
Tilly Bones, and their function in the fish is still unknown
(Tiffany et al 1980). They are dense and resemble antler tips

. at first glance, and the first impression is that they are some

kind of tool. Our zooarchaeologists indicated that they were
simply another, perhaps more preservable part of the fish
carcass, but they aré so common and useful looking that they
must have been collected for something. This could be said
about any frequently occurring faunal remains, but at Van
Horn we finally recovered some of these specimens with cut


-. IPOT, .. .-0 -

Figure 8. Pneumatized fish crevallee jack?) bones from Van Horn Creek shell mound, showing human modification: top) from
TUSSW, L5, with cut or worn notch; bottom) from TU5NCen, L2, with both ends cut off.

marks on them that looked like more than butchering marks
(Figure 8).
EthnobotanicalRemains. A sample consisting of charred
floral remains from the same proveniences as those whose
faunal materials were studied contained few surprises or
distinctions by component. As with the 1987 work, most of
the specimens were pine, hickory nutshell and acorn shell
fragments, which could have been fuel as well as food.
Species not previously known from the archaeological context
of the site are represented by 3 seeds: hornbeam, blackgum,
and possibly dogwood, all from Level 1. Other specimens
include indeterminate seeds and other plant parts. The general
environmental picture is one of mixed pine and hardwood
forest. The presence of pine may indicate a drier forest during
the Archaic from what exists today, where pine is rare and
tupelo, gums, and other hardwoods and cypress dominate.
Radiocarbon Date. Very little charcoal suitable for dating
was recovered from both seasons of excavations at Van Horn
Creek. For the 1987 work less than one gram of charcoal from
TU3, L6, thought to be undisturbed Late Archaic deposits,
produced an AMS radiocarbon date of 1120+75 years (Beta
26897), calA.D. 770-1030 (2-sigmaor 95% probability range;
Calib 4.3, Stuiver and Reimer 2000). Clearly this date is not
Late Archaic. Consultation with the experts at Beta Analytic
suggested the possibility that such a tiny piece of charcoal

could have traveled downward through the spaces between the
shells (those same spaces that were such trouble in the
pumping out of the water). After the 1993 work, another date
was desired but again little appropriate charred material was
recovered, and the grant funds had to be spent before all
materials were able to be processed. The goal was to obtain a
date on the Late Archaic component from as deep as possible,
so another sample was chosen from a 1987 unit, with better
stratigraphic context and well-correlated with the 1993 work.
Test Unit 3's stratification was similar to that of the 1993 TUs
5 and 6. In TU 3 Levels 6-9 produced no artifacts; then in
Level 10, right at the water table 144 cm below surface, fiber-
tempered ceramics were encountered in undisturbed Late
Archaic context (White 1994a:65). From this deep level the
<1 g of charcoal submitted for AMS dating is thought to be
representative of the later Late Archaic. Of course, this bit of
charcoal also may have traveled downward like the other one,
but the date returned was very reasonable: 3170+ 60 radiocar-
bon years (Beta-73523), cal. 1597-1314 B.C. Another date
was obtained in 1998 from a similar small piece of charcoal
from Level 11 just below the water table in TU3. The age
returned was 3150+50 (Betall9067), cal. 1520-1313 B.C.
(both dates calibrated at 2-sigma or 95% probability range
using Calib 4.3, Stuiver and Reimer 2000).
Summary. The evidence at Van Horn Creek shell mound


2003 VOL. 56(1)

jri .r

:li ~~-~-~%"i-~


suggests that the Late Archaic occupation, dated to about 1400
B.C., is represented by thick deposits from repeated use
throughout years, decades, or even centuries. The absence of
Late Archaic pottery in some levels and relatively high
frequency of it in others, by contrast with the continued
presence of clay lumps in all levels, tempts one to hypothesize
different activities for different individual (seasonal?) occupa-
tions. Late Archaic groups probably initially settled there for
a season of collecting because at the time it was a stream bank
and/or bayshore locale which provided easy access to both
terrestrial and aquatic species, freshwater, estuarine, and
marine. Possibly at the time of first settlement the barrier
islands had not even formed yet and the river was much
farther west. The materials recovered during the 1993
operations provide additional support for a good-sized Wood-
land occupation (perhaps corresponding with the later radio-
carbon date) during which people disturbed the top of the Late
Archaic deposits, dug at least one large pit, and left a thinner
stratum of cultural materials that consists of debris from
similar repeated use. The Woodland deposits did not cover the
Late Archaic shell midden uniformly either vertically or
horizontally. They may be up to 30 cm thick, but could not be
isolated completely due to mixing with earlier and later
materials. Fort Walton people (perhaps culturally continuous
with the Woodland people) then reutilized the site for one or
several brief occasions, leaving an even thinner deposit that
barely covered the earlier components and was similarly not of
uniform thickness or horizontal extent. They too harvested
some oysters and other saltwater species, but many more
freshwater clams, suggesting the river had moved eastward
and brought them closer. Noteworthy, perhaps, for social
archaeology, is that, if the Fort Walton people were now of a
more socially stratified and/or agricultural background, they
left no recognizable evidence of this. Occupation for all three
time periods shows identical subsistence on the rich wild
bounty of the coastal wetlands.

Sam's Cutoff Shell Mound

Logistics, Excavations, and Stratigraphy

To get to the second site tested, Sam's Cutoff shell mound
(8FR754), was even more of an adventure. This small oyster
mound rises <40 cm above the surrounding sawgrass marsh
adjacent to East Bay, and is 155 m north of Sam's Cutoff
(Figures 1, 9), a stream connecting East Bay and the East
River. The portion of the site above water is heavily forested;
it measures 56 m long and 16 m wide, and is oriented just
about north-south. The site was first recorded in 1985
(Henefield and White 1986). Very brief testing in 1991
(White and Estabrook 1994) produced only 4 microtools and
one fiber-tempered sherd, so we sought more and better
evidence in 1993. Logistics here were even trickier; smaller
pumps and pipes were hauled, bit by bit, onto and off our boats
then through the half-hour of thigh-deep muck covered in
sawgrass that reached over our heads. A friendly Coast Guard
helicopter crew airlifted the big mud hog pump in and out

Figure 9. Sam's Cutoff shell ridden, showing 1991 and
1993 excavations.

(Figure 10). However, since the site is an island of tall palms,
oaks, and cedars amid the sawgrass, the huge copter could not
get too close for fear of hitting its rotor blades. So it dropped
the mud hog into the marsh 50 m away. To move it that
torturous 50 meters we hooked up an electric winch carried in
from a boat trailer to the pump battery and wound the cable
around a tree, then had to leapfrog with plywood sheets in
front of the pump's tires to give it a flat surface to roll along as
the winch slowly reeled it over the grass and cypress knees and
up onto the site (Figure 11).
Despite the site's being surrounded by water, our 1991
dilemma had been the lack of water concentrated in one place
for waterscreening. At that time we dug a small sump hole off
the side of the site and let it sit until the mud cleared and it
became a continually filling pool. We used it again in 1993,
to establish an even more reliable, bigger pool for pumping
water to jet in the well points and for the big pump to empty
into. The big pool (affectionately called the swimming pool;
Figure 12) was made as a 3-m diameter circle of chicken wire
lined with roofing insulation covered by plastic sheeting, held
together, as was most of the whole project, with duct tape,
Jetting in the well points (and later coring) was made
easier at this site by excavating the first meter of each hole
with a posthole digger into the hard-packed oyster shells
(material from these holes was carefully screened). As the big




Figure 10. Helicopter transporting mud hog pump to Sam's Cutoff shell mound.

pump filled the big pool from the well points, a small pump
drew water from it back out again for waterscreening. The big
pool provided an effective reservoir, and was (surprisingly)
stable enough to produce an intrusive, blooming purple water
hyacinth. Two 3-hp pumps were needed for waterscreening
because, amid the constant flow of water everywhere, their
spark plugs kept getting wet. This, despite several varieties of
creatively designed shields, using bountiful quantities of
plastic sheeting, which sometimes melted when it brushed the
hot engines (apparently the design engineering ofthe standard
water pump fails to consider its use in the presence of a spray
of water).
One 2 x 4 m unit was opened, surrounded by the seven well
points, and named Test Unit 5, continuing the sequence from
1991. It was less than a meter northwest of TU 1, in which we
had excavated the deepest in 1991, hitting water at 33 cm.
Excavation was again in 15 cm arbitrary levels within eight 1
x 1-m squares. In two squares of this unit we again had only
modest success, reaching depths ofjust over 60 cm before the
water came rushing in despite the pumping. Late Archaic
microtools, cores, clay chunks, and more fiber-tempered
pottery were recovered, but excavation was considerably
slowed when we encountered, at only 4 cm depth, in the four
center 1 x 1-m squares of the unit, an unexpected human
Lack of any modem or historic evidence from the site as
well as skeletal characteristics were evidence that it could only
be a Late Archaic burial. In consultation with the State

Archaeologist, and considering the burial's shallowness and
exposure, permission was given to continue excavation.
Because of the care needed in recovering the burial, excavation
of the central 4 squares (out of the eight 1 x 1 units in TU 5)
was not extended below Level 1. In the NW and SW squares
Level 2 was reached and in the NE and SE, Level 4. No
artifacts were recovered in Level 4; however a few artifacts
from this depth came from well points and cores.
In TU 5, unlike in the units excavated in 1991, there was
oyster shell mixed in with the black sandy humus ofthe topsoil
that overlies the top of the shell midden. This humus layer is
15 20 cm deep, and in the other units it was more distin-
guishable from the slightlybrowner, deeper matrix containing
the shell. Apparently the shell occurred higher up in TU 5
because it had been disturbed to inter the burial. Beyond the
change from black to very dark brown there was no other
recognizable stratification in the 60 cm excavated. A core into
the NE square of TU 5, as well as 7 other cores around the
mound, showed the shell midden bottom to be at 177 cm below
summit ground surface, making the midden thickness to be
approximately 1.5 m. Only cores 3, 4, and 7 contained shell,
in a matrix of dark clay. The other cores contained the same
clay, which smelled of sulphur, but no shell. Geologist Joe
Donoghue of Florida State University examined the mineral
content of this clay and determined that it did not have a
marine origin. A summary of the soil stratification on the
summit of the mounded midden, as reconstructed from the
levels and the cores is as follows:


2003 VOL. 56(1)


Figure 11. Winching the mud hog pump through the sawgrass toward Sam's Cutoff shell mound (an island of palms and
cedars in the grassy marsh), pulling the wheels over plywood.

I. 0-15 cm, black (10YR2/1) sand/forest humus without shell
except where disturbed
II. 15-75 cm, very dark brown/dark brown (10YR2/2 to 3/2)
sand packed with shell
III. 75-100 cm, dark gray (10YR4/1) sand packed with less
IV. 100-130 cm, bluish/greenish gray (10YR5/1) sand with
less shell
V. 130-177 cm, black (2.5Y2/0) clayey sand with less shell,
much crushed
VI. 177 cm and below, bluish gray clayey sand, no shell

Materials Recovered

Artifacts and Radiocarbon Date. The small artifact
assemblage from Sam's Cutoff (Table 3) nicely augments the
even smaller volume of remains recovered earlier (White and
Estabrook 1994). Only one ceramic sherd was recovered
during 1993, but since the 1991 work also produced only one
sherd, this increases the sample size by 100%. Interestingly
this plain fiber-tempered sherd is unlike the 1991 sherd in
being rough and thick, similar to most of the fiber-tempered
wares in this valley. The 1991 sherd was very smooth-
surfaced and hardly appeared fiber-tempered at first glance
(White and Estabrook 1994: Figure 2). In fact, that is why we




Table 3. Cultural materials recovered from Sam's Cutoff shell
mound, 1993, Test Unit 5, by counts and weights (grams).

TU51 TU5 L2 TU5 L3 Core Well points TOTALS
SOIL 1.06m3 .6m3 .3 m' .02 m est..14m3 2.12 m3
fiber-t pl sherd 1(11) 1(11)
clay lump 10(21) _10(21)
clay ball 1 (16) 1 (.2) 1(5) 3(21)
microtool 2(2) 2(2) 4(71) 1 (11) 2(1) 11(87)
primary decort' 2 (3) 2 (44) 1 (6) 5 (53)
2ndary decort 5(11) 1 (.1) 2(2) 1(8) 9(21)
2ndary flake 6(14) 5(6) 1(1) 2(3) 14(24)
core 1 (32) 1(36) 2(43) 4(111)
block shatter 1 (2) 3 (4) 4(6)
ground stone 3 (342) 3(342)
Busycon frag 6 (19) 1 (1) 7(20)
shell columella 7 (59) 1(47) 8 (106)
Melongena 1 (19) 1 (19)
TOTALS 11(30) 42(529) 8(155) 9 (61) 10(67) 80(842)
*primary decortication flake= over 50% cortex; secondary decortication flake = under 50% cortex;
secondary flake = no cortex, secondary lithic reduction product.

broke off a corner to make sure. In doing so we exposed
unburned, undecayed fiber (White andEstabrook 1994:Figure
9) that was definitively identified as Spanish moss, Tillandsia
usneoides. That broken corer of the sherd was sacrificed to
be ground up for an AMS radiocarbon date, which was
returned at 3630 +/- 60 B.P., with a C13-adjustedage of 3720
+/- 60 years (Beta 68513, CAMS 10472), cal. 2292-1942 B.C.
( 2-sigma or 95% probability range; Calib 4.3, Stuiyer and
Reimer 2000), respectably early for fiber-tempered pottery in
this region. Other ceramic materials were clay chunks and clay
ball fragments, as at Van Horn Creek, though far fewer in
number. Also dissimilar to the material record at Van Horn
Creek was the discontinuous distribution of clay lumps, which
only occurred in Level 2; this distribution may be an artifact of
the small sample size and shallow depths reached.
In 1991, four microtools were obtained; in 1993, 11
microtools, as well as more cores and debitage. The micro-
tools are all sidescrapers or long thin needles. As at Van Horn
Creek there are no classic Jaketown perforators, though two
were recovered in the 1991 excavations at Sam's Cutoff. Eight,
Busycon columellae and other worked shell pieces, as well as
aMelongena corona (crown conch) shell also were recovered,
testifying to a limited shellworking industry similar to that at
VanHorn Creek. This considerably expands the worked shell
inventory from the site, as only one columella fragment had
been recovered in 1991.
Biotic Remains. Faunal remains analyzed (Fradkin 1994)
consisted of samples recovered from both 1991 and 1993
investigations at Sam's Cutoff(no previous professional faunal
identifications had been done for this site) by waterscreen and

flotation from Test Units 1, 3, and 5 (NE and
SE 1-m squares). Table 2 shows species
identified and Appendices F-H give detailed
quantifications. Fradkin identified 62 taxa
including rabbit, racoon, deer, rodent, many
turtles, snake, alligator, a dozen fish, and 29
taxa of invertebrates, including many shellfish,
and again several species of terrestrial, marine,
and freshwater snails. More of the species
were saltwater types, to be expected in this
mound of nearly all oyster. Predominant fish
species were marine catfish, sheepshead,
crevalle jack, and croaker. Oyster was so
dominant that only 19 Rangia shells appeared
in the sample analyzed. Even the assemblage
of tiny snails, while again predominantly
terrestrial species, contained more marine
species than freshwater. This assemblage is
consistent with the prediction that Sam's
Cutoff, farther east than Van Horn, was even
farther away from fresh water during the Late
Archaic. Ethnobotanical remains identified
from the same proveniences included mostly
pine and oak charcoal, acorn and hickory
nutshells, and a few seeds, including grape.
These are similar to those from Van Horn and
other Apalachicola shell mounds (White

Human Burial. The human burial at Sam's Cutoff lay
between only 2 and 20 cm deep. The skeleton was tightly
flexed, with the head to the east, knees pointing northeast and
face to the southwest. There were no grave goods. Burial
excavation consumed the largest part of our time at the site,
taking away from the effort to get older and deeper evidence
but adding the kind of data on human individuals that is
equally rare and valuable. Osteological analyses done by
graduate student Lorna Weill (Weill and White 1994) are,
detailed in the original field report (White 1994b). Dental
attributes such as attrition observed on the molars suggests a
diet of gritty, hard foods (Molnar 1972). The interesting
incisors are shovel-shaped on-both the labial and lingual
surfaces, and a mandibular torus is present. Shovel-shaped
front teeth are common in Native American (and north Asian
and other) populations (Mizoguchi 1985). As has been
documented for typical pre-agricultural populations in North
America (Larsen 1980), there is no evidence of cavities. The
postcranial skeleton shows a heavy reliance upon upper body
strength. Striations run laterally along the left humerus, and
the distal end of the left ulna is remodeled to support increased
musculature. The sex of the individual was originally reported
as indeterminate due to the fragmentary nature of the pelvic
bones. But during a visit to USF, forensic anthropologist
Stanley Rhine determined from the shallow sciatic notches that
it was female, though the chin exhibits a strong, square, more
male-looking shape (Bass 1971:73, Brothwell 1982:61). The
femoral circumference of 79 mm is marginally within the
female range, but the more reliable femoral length is370 mm,


2003 Voi. 56(l))


Figure 12. The mud hog pump and the big pool, which stored water for jetting in well points and for waterscreening.

well within the zone for females (Blackl978, Igcan and
Miller-Shavitz 1983:56).
The age of the young woman buried at Sam's Cutoff was
hard to determine; the pubic symphysis, a good indicator, was
unavailable. Attrition on the first molars suggests a range of
25 -35 years, but ossification of the long bone epiphyses to
their respective shafts, usually complete by age 19 (Bass
1971:166), had not been achieved. The head of the right
femur was not attached to the shaft but lay in the ground close
to the hip joint; other long bones were similarly unfused. The
third molars (wisdom teeth) had not erupted, and the degree of
development of the roots (Ubelaker 1989:72), along with all
the other clues, suggest the girl was approximately 16 years
old. Her stature, based on femoral length (Bass 1971) was

estimated to have been 142.5 cm or 4'7".
The skeleton looked fairly healthy. There is some enamel
hypoplasia, dental evidence of nutritional stress, and evidence
of an infection in the right lower canine and first premolar
teeth. According to Rhine, the gums were also infected and
receding (StanleyRhine, personal communication, 1997). The
right mandibular second molar exhibits a wide and flattened
root with a protruding deformation at the base, and there also
is a groove which extends up into the root from the base but
which does not appear to have reached the pulp. Mechanical
stress is suggested in a lower lumbar vertebra in which a lesion
has extended throughout most of the centrum. Deep striations
along the shaft of the humerus where the muscles attach, and
an osseous growth on the distal end of a metacarpal, confirm




the association with the physical activity involved in a fishing,
hunting, and gathering subsistence (Larsen and Ruff 1991).
Also of interest is the broken condition of the legs, which
may indicate a post-mortem or possibly shortly pre-mortem
process. The left and right femora are broken in half at
midshaft with suspiciously straight, clean edges. Similar
breaks have been reported at the midshaft ofhumeri as a result
of the weight of overlying soils or trampling of the ground
surface in skeletons which have lost their organic component
(Brett Waddell, personal communication to C. Wienker,
1994). On the other hand, Curtis Wienker, forensic physical
anthropologist at USF, who reviewed this analysis, reports a
similar clean break in forearm bones as a typical "parry
fracture" suffered when warding off a blow to the head (he had
firsthand evidence from being mugged one night in Tampa).
Interpretation of this burial is speculative. Based on the
flexed position and articulated nature of the skeleton, it
appears to be a deliberate burial, not someone who fell and
died alone in the forest, though the thin soil barely covered it.
The shells occurred higher up in the root mat than in areas
away from the burial, suggesting some minimal disturbance to
the ground for interment; there is no evidence of animal
gnawing that might have taken place if it were not buried.
More puzzling is the body's orientation. It was clearly flexed
and lying on the right side, but the face was turned away
toward the left and at least one hand was underneath. Was
this girl thrown down, hands perhaps tied behind her back,
legs whacked to break them in place, and quickly buried? Or
is this far too melodramatic a scenario to explain what is more
likely the result of natural processes over 4 millennia?
Other Late Archaic burials of this type do occur. A flexed
skeleton at Yellow Houseboat shell mound on the west side of
the Apalachicola delta is probably of this age (White
1994a:88-114), and others are known from shell middens
throughout Florida. But more research is needed on them.
We expect that Late Archaic society was mostly egalitarian,
but by this time they are burying the dead in cemeteries, for
example, in Kentucky at Indian Knoll (Webb 1946). Though
there are no grave goods at such cemeteries, they imply more
of a connection or special dealing with death than sticking
someone in the pile of shell and fish bone garbage before you
leave camp forever.
Summary. Evidence recovered in 1993 at Sam's Cutoff
shell mound supports earlier conclusions that the site is a
single component Late Archaic settlement and that it was in
more of a saltwater environment than was Van Horn Creek
and other Archaic shell middens to the west. A very small
amount of faunal remains from the core into the bottom of
TU5 are the same species as the rest of the samples, suggesting
continuity of adaptation. The radiocarbon date obtained places
the latest occupation at around 4000 years ago, about 2100
B.C. Possibly the mound was built up by many successive
short-term occupations, perhaps in seasonal fashion, though
there is so far no determination of seasonality. The inhabitants
left a burial of one teenage girl at the end of the site's period
of use.

Delta Lobes and Fishes

Environment and subsistence during the Late Archaic are
well documented with the data from this project, which lend
further support to the hypothesis that the fresh water of the
river was farther to the west prior to 4000 years ago, then
moved eastward. The fluvial history is tied with sea level rise
at the end of the Pleistocene. The Apalachicola and its
distributaries formed various delta lobes as alluvium built up
at the mouth. Paleo-channels have been documented farther
west; today's Lake Wimico and other streams are former
segments of the river's main channel, and evidence of forma-
tion of earlier delta lobes is present north and west of the
modern delta. Rising sea level may have drawn the river
channel eastward, abandoning those earlier lobes (Donoghue
and White 1994). Fish and shellfish collected by Late Archaic
populations at Sam's Cutoff and Van Horn Creek included
more oysters and other saltwater species than their contempo-
raries on the west side of the delta.
Sam's Cutoff is the only shell mound so far known in the
Apalachicola delta to have no cultural deposits later than the
Late Archaic. This may be because its original low elevation
and proximity to the bay resulted in its near inundation
through the centuries, making it lower, less attractive, and less
visible to later populations. At Van Horn Creek the original
hypothesis was that the later cultural components) are
associated with more freshwater species. The 1993 work
supports this view but shows evidence of both freshwater and
marine environments earlier in the Late Archaic, suggesting
possible fluctuation in both the river's movement eastward and
the sea levels. Future work will emphasize better dating of
these subsistence shifts and correlations with what is known
about sea level changes.
The picture of estuarine subsistence during the Late
Archaic is greatly enhanced by the new data from 1993.
Several animal species have been added to the list of taxa for
Van Horn Creek and an assemblage of fauna for Sam's Creek
is described for the first time. Probably more important as
food than all the shellfish, several fish species seem to have
been the most desirable, especially marine catfish. There also
is a high frequency of gar fish remains, as at other delta shell
middens, possibly because they are so easy to catch. The
assemblages of plant remains indicate use of pine and hard-
woods for fuel, utilization of nuts and fruits such as grape, but
little indication of seasonality or much of the vegetable diet.
It is possible that the prehistoric people ate mostly meat,
predominantly fish and turtle and lesser amounts of shellfish.
More likely, the kinds of wetland and other plants utilized,
from swamp cabbage (palm hearts) to smilax root, left no
preservable parts. Future research should certainly involve
examination of saturated deposits for pollen and other less
obvious evidence. It was hoped that excavation of wet deposits
would produce more perishable artifacts and biotic remains,
Clearly these shell middens were deposited on dry land. Rising
and fluctuating water, from daily tides to larger-scale sea level
changes, later partially inundated the sites.
The enormously rich river swamp/estuarine environment,


2003 VoL. 56(1)


even if drier during the Late Archaic, provided a bounty of
species useful for food, tools, and other functions. Several
studies suggest prehistoric shell middens were occupied during
specific seasons only (e.g., Claassen 1986), but most species
utilized at the Apalachicola shell mounds were available year-
round (though the snow goose from Van Horn Creek suggests
winter occupation). The most notable aspect of subsistence is
the lack of significant change through time. The later compo-
nents at Van Horn Creek present a very similar record of
fishing, shellfishing, and hunting, from the Late Archaic
through Fort Walton times, some 4000 years. Though the
species may vary somewhat due to different local aquatic
regimes (saltwater or fresh), the pattern is the same.

External Connections

The Late Archaic material culture at both sites is easily
consistent with the range encompassed by the Poverty Point
Complex first defined in Louisiana and recognized all along
the northern Gulf Coast (Broyles and Webb 1970, Gibson
2000, Webb 1977, Webb and Gibson 1991, Byrd 1991). This
includes Florida, where the Elliott's Point Complex was the
name Lazarus (1958) gave to the contemporaneous adaptation
along the northwestern coast that had Poverty Point clay
objects, apparently using the separate term because the
materials were in Florida, not Louisiana. Dating of Elliott's
Point materials is unclear; Fairbanks (1959:98) notes that
Lazarus's lack of dates, then as now, showed only that the
materials were earlier than Deptford. The concept of Elliott's
Point as a localized adaptation on Choctawhatchee Bay
(Thomas and Campbell 1991) is now clearly inappropriate, as
the materials occur all along the Florida northwest coast, and
items such as clay balls and other Poverty Point-related
artifacts have come from the Atlantic coastal area and also
farther down the Gulf near Tampa. Another unclear aspect of
this time period is the relationship of clay balls and fiber-
tempered ceramics, which overlap, but inconsistently.
Calvin Jones (1993) identified some 90 sites with Elliott's
Point components in the Florida panhandle, but the data
remain unpublished. This paper describes the first in the
Apalachicola valley to be investigated in such detail and
radiocarbon-dated. Social and economic connections 3000-
4000 years ago across the northern Gulf Coast could have been
in the form of long distance exchange, but other systems are
more likely. In the Apalachicola delta the characteristic status
items known from Poverty Point, such as elaborate lapidary
work or large earthworks, are nonexistent. Connections
leading to the similarities in lithic industries and clay ball
cooking were probably those of general domestic tasks done in
a similar way in similar coastal wetland environments.
Continuing analysis will examine the role of these sites and
the Apalachicola delta area within the network of Poverty
Point-related cultures all along the northern Gulf Coast. The
microtools newly recovered by this project expand the data set,
providing enough specimens for future work on use wear and
manufacturing trajectories. The Late Archaic burial addition-
ally offers insights into the life of one person who lived during

this time period. The fiber-tempered ceramics, extremely rare
at Sam's Cutoff during the earlier Late Archaic and more
numerous later in time at Van Horn Creek, are still the subject
of debate concerning their role in Poverty Point-related
material culture. The best interpretation here is that they
emerged sometime before 2000 B.C., developed slowly and
lasted well over 1000 years. A similar puzzle is the smooth or
grooved clay balls, few in the Apalachicola shell mounds, but
certainly present and consistent with Poverty Point material
culture, and hypothesized to be for cooking. Far more numer-
ous are the chunky fired clay lumps, also thought to be for
cooking. Some are smoothed and others, rough. They could be
from broken clay balls, but some look more like fragments of
larger clay cooking surfaces that might have cracked into
pieces. If they were for boiling, the smaller or smoother bits
might be fragments sloughed off during use (McGee 1995;
Wheeler and McGee 1994). All these clay items are more
numerous in deeper levels at Van Horn, but not at Sam's
Cutoff. The hypothesis that their use for cooking predates the
emergence of pottery needs careful testing, as does the idea
that ceramic vessels made them unnecessary for cooking and
replaced them.
Radiocarbon dates are statistical entities and a single one
from a site is insufficientfor definitive statements, but compar-
ison of the dates from this project is interesting. The cali-
brated ranges of the dates from Van Horn Creek and Sam's
Cutoff are farther apart than the 2 standard deviation range
that assures 95% probability of correctness (in fact, they are
2.3 sigmas or 550 years apart). Thus there is a high probabil-
ity that the sites were occupied at two different times. The
intercepts of the two dates with the calibration curve are at
1420 B.C. for Van Horn Creek and 2070 B.C. for Sam's
Cutoff, a 650-year difference. As the Late Archaic component
at Van Horn is more extensive, up to 3 m thick, it may indicate
longer occupation than that at Sam's Cutoff, which is 1.75 m
thick at most (and deeper deposits may be from even earlier
Archaic times at both sites). The simplest explanation is that
Sam's Cutoff was abandoned earlier during the Late Archaic,
when they were just starting to make pottery, and Van Horn
Creek was occupied later and possibly continually attractive
because it was more visible, higher above water.

Reflections on Methods

Details of method are necessary to relate, not only because
they may be useful in future research, but also because the
manner in which the archaeological record is extracted is
enormously important to understand for doing analysis and
interpretation. I include the story of our fieldwork because
knowing how archaeological knowledge is produced is crucial
for anyone trying to evaluate it (and it was good adventure!).
Usually archaeological excavation below the water table is
easier. At sites in sandy lake bottoms or coquina shell
middens, well points can be jetted in quickly, though wet site
excavation is still more difficult and expensive than conven-
tional digging (e.g., Doran 2002; McGee and Wheeler 1994;
Purdy 1988 ). Oyster or clam shell deposits are not like peat




or small coquina shells in sand. Apparently a larger volume
of air is retained in the spaces around the big shells, so
pumping may be less efficient, like sucking air when trying to
drink through a perforated straw. Successful dewatering
strategies in archaeology have dealt with problems through
good engineering. For example, on the Florida east coast at
Groves' Orange Midden, piston cores around the inundated
midden of small shell, sand, and peat established depths of
deposits, and shortening well points drained water trapped
above deeper deposits (Wheeler and McGee 1994:333-338).
At Windover, during draining of the pond and peat, sand was
packed around the well points mostly to anchor them and filter
the water at the intake point (Doran and Dickel 2002). In the
estuarine Apalachicola wetlands, it was not possible to bring
in heavy equipment (especially with our tiny budget), and the
high pressure of our rentedjet pump was nowhere near enough
to jet in the well points without heavy physical labor of
pounding them in. Coring by hand through the shell was
similarly impossible at Van Horn Creek, and difficult at Sam's
cutoff, though it was not difficult off the shell in the swamp
muck. Pouring sand around the well points did not help,
though it may not have been enough sand. Raising the well
points may have helped if water was trapped above them.
Other ways of increasing our success may have been pumping
more steadily and longer, and using more well points.
Dewatering to excavate deep deposits on the Tennessee-
Tombigbee Waterway in northeast Mississippi was done by
machine excavation of drainage trenches surrounding a block
of earth, then pumping water from the trenches and isolating
this dry block for excavation units (White 1983:205-207).
While this is a successful method, it goes to the extreme in one
of the negative aspects of dewatering: there is always damage
from less-controlled excavation of portion ofthe archaeologi-
cal deposits in order to get the machinery and equipment
installed. At the shell middens of northwest Florida we tried
to do as little damage as possible, but the nature of the shells
themselves may have been the biggest problem. Rangia shell
mounds in low estuarine settings along the Gulf Coast have
never been easy to investigate. In the Mobile delta Trickey
and Holmes (1971:117) found earthen dikes to be only
partially effective. InLouisiana excavation far below the water
table at Rangia mounds was done with elaborate engineering
and expensive tools such as helicopters and large cofferdams
provided by oil companies (Neuman 1976). We expected the
Apalachicola project to be difficult, but gained valuable
knowledge for tackling wet deposits in the future. We also had
great luck: If this project had been attempted during the
summer of 1994 instead of 1993, the two tropical storms and
record 100- to 500-year flood that raised the water level two
meters would have made it impossible; the entire lower delta,
shell mounds and all, was submerged.


Ithank the USF 1993 summer field school crew and many people
of Apalachicola and Franklin County, especially Apalachicola
National Estuarine Research Reserve director Woody Miley,
education coordinator Erik Lovestrand, and staff members Dora

McCarthy (who videotaped the helicopter), Steve Travis (who
designed our pumping logistics) and Pat Millender and Jimmy Moses
(who slogged through the swamps all summer with us, fixed balky
equipment, and kept laughing). The Florida Division of Historical
Resources provided a Historic Preservation grant, and grants
specialist Robert Taylor cheerfully waded through the muck to visit.
Student Lorna Weill did the skeletal analysis with advice from USF
physical anthropologists Lorena Madrigal and Curtis Wienker.
Florida State University geologist Joe Donoghue and his graduate
student studied soils and helped haul out 15-foot pipes and 60-lb
pumps. ZooarchaeologistArleneFradkin, then atthe FloridaMuseum
of Natural History, analyzed the faunal remains, and ethnobotanist
Elisabeth Sheldon of SITE, Inc., in Montgomery, Alabama, did the
flora. Archaeologist Karen Walker did the original analysis of Van
Horn Creek fauna that suggested major environmental change after
the Late Archaic. Ryan Wheeler, Ken Sassaman, and other reviewers
provided great ideas and references. I also appreciate the U.S. Coast
Guard helicopter, and the support of the Friends of the Reserve.
Finally, I owe great thanks for the field lodging donated in 1993 by
Bob and Edda Allen at their East Point establishment, Sportsman's
Lodge, "the fishing capital of the world."

References Cited

Bass, William M.
1971 Human Osteology: A Laboratory andFieldManual of the
Human Skeleton. Missouri Archaeological Society, Colum-

Black, T.K., II
1978 A New Method for Assessing the Sex of Fragmentary
SkeletalRemains: Femoral Shaft Circumference. American
Journal ofPhysical Anthropology 48:227-232.

Brothwell, D.
1982 Digging Up Bones, 3rd ed. British Museum, London.

Broyles, Bettye J., and Clarence H. Webb (editors)
1970 The Poverty Point Culture. Southeastern Archaeological
Conference Bulletin No. 12, Morgantown.

Byrd, Kathleen M. (editor)
1991 The Poverty Point Culture: LocalManifestations, Subsis-
tence Practices, and Trade Networks. Geoscience andMan
29. Louisiana State University, Baton Rouge.

Claassen, Cheryl
1986 Shellfishing Seasons in the Prehistoric Southeastern United
States. American Antiquity 51:21-37.

Donoghue, Joseph F., and Nancy Marie White
1994 Late Holocene Sea Level Change and Delta Migration,
Apalachicola River Region, Florida. Journal of Coastal
Research 11(3):651-663.

Doran, Glen H. (editor)
2002 Windover: Multidisciplinary Investigations of an Early
Archaic Florida Cemetery. University Press of Florida,

Doran, Glen H., and David N. Dickel
2002 Wellpointing and Excavation Procedures: How to Dewater
a Site. In Windover: Multidisciplinary Investigations ofan


2003 VoL. 56(1)


Early Archaic Florida Cemetery, edited by G. Doran, pp.
288-305. University Press of Florida, Gainesville.

Fairbanks, Charles H.
1959 Additional Elliot's Point Complex Sites. The Florida
Anthropologist 12:95-100.

Fradkin, Arlene
1994 Prehistoric Estuarine Adaptations in the Lower
Apalachicola River Valley: The Archaeological Faunal
Assemblages from Van Horn Creek and Sam's Cutoff Shell
Mounds. Report submitted to the Dept. of Anthropology,
University of South Florida. Environmental Archaeology
Range, Florida Museum of Natural History, Gainesville.

Gibson, Jon L.
2000 The AncientMounds ofPoverty Point. University Press of
Florida, Gainesville.

Henefield, Susan M., and Nancy Marie White
1986 Archaeological Survey in the Middle and Lower
Apalachicola Valley, Northwest Florida, 1985. Report to
the Division of Archives, History and Records Manage-
ment, Tallahassee.

Hunter, Donald G.
1975 Functional Analyses of Poverty Point Clay Objects. The
Florida Anthropologist 28:58-71.

Iscan, M.Y., and P. Miller-Shavitz
1983 Determination of Sex from the Tibia. American Journal of
Physical Anthropology 64:53-57.

Jones, B. Calvin
1993 The Late Archaic Elliott's Point Complex in Northwest
Florida. Paper presented at the Florida Anthropological
Society Meeting, Clearwater, May 1993.

Larsen, C. S., and C.B. Ruff
1991 Biomechanical Adaptation and Behavior on the Prehistoric
Georgia Coast. In What Mean Those Bones? Studies in
Southeastern Bioarchaeology, edited by M. L. Powell, P.
Bridges, andA. Mires, pp. 102-113. University ofAlabama
Press, Tuscaloosa.

Larsen, Clark Spencer
1980 Dental Caries: Experimental and Biocultural Evidence. In
Skeletal Biology ofAboriginal Populations in the South-
eastern United States, edited by P. Willey and F. Smith.
Tennessee Anthropological Association Miscellaneous
Paper No. 5. Knoxville.

Lazarus, William C.
1958 A Poverty Point Complex in Florida. The Florida Anthro-
pologist 11:23-32.

Livingston, Robert J.
1984 TheEcology oftheApalachicolaBaySystem:An Estuarine
Profile. Report to the Fish and Wildlife Service, U.S.
Department of the Interior, Washington, D.C. Department
of Biological Science, Florida State University, Tallahas-

McGee, Ray M.
1995 Fired-Clay Objects: Cooking Technology from the Florida
Shell Mound Archaic at Groves Orange Midden (8-VO-
2601) Florida. Unpublished M.A. Thesis, Department of
Anthropology, University of Florida, Gainesville.

McGee, Ray M., and Ryan G. Wheeler
1994 Stratigraphic Excavations at Groves' Orange Midden, Lake
Monroe, Volusia County, Florida: Methodology and
Results. The Florida Anthropologist 47: 333-349.

Mizoguchi, Yuji
1985 Shovelling: A Statistical Analysis of its Morphology.
University of Tokyo Press, Tokyo.

Molnar, S.
1972 Tooth Wear and Culture. Current Anthropology

Neuman, Robert W.
1976 Archaeological Techniques in the Louisiana Coastal
Region. Louisiana Archaeology. Bulletin of the Louisiana
Archaeological Society No. 3:1-22.

Purdy, Barbara A. (editor)
1988 Wet Site Archaeology. Telford Press, Caldwell, New

Stuiver, M., and P. J. Reimer
2000 Calib 4.3 (available online). Radiocarbon 35:215-230.

Thomas, Prentice M., and L. Janice Campbell
1991 The Elliott's Point Complex: New Data Regarding the
Localized Poverty Point Expression on the Northwest
Florida Gulf Coast, 200 B.C.-500 B.C. In The Poverty
Point Culture, Local Manifestations, Subsistence Prac-
tices, and Trade Networks, edited by K. Byrd, pp. 103-120.
Geoscience and Man 29. Louisiana State University, Baton

Tiffany, William J., III, Robert E. Pelham, and Frank W. Howell
1980 Hyperostoses in Florida Fossil Fishes. Florida Scientist

Trickey, E. Bruce, and Nicholas H. Holmes, Jr.
1971 A Chronological Framework for the Mobile Bay Region.
Journal ofAlabama Archaeology 17 (2):115-128.

Ubelaker, Douglas H.
1989 Human Skeletal Remains: Excavation, Analysis, Interpre-
tation. 2nd ed., Smithsonian Institution, Taraxacum Press,
Washington, D.C.

Waddell, Brett
1994 Personal communication concerning bone fractures from
University of Winnipeg, Manitoba, to Curtis Wienker,
USF, through SMTP Mail, Internet.

Walker, Karen J.
1994 Faunal Remains from Apalachicola River Sites: Depot
Creek and Van Horn Creek Shell Mounds (8Gu56,
8Fr744). In Archaeological Investigations at Six Sites in
the Apalachicola Valley, NorthwestFlorida, by N. White,




Appendix 1A, pp. 227-256. National Oceanic and Atmospheric printed 1999 by University Press of Florida, Gainesville.
Administration Technical Memorandum NOS SRD 26. Department
of Commerce, Washington, D.C.

Webb, Clarence HH. . ‘
1977 The Poverty Point Culture. Geoscience and Man 18.
Louisiana State University, Baton Rouge.

Webb, Clarence H., and Jon L. Gibson
1991 Studies of the Microflint Industry at Poverty Point Site.
Geoscience and Man 22:85-101. Louisiana State Univer-
sity, Baton Rouge.

Webb, William S.

1946 Indian Knoll, Site Oh 2, Ohio County, Kentucky. The
University of Kentucky Reports in Anthropology and
Archaeology, Vol 4, No. 3, Pt. 1. Lexington.

Weill, Lorna A., and Nancy Marie White
1994 Swamp Muck, Helicopters, and a Puzzling Body in the
Wilderness:. 1993 Excavations at Apalachicola Shell
Mounds. Paper presented at the Florida Anthropological
Society meeting, Fort Lauderdale.

Wheeler, Ryan J., and Ray M. McGee
1994 Technology of Mount Taylor Period Occupation, Groves’
Orange Midden (8VO2601), Volusia County, Florida. The
Florida Anthropologist 47:350-379.

White, Nancy Marie
1994a Archaeological Investigations at Six Sites in the
Apalachicola River Valley. National Oceanic and Atmo-
spheric Administration Technical Memorandum NOS SRD
26. Department of Commerce, Washington, D.C.

1994b Apalachicola Shell Mound Excavations. Report to the
Division of Historical Resources, Florida Department of
State, Tallahassee. Department of Anthropology, Univer-
sity of South Florida, Tampa.

1987 = Shell Mounds of the Lower Apalachicola Valley, North-
west Florida. The Florida Anthropologist 40: 170-74.

1983. Archaeological Investigations in the Upper Tombighee
Valley, Mississippi, Phase II. Report to the U.S. Army
Corps of Engineers, Mobile District. University of West
Florida Office of Cultural and Archaeological Research
Report of Investigations No. 4, Pensacola.

1982 The Curlee Site (8Ja7) and Fort Walton Development in
the Apalachicola-Lower Chattahoochee Valley, Northwest
Florida. Ph.D. dissertation, Department of Anthropology, |
Case Western Reserve University, Cleveland. University
Microfilms International, Ann Arbor.

White, Nancy Marie, and Richard W. Estabrook
1994 Sam’s Cutoff Shell Mound and the Late Archaic Elliott’s
Point Complex in the Apalachicola Delta, Northwest
Florida. The Florida Anthropologist 47-61-78.

Willey, Gordon R.
1949 . Archaeology of the Florida Gulf Coast. Smithsonian -
Miscellaneous Collections 113. Washington, D.C. Re-

Appendix A. Faunal remains from Van Horn Creek shell mound, Test Unit 6, Levell, Fort Walton component (with some
mixed materials from earlier deposits; weights in grams).

TAXON Waterscreen (1x2x.15m= Flotation (9 liters of soil)
300 liters of soil)
NISP* MNI** Weight NISP MNI* % Weight %

SyMlagus spp. 4 1 1.2
Sigmodon hispidus 1 1 0.2 8 1 0.6 0.6
eofiber alleni 9 1 1.1
Rodentia 5 0.4 6 1 0.6 0.4
Odocoleus virginianus 1 1 0.7
Small Mammalia 1 0.1
Large Mammalia 7 8.7
Kinostemidae 1 1 0.1
Cf. Terrapene carolina 1 1 0.6
Emydidae 21 1 37.6
Tionyx ferox 1 1 0.7
Testudines 38 8.29 2 1 0.6 0.3
Crotalus spp. 9 1 11.1
Serpentes 2 0.2
Alligatormississippiensis 24 2 21.6 1 1 0.6 4.2 0.2
Tetrapoda 4 0.1 3 0.2
Lepisosteus spp. 40 2 8.7 22 1 0.6 0.7
Amia calva 1 1 0.2
Elops Saurus 2 1 0.1
Adus fells 4 1 1.2
Bagre marinus 1 1 0.1
Ariidae 10 1 1.6 1 1 0.6 <.1
Cf. Aridae 1 0.2
Caranx hippos 1 1 0.7
Caranxsp. 1 1 0.2
Archosargus probatocephalus 14 3 10.5 6 1 0.6 0.3
Cf. Archosargus probatocephalus 2 1.1
Pogonias cromis 1 1 0.1
Sciaenops ocellatus 2 2 2
mugl sp. 1 1 0.2
Sparidae/Sciaenidae 3 <.1
Perciformes 16 9.2
Osteichthyes 103 12.7 113 1.7 0.1
Vertebrata 168 33.6 326 6.8 0.3
TOTAL VERTEBRATA 497 27 174.7 491 7 4.1 15.2 0.6

Crassostrea virginica frags 8 4.7 425.8 17
Rangia cuneata 296 115 67.7 1380 55.2
Polymesoda caroliniana 25 12 7:1 43 1.7
Bivalvia frags 610 24.4
Oligyra orbiculata 11 11 6.5 0.2
Euglandina rosea 8 2 1.2 1.3 0.1
lyphyalinia indentata 1 1 0.6 <.1
Zonitoides arboreus 10 10 5.9 0.1
Polygyra pustula 4 4 2.4 0.1
Pomacea paludosa 1 1 1.5
Mollusca frags 23.1 0.9
TOTAL INVERTEBRATA" 1 1 1.5 163 95.9 2483.6 99.4

TOTALS 498 28 176.2 170 100 2498.8 100
* Minimum number of individual specimens; ** Minimum number of individual animals; not included are higher level taxa (class, order, family) unless
clearly representing additional individuals; also not included are barnacles; *** Most shells not saved from waterscreened levels; flotation samples give
representative frequencies and percentages



Appendix B. Faunal remains from Van Horn Creek shell mound, Test Unit 6, Level 3, Late Archaic component (weights
in grams).

Taxon Waterscreen Flotation (9 liters of soil)
(1x2x.15m = 300 liters
of soil)
NISP* MNl* Weight NISP* MNI" % Weight %
Kinostemidae 2 1 0.4
Terrapene carolina 1 1 1.9
Tionyx ferox 1 1 0.3 0.4
Pseudemys floridana 6 1 4.7
Emydidae 1 0.5
Lepisosteus spp. 16 2 2.4 6 1 0.3 0.4
Sops saurus 1 1 0.3 0.1
Arus felis 10 1 1.6 3 2 0.6 0.3
Bagre marinus 1 1 0.2
Ariidae 7 1 2.8 2 0.1
Caranx hippos 1 1 0.6
Carangidae 1 0.3
Archosargus 5 3 1.5
Sciaenops ocellatus 1 1 0.3 0.3
Sparidae 2 0.4
Perciformes 7 4 2 1 0.3 1.1
Osteichthyes 38 6.5 87 3.6 0.1
Vertebrata 22 1.8 34 1.6
TOTAL VERTEBRATA 120 12 29.5 137 7 2 7.8 0.2

Balanidae 1 4.7 274 13.6 0.4
Ischadium recurvum 2 1 0.1 629 7 2 26 0.7
Crassostrea virginica 2 1 12.9 frags 82 23.1 3605 97.5
Rangia cuneata 7 5 1.4 27.8 0.7
Bivalvia 1 0.4 21 9.3 0.2
Oligyra orbiculata 167 162 45.6 6.6 0.2
Nassarius acutus 1 1 0.3 <.1
Detracia sp. 1 1 0.3 <.1
Gastrocopta pellucida 2 2 0.6 <.1
Pupillidae 1 1 0.3 <.1
Euglandina rosea 1 1 0.3 <.1
Helicodiscus parallels 38 38 10.7 0.2
Glyphyalinia indentate 19 19 5.4 0.1
Nesovitrea dalliana 2 2 0.6 <.1
Zonitoides arboreus 2 2 0.6 <.1
Polygyra cereolus 25 25 7 0.8
TOTAL INVERTEBRATA"* 6 2 18 348 98 3689.6 99.8

TOTALS 126 14 47.5 355 100 3697.3 100

* Minimum number of individual specimens
Minimum numberof individual animals; not included are higherleveltaxa (class, order, family) unless clearly representing additional individuals;
also not included are barnacles
Most shells not saved from waterscreened levels; flotation samples give representative frequencies and percentages


2003 VOL. 56(1)

Appendix C. Faunal remains from Van Horn Creek shell mound, Test Unit 6, Level 5, Late Archaic component (weights
in grams).

raxon Waterscreen Flotation (9 liters of soil)
(1x2x.15m= 300 liters
of soil)
NISP* MNI** Weight NISP* MNI** % Weight %
docoileus virginianus 1 1 1.20
kinostemon sp. 1 1 0.44
inosternidae 1 0.26
f. Terrapene Carolina 1 1 0.77
episosteus spp. 9 1 1.05 16 1 0.7 0.5
ius felis 5 3 1.37 5 2 1.4 1.4
,riidae 3 0.3
agre matinus 6 1 4.2
,riidae 2 0.3
Archosargus 3 2 0.8 2 1 0.7 <.1
icropogonias undulates 2 1 0.3 2 2 1.4 0.5
Iugil sp. 1 1 0.1 1 1 0.7 <.1
erciformes 2 1 1.2
steichthyes 18 1.8 71 1.8
ertebrata 22 2.4 47 1.4
FOTALVERTEBRATA 74 13 16.1 147 7 4.9 6 0.1

3alanidae 1 0.1 204 6.1 0.1
schadium recurvum 3 1 0.8 414 15 10.4 15.6 0.4
1ytilidae 1 <.1
3rassostreavirginica 2 1 6.9 frags 63 43.8 4180.1 98.3
angia cuneata 7 8 4.2 18.6 0.4
olymesoda caroliniana 4 2 1.4 .3.1 0.1
artesia striata 9 7 4.9 0.1
3ivalvia 44 23.1 0.5
ligyra orbiculata 21 18 12.5 0.7
omacea paludosa 1 1 0.7 0.8
uglandina rosea 1 1 <.1
hysella sp. 1 1 0.7 <.1
elicodiscus inermis 2 2 1.4 <.1
elicodiscus parallels 11 11 7.7 <.1
awaiia miniscula 1 1 0.7 <.1
.onitoides arboreus 7 7 4.7 0.1
olygyra cereolus 3 3 2.1 0.1
IOTAL INVERTEBRATA*** 7 3 7.9 137 95.1 4248.3 99.9

'OTALS 81 16 23.99 144 100 4254.3 100

Minimum number of individual specimens
** Minimum number of individual animals; not included are higher level taxa (class, order, family) unless clearly representing additional individuals;
also not included are barnacles
Most shells not saved from waterscreened levels; flotation samples give representative frequencies and percentages



Appendix D. Faunal remains from Van Horn Creek shell mound, Test Unit 6, Level 7, Late Archaic component (weights
in grams).

TAXON Flotation (9 liters of soil)
Waterscreen (1x2x.15m=
300 liters of soil)
NISP* MNI* Weight NISP* MNI** % Weight %
Large Mammalia 1 1 3.1
Chen caerulescens 1 1 0.5
Alligator mississippiensis 1 1 0.4
episosteus spp. 19 1 1.3 10 1 0.8 0.1
rius felis 3 2 0.9 2 1 0.8 0.2
Bagre marinus 5 1 0.8
Aridae 11 1 1.1 1 0.1
Archosargus probatocephalus 1 1 0.8 >.1
Caranx hippos 1 1 19.9
Cf. Caranx hippos 4 14.7
Cynoscion nebulosus 1 1 0.4
Cynoscion sp. 1 0.2
Sciaenops ocellatus 1 1 0.1
Sparidae/Sciaenidae 1 0.2
Mugilspp. 2 2 0.3 1 1 0.8 0.1
Perciformes 15 10.6
Osteichthyes 111 7.9 105 1.5
Vertebrata 136 7.3 50 1
TOTAL VERTEBRATA 314 13 69.5 170 4 3.3 2.9 0.1

Balanidae 249 2.8 0.1
schadium recurvum 3 1 0.7 308 5 4.1 7.7 0.2
Mytilidae 3 0.1 3 0.1
Crassostrea virginica 20 6 232.7 frags 51 42.5 3365 99.3
Rangia cuneata 1 1 0.8 8 0.2
Martesia striata 1 1 0.8 <.1
Bivalvia 1 0.3 1 0.3
Neritina reclivata 1 1 0.3
Oligyra orbiculata 19 19 0.9 5 5 4.2 0.3
Busycon contrarium (artifact?) 1 1 134.5
Hydrobiidae I 1 1 0.8 <.1
Nassarius acutus _1 1 0.8 <.1
Odostomia impressa 17 17 14.2 0.1
Gastrocopta pellucida 1 1 0.8 <.1
Helicodiscus intermis 4 4 3.3 <.1
Helicodiscus parallels 6 6 5 <:1
Euconulus chersinus 1 1 0.8 <.1
Glyphyalina indentata 5 5 4.2 <.1
Glyphyalina spp. 2 2 1.7 <.1
Hawaiia miniscula 11 11 9.2 <.1
Zonitoides arboreus 3 3 2.5 <.1
Euglandina rosea 1 1 0.3
Polygyra cereolus 10 4 0.2 1 1 0.8 <.I
Mollusca 2 0.1
TOTAL INVERTEBRATA"* 61 33 370.1 116 96 96 3384.3 99.9

TOTALS 375 46 439.5 120 100 3387.2 100

Minimum number of individual specimens
Minimum number of individual animals; not included are higherlevel taxa (class, order, family) unless clearly representing additional individuals;
also not included are barnacles
Most shells not saved from waterscreened levels; flotation samples give representative frequencies and percentages


2003 VOL. 56(1)

Appendix E. Faunal remains from Van Horn Creek shell mound,Test Unit 6, Level 9 (Preceramic?) Late Archaic
component (flotation sample only, 9 liters of soil, below water table; weights in grams).

TAXON NISP MNI** % Weight %

Tetrapoda 2 1 0.4 0.2
Lepisosteus spp. 27 1 0.4 0.4
Anus felis 3 2 0.8 2.4
Bagre marnus 2 1 0.4 0.2
Ariidae 1 0.4
Archosargus 8 1 0.4 1.5
Pogonias cromis 1 1 0.4 0.1
Sparidae/Sciaenidae 1 <.1
Osteichthyes 349 4.4 0.1
Vertebrata 477 5.2 0.1
TOTAL VERTEBRATA 871 7 2.7 14.8 0.3

Balanidae 23 0.2
Ischadium recurvum 16 1 0.4 0.4
Crassostrea virginica frags 51 20 2510 52
Rangia cuneata 264 137 53 1770 36
Polymesoda caroliniana 25 11 4.3 29.6 0.6
Mercenaria sp. 1 1 0.4 8.5 0.2
Bivalvia frags 540 11
Neritina reclivata 1 1 0.4 <.1
Oligyra orbiculata 19 14 5.4 0.3
Odostomia impresssa 4 4 1.6 <.1
Helicodiscus inermis 3 3 1.2 <.1
Helicodiscus parallels 5 5 1.9 <.1
Glyphyalinia indentata 2 2 0.8 <.1
Hawaiia minuscule 9 9 3.5 <.1
Zonitoides arboreus 4 4 1.6 <.1
Polygyra cereolus 14 8 3.1 0.2
TOTAL INVERTEBRATA*** 251 97 4859 100

TOTALS 258 100 4874 100

* Minimum number of individual specimens
* Minimum number of individual animals; not included are higher level taxa (class, order, family) unless clearly representing additional individuals;
also not included are barnacles
Most shells not saved from waterscreened levels; flotation samples give representative frequencies and percentages



Appendix F. Faunal remains from Sam's Cutoff shell mound, Test Unit 1 (no flotation sample taken from Level 1; weights in grams).

Level 1 Level 2 Level 3
Waterscreen (1x x.15m Watecreen(x.15m Flotaton (9 rsof Waterscreen (x1x.15m Flotation (9 liters of soil)
Rodentia 1 1 0.8 <.1
Tetrapoda 7 0.2
Serpentes 1 1 0.1
Lepisosteus spp. 9 2 2.08 8 1 0.8 0.5 5 1 0.5 8 1 0.6 0.2
Aridae 1 1 0.6 0.1
Arus Felis 1 1 0.1
Baare marinus 1 1 0.71__
Archosarus _1 1 0.8 <.1
Micropoaonias undulatus 2 2 1.2 0.5
Perciformes 1 1 0.6 0.9
Vertebrata 24 0.7
TOTAL VERTEBRATA 10 3 2.79 17 3 2.4 0.7 7 3 0.7 36 5 3.1 2.4 0.1

Balanidae 1 <.1 7 0.2 22 1.6 0.1
Ischadium recurvum 1 1 <.1 2 1 0.01 258 5 4 4.6 0.2 7 2 0.7 425 1 0.6 9.3 0.3
Mytilidae 2 <.1
Crassostrea virainica 8 8 370.7 16 11 528.60 fraqs 68 54.8 2415 99.7 12 8 403.4 fraq 95 57.9 3240 99.4
Rangia cuneata 1 1 4.6 1 1 5.97 6 5 76.4
Polymesoda caroliniana _1 1 3.7 2 1 0.6 4.2 0.1
Pholadidae_ 1 <.1
Martesia striata 2 1 0.8 <.1 1 1 0.6 <.1
BivaMa 2 1 2 2.5 0.1
Neritina reclivata 1 1 0.33
Olivqra orbiculata 3 3 0.12 5 3 2.4 0.2 7 6 3.7 0.2
Columbellidae 1 1 0.6 <1
Odostomia imressa 1 1 0.8 <.1 1 1 0.6 <.1
Detracia floridana_ 1 1 0.6 <.1
Melampus bidentatus 1 1 0.8 0.1 1 1 0.6 0.1
Gastrocopta pellucida _2 2 1.2 <.1
Helicodiscus parallelus 18 18 14.5 0.1 14 14 8.5 0.1
Glyphyalinia sp. __2 2 1.2 <.1
Hawaiia minuscule 9 9 7.3 <.1 15 15 9.2 <.1
Nesovitrea dalliana 2 2 1.2 <.1
Striatura meridionalis 1 1 0.6 <.1
Polyvyra cereolus 8 8 0.3 24 15 12.1 0.6 24 15 9.2 <.1
Mollusca 1 0.15
TOTAL INVERTEBRATA*** 13 10 375.4 32 25 535.48 121 97.6 2421.8 99.9 26 16 484.2 159 96.9 3258.6 99.9

TOTALS 13 In 375 4 42 28 538 27 124 100 2422. 6 10 33 19 Q 484 l 164 itnn 3 i n
* Minimum number of individual specimens
** Minimum number of individual animals; not included are higher level taxa (class, order, family) unless clearly representing additional individuals; also not included are barnacles
Most shells not saved from waterscreened levels; flotation samples give representative frequencies and percentages

Appendix G. Faunal remains from Sam's Cutoff shell mound Test Unit 3 (no flotation sample taken from Level 1; weights in grams).

Level 1 Level 2 Level 3
Waterscreen (lx1x.15m Waterscreen Flotation (9 liters of soil) Waterscreen (1xlx.15m Flotation (9 liters of soil
Odocoileus virqinianus 1 1 0.3
Kinostemon sp. 1 1 0.4
Kinostemidae 2 0.4
Testudines 2 0.6 1 1 0.4 0.1
Lepisosteus spp. 10 1 0.6 11 1 0.4 0.3
Aridae 4 3 1.9 1 1 0.1
Aius felis 1 1 2 0.1 1 1 0.2 1 1 0.4 0.1
Baqre matinus _8 1 3.4 2 1 1.1 2 1 0.4 0.2
Micropooonias undulatus
Perciformes 3 1 0.8 2 1 0.4 0.4
Osteichthes _1 0.1 18 0.6
Vertebrata _6 0.3 16 0.6
TOTAL VERTEBRATA 38 8 8.6 1 1 2 0.1 3 2 1.3 52 5 2.1 2.2

Balanidae 2 0.6 3 <.1 41 4 0.1
Ischadium recurvum 1 1 <.1 2 1 0.2 64 1 2 0.9 0.1 1 1 0.2 436 2 0.8 14.8 0.3
Mytilidae 8 0.1
Crassostrea virginica 15 9 367.5 96 18 729.1 fraqs 35 68.6 895 99.7 7 5 158.8 frags 132 55.1 4710 99.3
Ranqia cuneata 1 1 6.9 3 3 20.8 3 3 33.3 1 1 0.4 8.2 0.2
Polymesoda caroliniana 2 2 6.3
Martesia striata 3 2 3.9 <.1
Bivalvia 11 5.6 4 1.6 fraqs 1.2 0.1
Neritina reclivata 3 3 2.4 3 3 1.2 1.3
Oliqyra orbiculata 3 3 0.2 3 1 2 <.1 15 13 5.4 0.6
Nassarius acutus 1 1 0.1
Odostomia impress 5 5 2.1 <.1
Melampus bidentatus 1 1 0.1 1 1 2 <.1 1 1 0.4 <.1
Ellobiidae 1 1 0.4 <.1
Physella sp. 1 1 0.4 <.1
Gastrocopta pellucida 1 1 0.4 <.
Helicodiscus parallelus 5__ 5 9.8 0.1 21 21 8.6 0.1
Hawaiia minuscule 29 29 11.9 0.1
Nesovitrea dalliana 1 1 0.4 <.1
Polyqyra cereolus 10 10 0.3 10 5 9.8 0.1 2 2 0.1 38 25 10.3 1
Mollusca 1 0.2 238 97.9 4740.1 99.9
TOTAL INVERTEBRATA"* 28 11 380 128 42 761.9 50 98 897.6 99.9 13 11 192.4 243 100 4742.3 100

TOTALS 28 11 380 166 50 770-5 51 100 8977 1O0 16 13 193-7

Minimum number of individual specimens
Minimum number of individual animals; not included are higher level taxa (class, order, family) unless clearly representing additional individuals; also not included are barnacles
*Most shells not saved from waterscreened levels; flotation samples give representative frequencies and percentages

Appendix H. Faunal remains from Sam's Cutoff shell mound, TU5, NE and SE 1 x 1 m squares (no flotation sample taken from Levels 3 and 4; weights in grams).

Level 1 Lel 2 L 2 Level 3 Level 4
SWaterscreen(1x2x.15 Flotation (9liters of soil) Wate screen x2x.15 Flotation (9 liters o soil) Waterscreen (lx2x.15m Waterscreen
Svlwaqus spp. 3 1 1.1
Rodentia 1 1 0.1
Procon lotor 1 1 0.2 1 1 0.3 1 1 2 0.1
Mammalia 1 1 17 <.1 4 2.7
Kinosternidae 4 1 0.6
Emvdidae 1 1 0.3
Testudines 1 10 2.2 1 1 0.3
Tetrapoda 1 <.1 1 <.1
Colubridae 1 1 0.1
Serpentes 3 1 1.3 1 1 0.1
Alliaator mississippiensis 5 1 25.8 5 1 4.3
Lepisosteus cf. oculatus 2 2 1
Lepisosteus spp. 1 1 0.1 4 1 17 <.1 0.1 138_ 19.2 25 1 2.2 0.9 0.1 116 1 10.7
Arius felis 10 2 1.9 5 1 0.9
Ariidae 8 3 4.1 2 0.4
Bare marinusI 1 1 0.1
Caranx hippos 6 1 7.9 6 2 18.2 5 2 7.8
Cf. Caranx hippos 2 4.5 5 18.9
Caranx so. 1 0.5 1 0.8
Cf. Caranqidae 1 1.1 9 8.9 4 9
Archosarus probatocephalus 2 1 17 0.1 0.1 3 2 1.1 1 1 2.2 0.1 4 1 0.3 1 1 0.1
Cf. Archosaraus 1 1
Micropoqonias undulatus 1 1 0.2
Sparidae/Scienidae 1 <.1
Sciaenops ocellatus 2 1 0.3
Perciformes 1 1 0.3 12 11.1 17 19.5 4 7.4
Osteichthyes J 60 7.1 3_ 0.1 39 6.2 4 1.3
Pdisds sp. 1 1 <.1
Vertebrata 9 0.1 0.1 108 15.2 23 0.5 91 7.1 7 2.3
TOTAI VFRTFRRATA 3 3 0 6 16 3 I50 10 2 0 3 387 19 110j 4 55 3 i 7 1 0 l 1 3as in I Q i .9r 1 4 1

Appendix H, continued.

Waterscreen(1x2x.15 Flotation (9 liters of soil) Watescreen 1x2x.15 Flotation (9 liters of soil) Waterscreein (1x2x.15m Waterscreen
TAXON NISP* MNI** Wt. NIS MNI** % Wt. % NIS MNI** Wt. NIS MNI* % Wt. % NISP* MNI** Wt. NISP* MNI** Wt.
Balanidae 9 0.1 1 1.8
Ischadium recurvum _1 1 0.1 3 1 2.2 <.1 1 1 0.2
Crassostrea virqinica 3 2 108.3 frags 3 50 55.3 99.7 50 19 726.2 fraqs 32 71.1 1175 99.8 13 4 40 9 7 242.5
Ranqia cuneata 1 1 12 2 2 38.3
Polymesoda caroliniana 1 1 6.2 -
Macrocallista nimbosa 1 1 3.8 -
BivaMa 1 0.2
Martesia striata 2 2 4.4 <.1
Neridna reclivata 3 3 2.6 5 5 5.4 1 1 0.4
Helicodiscus parallels 2 2 4.4 <.1
Hawaia miniscula 2 2 4.4 <.1
Oliqyra orbiculata 1 1 0.1 2 2 0.1
Busycon contrarium 1 1 8.8 1 1 46.8
Melampus bidentatus 1 1 0.1 3 2 0.1
Euqlandina rose 1 1 0.2
Polyqyra cereolus 16 15 0.5 3 3 6.7 <.1 20 17 0.5
AZtmrnriv 1 1 1
Mollusca 1 1.2 2 <.1
TOTAL INVERTEBRATA** 3 2 108.3 3 50 55.3 99.7 76 42 740.8 42 100 1175. 99.8 51 34 117 13 111 281.4

TOTALS 6 5 1108g 6 10 55 4 150 463 61 851 1 45 100 1176 7 10 356 44 1 214 39 1 15 309.35

*Minimum number of individual specimens
Minimum number of individual animals; not included are higher level taxa (class, order, family) unless clearly representing additional individuals; also not included are barnacles
*** Most shells not saved from waterscreened levels; flotation samples give representative frequencies and percentages


An Endowment to

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

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

Inquiries and gifts 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.



1 Department ofAnthropology, University ofMiami, P.O. Box 248106, Coral Gables, FL 33124

2 P.O. Box 20, Terra Ceia Island, Florida 34250

SStarbridge Chiropractic, Murrayville, Georgia 30564

In this report we present results of three seasons of archae-
ological research at the Mystery River Point archaeological site
(8S011), a large shell midden in the France Environmental
Research Area within Lemon Bay Park in Englewood, Florida.
This research was conducted by the authors and several
students from New College ofFlorida, in January 1985, 1987,
and 1989, under direction of the second author. During the
first season, a detailed survey and surface collection of the site
were conducted and topographic maps of the main features of
the site, its vegetation, and artifact density distribution were
produced (Burger et al. 1986). During the second season, we
excavated two 1.5 x 1.5 m test units near the center of the
midden and spent several months analyzing the materials
recovered. The third season consisted of excavation of a third
1.5 meter square test unit on the midden (Burger et al. 1992).
The France Environmental Research Area is a 23 acre
nature preserve near Englewood, Florida, in Sarasota County.
In 1980, it was donated to New College by the Annita and
Jacob France Foundation of Baltimore, Maryland, for the
purpose of socially-significant academic and scientific study of
ecology and environment. The area lies along Lemon Bay at
the end of an unimproved road and consists of a freshwater
bayhead in pine flatwoods that drains into a small embayment
fronted by coastal shell midden. This natural community is
the home of several endangered and threatened species,
including coontie, indigo snake, gopher tortoise, scrubjay, and
bald eagle.
The archaeological site was originally designated "Mystery
River Point" by William Plowden in 1953, who recorded it
with the Archeological Site Survey of the University of
Florida, precursor to today's Florida Master Site File. He
described it as a shell midden located on the thirdid point
north of the Lemon Bay fisheries, on the south side of the
point jutting into the bay from the east side...just south of
Forked Creek." Plowden made a small surface collection of
ceramic material which is now in the collections of the Florida
Museum of Natural History, Gainesville (FLMNH cat.
Figure 1 shows that the site is located on the eastern side
of Lemon Bay, toward the north end of the bay and almost
directly across the water from the prehistoric Manasota Key
Cemetery (8S01292). The midden occupies most of a small
sand spit that effectively encloses the mouth of Plowden's

rather ostentatiously named "Mystery River." The "river" is
actually the drainage system of a freshwater bayhead located
in the adjacent pine flatwoods. One can hypothesize that in an
earlier period of slightly lower sea level this drainage was a
small creek, partially tidal, perhaps partially springfed, which
intermittently flowed to its largely enclosed mouth, there
providing optimal oyster habitat. With a rise in sea level, the
mouth silted in and mangrove vegetation became dominant.
In the late 1950s, Sarasota County extensively ditched the
clogged backwater area in an attempt at mosquito control.
Certain limited areas of the site were affected by this ditching.
Figure 2 shows the site to be a shell midden ridge extend-
ing approximately 265 m east-west, then 265 m southeast-
northwest, varying in width from 15-45 m. Elevation averages
about 0.75 m, with two areas rising to 1.5 m; the Mystery
River Point midden is a long and narrow site like many along
Lemon Bay (Luer 1999a:6). At the easternmost high point
scattered evidence of looting was discovered-- several large
holes had been dug into the mound and a small, palm-thatched
shelter set up under the vegetation. None of this activity
appeared to be recent. To the immediate north of the most
elevated areas of the midden lies a backwater area that was not
surveyed, but would have provided a very productive oyster
area during prehistoric times.

Research Orientation

The 1985 field survey provided preliminary ceramic and
artifactual evidence as well as the baseline topographic map of
the prehistoric midden. These surface collections indicated an
occupation during the Weeden Island Period (600-1000 C.E.),
while Plowden's ceramics collection, including examples of
semi-sand/semi-fiber tempered Norwood sherds, indicated the
presence of a much earlier occupation, dating to circa 1000-
500 B.C. (Plowden 1953; Burger et al. 1986:12). The excep-
tional temporal gap between the early materials found by
Plowden and those in the 1985 surface collection indicated the
possibility of a very long occupational sequence, representing
many of the ceramic traditions in this area of Florida.
Another research goal was our desire to clarify important
questions about contemporary models of the nature of sites at
the regional level (Milanich 1994). George Luer and Marion
Almy designated a regional adaptation within the Weeden


MARCH 2003

VOL. 56(1)



Figure 1. Lemon Bay and sites
mentioned in the text

River Point


A Mound

2 km
I -

Figure 1. Lemon Bay area with sites mentioned in the text.


2003 VOL. 56(1)


Island tradition as a distinctive cultural sub-area, named the
Manasota archaeological culture (Luer and Almy 1982).
According to their definition of cultural traits the Manasota
culture can be recognized as follows:

Archaeologically the Manasota culture is characterized
by...sites which yield evidence of an economy based on
fishing, hunting, and shellfish-gathering. The sites yield
evidence of burial practices involving primary, flexed
burials...Ceramic manufacture was limited to sand-tempered,
undecorated...pottery such as...flattened-globular bowls and
pots with a converged orifice. Many shell tools were
used...including fighting conch shell hammers, left-handed
whelk shell "spokeshaves"...collumellae and hammers...
there was little use of stone tools...Bone tools include barbs
and simple points made from longbones. From our fieldwork
and our analysis of published site reports, we realized that
the Manasota culture had a particular distribution in time;
that is, from about 500 B.C. to about A.D. 800. Our dating
of the culture was based primarily on stratigraphic evidence,
especially ceramic inventories and carbon-14 dates. [Luer
and Almy 1982:37]

Luer and Almy defined the geographic limits of the
Manasota culture as including an area from just north of
Tampa Bay, southward to the north edge of Charlotte Harbor.
If these limitations are correct, the Mystery River Point site
could be representative of the extreme southern limit of this
archaeological culture. If it deviated significantly from the
Manasota pattern, this too would shed light on the geographic
boundaries and/or border areas of the proposed culture.

The 1987 and 1989 Field Seasons

Two areas of the Mystery River Point midden rise to 1.5 m.
The westernmost of these areas is a long ridge, running
northwest-southeast. This ridge rises from the water, reaching
a maximum elevation of 1.5 m approximately 10 m from the
water, and then drops off sharply toward the land side. Two
1.5 m square test units were placed on the summit of this
ridge: Test Unit A was placed on the shore side of the ridge,
just below maximum elevation, and Test Unit B was placed
approximately 10 m to the north, at maximum elevation on the
ridge, closer to the land side. Test Unit C was placed approxi-
mately 165 m east of Test Unit A on the shore side of the
second, more eastern elevated ridge.
Excavation in all units revealed that the matrix of the
midden is mostly oyster shell, with considerable amounts of
scallop and a variety of other shell in varying amounts. The
midden is characterized by layers of essentially whole, air-
spaced shell, overlaying areas of fine to very fine diminutized
shell mixed with dark soil and some cultural and faunal
remains, which in turn overlay more whole air-spaced shell.
In Test Units A and B, this pattern was visible until the
watertable was reached at approximately 1.75 mbelow ground
surface. Problems with the pumps used to de-water the test
units necessitated the removal of an additional 70 cm of
cultural strata as single artificial levels. At approximately 2.5
m below ground level we encountered a layer of very dark

mud, overlying white sand, which we interpreted as culturally
sterile. In Test Unit C, the sterile white sand level was
approximately 70 cm higher than in Test Units A and B.
Within each cultural stratum we excavated in artificial 10 cm
levels, using wooden plywood sleeves to help prevent collapse
as the units progressed. Test Unit A revealed five distinguish-
able cultural strata, while Test Units B and C each revealed
four distinct strata.
The distinction of cultural strata in the matrix of the units
is somewhat speculative. Each is defined by an undulating
surface of fine diminutized shell intermixed with dark gray to
dark brown soil, and yielding a higher concentration of
cultural remains than the underlying whole air-spaced shell
areas. Below water level, separate depositions were distin-
guished by changes in the composition of the matrix. The
existence of cultural strata below the water table is interpreted
as evidence of a lower sea level during the period in which the
midden began.
A number of organic remains were recovered from the
lowest levels of the midden. These artifacts and ecofacts were
remarkably well preserved-- due to the anaerobic hydrogen
sulfide environment, fragments of charred stakes, carved
wood, pine bark, and acorns were recovered. Two samples of
wood from the basal deposits of muck, diminutized shell, and
sand which underlay the whole shell below water table in Test
Unit B were submitted for radiocarbon dating (Table 1). A
calibrated date of A.D. 425 +/- 50 years suggests the initial
occupation of the Mystery River Point midden was midway
through the fifth century. The midd6n appears to have been an
active living area for at least 200 years, because two samples
of charcoal from within overlying whole shell matrix in
separate test units were dated to a range of A.D. 550 +/- 50 to
A.D. 640 +/- 50 years (Table 1). Final occupation of the site,
from these data and a single sherd ofPinellas Plain ware from
30 cm below ground level in Test Unit B, is extrapolated to at
least A.D. 800. No later ceramic materials were discovered
during surface reconnaissance. Thus, our present evidence
suggests the Mystery River Point site was occupied for a period
of at least 400 years, between A.D. 400 and 800. The discov-
ery of datable anaerobic deposits is perhaps one of the most
significant contributions the Mystery River Point site has to
offer to our understanding of the prehistory of southern
Charcoal samples from within whole shell matrix were
recovered from Test Unit C, although they have not yet been
subjected to radiocarbon analysis. Because culturally sterile
white sand, or what we believe was the original beach sand
surface prior to occupation, was encountered 70 cm higher in
Test Unit C and above the present water table, we postulate
that this area of the midden predates the most elevated point
further west where Test Units A and B were located. It
appears the site began as an occupation on an existing sand
spit, then grew westward through a combination of midden
deposition and long shore drift of sand and sediment. Thus,
the unanalyzed charcoal samples may generate dates even
earlier than A.D. 425 that might reflect Plowden's Norwood





Lemon Bay

Figure 2. Mystery River Point

Mystery River Point
8 So 11
Englewood, Florida

New College Research Program
1985 -1989
Bill Burger, Principal Investigator

25 cm contour interval
0 50 m

Original Map by B. K. Sullivan

Figure 2. Contour map of Mystery River Point with excavation locations.


Table 1. Radiocarbon dates from Mystery River Point.

Lab # Material Context Depth Uncorrected 13C/PC Corrected One Sigma Two Sigma
UGa- and Unit Below Date Correction Date Calibrated Calibrated
Ground Range Range

5662 wood sand 2.22 m 1610 B.P. 50 -26.97 428 50 A.D. 406-534 A.D. 264-561
whole shell
water table,
Test Unit B

5663 wood sand 2.50 m 1610 B.P. 60 -28.22 428 60 A.D. 395-536 A.D. 261-599
whole shell
water table,
Test Unit B

5664 charcoal whole shell, 0.60 m 1510 B.P. 60 -27.77 549 60 A.D. 441-636 A.D. 419-657
Test Unit A

5665 charcoal whole shell, 0.30 m 1410 B.P. 50 -27.76 646 50 A.D. 604-662 A.D. 543-688
Test Unit B

Radiocarbon age determination analysis conducted by Center for Applied Isotope Studies, University of Georgia, April 1987. Calibration using 20-year
atmospheric record by University of Washington Quaternary Isotope Lab Radiocarbon Calibration Program Rev. 4.3.



The overwhelming majority of ceramic materials recovered
from all levels within the three units and the surface collection
were fragments of sand-tempered plain pottery (Table 2).
Projected vessel forms consist of simple open bowls and
globular pots with slightly converged orifices. Thickness
curves show medial body sherd thickness to average at 7.5
mm-8.5 mm, varying only slightly through time. Lip styles
are chiefly pointed/rounded, with a few flattened examples. A
single chamfered lip was recovered from the initial (earliest)
cultural deposition at Test Unit A.
Only very minor amounts of other ceramic types were
recovered. Sherd-temperedplain, an otherwise sand-tempered
ware with very rare inclusions of sherds or clay, occurred in all
units from initial to final occupations, but in very slight
amounts and is consistent with ceramics from other Manasota
sites such as the Roberts Bay site (8S056) (Luer 1977). Three
sherds of St. Johns Plain, a type having the same long tempo-
ral range as sand-tempered plain, were recovered from Test
Unit B in upper levels as well as in the controlled surface
collection. The presence of Belle Glade Plain sherds in the
surface collection suggests contact with the Lake Okeechobee
or Caloosahatchee regions late in the occupation of the site. A
single sherd ofPinellas Plain, which is diagnostic of post A.D.

800 occupation, was recovered from an upper level at Test
Unit B. We believe this isolated artifact may be indicative of
a later, Safety Harbor/early Mississippian occupation, some-
where in the vicinity of Mystery River Point, perhaps the
nearby Englewood Mound (8S01).
The Mystery River Point ceramic sample is consistent with
what is known for the Manasota culture. Luer and Almy
(1982:44) state:

One of the distinctive Manasota traits was the manufacture
of sand-tempered, undecorated flattened globular bowls and
pots with a converged orifice. These vessels are easily
recognized from distinctive inward-curving rim sherds with
chamfered or rounded lips. Sherds from these vessels are
often more than 1 cm in was the predominant
kind of ceramic made from about 300 BC to AD 700 along
the central peninsular Gulf Coast. From about 300 BC to
about AD 400 the pottery was especially thick and chamfered
lips were common. Gradually, around AD 400, pottery
became slightly thinner and only rounded lips were made.

None of the research undertaken by New College recovered
evidence of the very early occupation suggested by Plowden
(1953). Examinationby Burger of the 1953 surface collection
at the Florida Museum of Natural History showed that it does
contain a limited number of Norwood ceramic sherds (semi-
sand/semi-fiber tempered), which would indicate an occupa-
tion of 1000-500 B.C. Further excavation might recover



Table 2. Ceramic counts from Mystery River Point.

evidence of an initial Norwood component.

Subsistence Activities and Seasonality

Some of the most useful information derived from limited
excavation in shell middens relates to variability in resource
exploitation. Significant understanding of site activity can be
gained through an examination offaunal remains and artifacts
reflecting subsistence activities and seasonality. Tables 3 and
4 present lists of the variety of species recovered at the Mystery
River Point site. This sample was obtained by screening all
matrix through 3 mm (1/8 in) mesh. In terms of minimum
number of individuals, our samples indicate a subsistence
strategy primarily based on the exploitation of marine re-
sources through fishing and the collection of shellfish,
augmented by hunted birds and mammals.
Massive shell midden deposits point to the importance of
shellfish in the ancient diet of the inhabitants of Lemon Bay
(Table 3). The predominant species found at Mystery River
Point, by gross volume of depositions, are oyster (Crassostrea
sp.), small conchs (Gastropoda spp.), scallop (Aequipecten
sp.), and quahog (Mercenaria sp.). The sheer mass of shell
debris obscures the equivalent importance ofvertebrate species
(Table 4). For example, while our deer remains are not
numerous, edible meat value for deer is high and non-meat
remains (bones, hides, sinew) represent an important source of
other useful raw materials. Preliminary analysis of skeletal

elements shows a predominance of limb bones, indicating deer
were likely butchered elsewhere, probably at kill sites, with
primarily haunches brought back to living areas.
There are several potential seasonality indicators present
in the data. The occurrence of sea turtle bone and shell
fragments suggests an occupation during the late summer
months, when the animals come ashore to lay eggs and would
be easier to kill. The presence of duck, and especially loon,
seems to indicate an autumn through spring presence at the
site. Clearly, the site was utilized as a resource base at various
seasons, although explicit evidence for a year-round occupa-
tion was not found.
The presence of large spotted grouper vertebrae in the
sample raises a number of questions. Did the ancient inhabit-
ants of Lemon Bay have the technology to pursue deep water
fishing (as this large grouper is not accessible close to shore),
or is this an indicator of possible scavenging of the Gulf
beaches? The Manasota culture sites of Roberts Bay and
Venice Beach also have yielded large deep water fish remains
and evidence suggests the technology and skills necessary to
pursue deep water fishing were in use during this time (Fraser
1980:77; Luer 1977:130, 1999a:1,10).
An anomaly exists concerning the exploitation of shark at
the site. While a significant number of shark vertebrae were
found in the test units, we lack a comparable sample of shark
teeth. This could be related to several factors, including
sample size, or could be due to specific utilization:of shark

Sand Sherd Limestone Weeden Belle St. Lime- Pinellas Unid.
Temp. Temp. and Sherd Island Glade Johns stone Plain
Plain Temp. Plain Plain Plain Plain Temp.

Surface Collection 265 2 1 4 4 8 1

Unit A
Stratum A 42
Stratum B
Stratum C 28 4
Stratum D 80 1
Stratum E 85 1 1
19 1 1____ _
Total 254 7 1 1

Unit B
Stratum A 136 7 3 1
Stratum B 140 5 1
Stratum C 65
Stratum D 2
Total 343 12 1 4 1 1


2003 VOL. 56(1)


Table 3. Invertebrate faunal remains recovered
from excavations.

Blue (Callinectes sapidus)
Stone (Menippe mercenaria)
Apple murex (Murexpomum)
Atlantic moon snail (Polinices duplicatus)
Banded tulip (Fasciolaria hunteria)
Bay scallop (Aequipecten irradians)
Eastern oyster (Crassostrea virginica)
Crown conch (Melongena corona)
Fighting conch (Strombus alatus/pugilis)
Horse conch (Pleuroploca gigantea)
Giant Atlantic cockle (Dinocardium robustum)
Lettered olive (Oliva sayana)
Lightning whelk (Busycon contrarium)
Ponderus ark (Noetia ponderosa)
Prickly cockle (Trachycardium egmontianum)
Southern quohog (Mercenaria campechiensis)
True tulip (Fasciolaria tulipa)
Surf clam (Spisula solidissima)
Marsh clam (Polymesoda caroliniana)

teeth, either as knives/scrapers, trade items, or both (Kozuch
There is a noticeable lack of large Busycon whelk shells.
While the species is represented, no exceptionally large
examples, as can be found at other sites in the area, are evident
in the excavation sample or from the surface collection. This
may be a reflection of relative species absence due to lower
salinity levels caused by closing of Gulf passes to Lemon Bay
in prehistoric times, or to previous over harvesting and
deposition at other earlier sites elsewhere along the Bay.

Tools and Technology

The tool complex of the site is consistent with a primarily
marine resource oriented subsistence strategy. Shell tools
include Busycon percussive and cutting-edged, Pleuroploca
percussive, Strombus percussive, Noetia net weight, and
Mercenaria percussive and scraping tools. In addition to
eleven small points fashioned from deer bone splinters,
probably used as gorges or compound fish hook barbs, we
recovered one deer ulna awl. Lithic technology was very rare
in the Mystery River Point sample, with no grinding stones
present and only a single, small non-utilized flake of silicified
limestone found in the upper levels of Test Unit C. A natu-
rally bi-lobed piece of limestone with percussive scarring on
opposite ends was also recovered from the same unit. This
small sample may be due to the limited amount of excavation
carried out, but more probably reflects the absence of suitable

lithic material sources in southern Florida, as well as the
possibility that Mystery River Point had an exceptionally
specialized function focused on marine resource exploitation.
A number of recovered artifacts would best be described as
ornamental. These include four perforated scallop
(Aequipecten) valves too fragile for use as net weights, an
Oliva bead, a perforated deer phalange, and a flattened
pendant of fossilized bone. Identical scallop and Oliva beads
were found by Bullen at the Sarasota County Mound
(8S016/23) at Paulsen Point in Englewood, Florida (Bullen
1971:21). This nearby shell midden site overlaps in occupa-
tional history with Mystery River Point, and both middens
probably represent the remains of a single archaeological
culture whose members lived throughout Lemon Bay during
Manasota times.


Although limited, the research conducted by New College
of Florida in 1985, 1987, and 1989 at Mystery River Point
provides significant information about this ancient habitation
site on Lemon Bay. The size of the Mystery River Point
midden suggests it represents the remains of a moderate
village. The primary occupation of the site occurred between
A.D. 400-800 and may have been partially dependent upon
seasonally available resources. The estuarine environmentwas
intensively exploited for shellfish, fish, and sharks, with
capture techniques including hook, nets, and probably deep
water fishing. A fresh water source, so often absent from
coastal shell middens, was immediately adjacent to the site,
and the small creek known as "Mystery River," would have
helped create optimal salinity conditions for the growth of
oysters in its embayed mouth, as it does today. Easily gath-
ered, shellfish were percussively opened with shell tools.
The adjoining terrestrial environment also was exploited.
The surrounding catchment area of the site includes a mosaic
of pine flatwoods, hardwood hammocks, ponds, and fluvial
systems. Game was available and included mammals, espe-
ciallywhite-tailed deer, reptiles, birds, and amphibians. While
no definitively terrestrial hunting technology was recovered,
deer bone points could have been used as projectile points.
Other activities reflected in the archaeological record of the
site include wood working, meatprocessing, ceramic manufac-
ture, and plant food processing.
The rich environment of Lemon Bay appears to have
supported sizable populations in ancient times, of which the
Mystery River Point settlement was only one part (see Luer
1999a for a recent summary). Within a 10 km radius of
Mystery River Point, at least three other sites with Manasota
components have been excavated (Paulsen Point, Manasota
Key Cemetery, and the Cedar Point Shell Heap) that suggest
our site may have been part of flourishing community around
Lemon Bay (Bullen 1971; Dickel 1991; Luer 1999b; Willey
1949). The many middens in the Blind Pass area ofManasota
Key suggest even more occupation from this period may be
identified (Luer 1999a).




2003 VoL. 56(1)

Table 4. Vertebrate faunal remains recovered from


White-tailed deer (Odocoileus virginianus)

Opossum (Didelphis virginiana)

Cetacean (Tursiops sp.)?


Box (Terrapene carolina)
Chicken (Deirochelys reticularia)
Mud/Musk (Kinosternidae)

Rattlesnake (Crotalus adamanteus)

Alligator (Alligator mississippiensis)

Sea Turtle (Caretta sp.)?


Siren (Siren lacertina)


Common loon (Gavia immer)

Ducks: (Anatidae)

Owl (Tytonidae)

Jack (Caranx spp.)
Trout/Redfish/Drum (Sciaenidae)
Sea Catfish (Galeichthys felis)
Snook (Centropomus undecimalis)
Sheepshead (Archosargus probatocephalus)
Southern puffer (Sphaeroides nephelus)
Spiny boxfish (Chilomycterus schoepfi)
Spotted grouper (Stereolepsis gigas)
Toadfish (Opsanus spp.)
Flounder (Paralichthys lethostigma)


Lemon (Negaprion brevirostris)
Nurse (Ginglymostoma cirratum)
Blacknose (Carcharhinus acronotus)

Rays (Dasyatis sp.)?

Excavations at the Sarasota County Mound at Paulsen —

Point show that it was similar in many ways to the Mystery
River Point midden. While there was evidence of a longer
range of occupation at Paulsen Point (1000 B.C. - A.D. 1350),
there was a nearly identical artifactual and especially ceramic
inventory (Bullen 1971). Bullen had the opportunity to
analyze materials from more extensive excavations conducted
by the Sarasota County Historical Commission at 88016/23,
thus his research yielded a much larger and more representa-
tive data set, so it is not surprising that the range of occupation
at Paulsen Point is greater than currently documented for
Mystery River Point. The main occupation of Paulsen Point
seems to overlap with the primary occupation of Mystery River

Point around A.D. 400 (Bullen 1971:13). Bullen speculated |

that the relative lack of burials at Paulsen Point suggested the

Englewood burial mound (8SO1), excavated in 1934 by

Marshall Newman and reported by Gordon Willey in 1949,

was the focus of burial and perhaps ceremonial activity for the
inhabitants of the small settlements represented by midden
debris around Lemon Bay (Bullen 1971:28). Willey suggested
that the size, density, and number of burials within the
Englewood Mound supported its use over time by the inhabit-
ants of many village sites along the water, although the
Englewood Mound seems to post-date the main occupation of
Mystery River Point (Willey 1949:126). The Manasota Key
Cemetery (8801292), directly across the bay from Mystery
River Point, may be another location linked culturally with the
Mystery River Point midden peoples. The Manasota Key
Cemetery, where 120 adult burials were excavated in the late
1980s, had only four radiocarbon dates analyzed (Dickel
1991:2). The latest of these yielded a date of A.D. 325 +/- 80
years, which is contemporary with the occupation of Mystery
River Point (Dickel 1991:2). While we are relatively certain
that mounds served discrete functions later in time in this area,
itis not clear to what degree mounds were still multifunctional
at the time of the Mystery River Point settlement. Both
Paulsen Point (8S8O016/23) and Cedar Point Shell Heap
(8CH8/8CH61) were shell middens with artifactual material as
well as human burials, and given the very limited amount of
excavation at Mystery River Point, it is impossible to rule out
the possibility that human burials also were placed in this
midden. Luer and Almy (1982:46) have stated that over the
course of the Manasota Period, native peoples adopted the
practice of interring their dead in sand mounds dedicated to
this purpose, and certainly by Safety Harbor times (post A.D.
900) secondary burial in sand mounds predominates. Within
the Lemon Bay area, we have good evidence that this profound
change in the treatment of the dead happened gradually, since
there are a number of shell middens with primary burials that —
appear to be contemporary with dedicated sand burial locations
like the Manasota Key Cemetery.

Further study within the proposed Manasota Culture area
might supply much needed information about the arrangement
of settlement and the degree of social integration along Lemon
Bay. As defined by Luer and Almy (1982), the Manasota
Culture is a cohesive regional adaptation that extends from
Tampa Bay to Charlotte Harbor during the period of 500 B.C.
- A.D. 800, and the presently known settlement of Mystery
River Point falls within one phase of this very long time
period, the Weeden Island-related phase from A.D. 400-800.
There is substantial evidence that during this time throughout
the Central Peninsular Gulf Coast, a regional level of integra-
tion can be discerned in the archaeological record of many
coastal sites. The inventory of ceramic, artifactual, and faunal
data we recovered from the limited Mystery River Point
excavations is consistent with the patterns and evidence
defined as the Manasota Culture. In fact, it would be difficult
to understand the small Mystery River Point village outside a
regional pattern of subsistence and cultural practice. The rich
coastal environment of aboriginal Lemon Bay, as well as most
of the Central Gulf Coast, would have provided an ideal area
in which ancient peoples could have periodically moved camp,


traded with neighbors, and traveled reasonable distances for
ceremonial or economic obligations. Whether the numerous
settlements around Lemon Bay and the southern Manasota
area were held together by a common environmental adapta-
tion or more profound social and political mechanisms is a
question that remains to be answered by future research. Luer
has recently suggested that simple chiefdoms emerged in the
Lemon Bay area around A.D. 700 (Luer 1999a:3). Sites like
Mystery River Point represent the relative wealth or prosperity
that may have laid the groundwork for the emergence of
stratification. The sheer number of sites along Lemon Bay
from late Manasota times appears supportive of Widmer's
hypothesis that population growth and subsequent control of
resources was also a causal factor in the emergence of chief-
doms by circa A.D. 700 (Widmer 1988).
A great amount of research remains to be carried out at
Mystery River Point, along Lemon Bay, and within southern
Sarasota County. Important questions about the duration and
nature of the Manasota archaeological culture at its southern-
most extent could be tested at Mystery River Point. More
information about the ancient exploitation of the environment,
including the pine flatwoods and freshwater bayhead adjacent
to the midden, is plentiful at the site and an important resource
for future research. The anaerobic deposits that yielded
datable carbon samples hold the potential of more well
preserved organic material. Fortunately, Mystery River Point
is located within a perpetual easement of the Jacob and Annita
France Environmental Research Area, within Sarasota
County's LemonBay Park, and professional investigations will
be possible at the site for many years to come.


The authors would like to thank the New College Foundation for
generously sponsoring this research. We would also like to thank Dr.
Anthony P. Andrews of the Division of Social Sciences as well as
Julie Morris and Jonathan Miller of the New College Environmental
Studies Program for facilitation of the excavations. We also thank
the fine crew members from New College: J. Brandehoff, J. Hover,
C. Jacob, T. Lee, H. Mahan, M. Metcalf, M. Pool, L. Roggow, G.
Vail, P. Vaughn-Brown, S. Waldman, and S. Zatinsky for their
contribution to the project. Traci Ardren would like to thank George
Luer for editorial assistance with this publication and Michael C.
Owens for assistance with the figures.

References Cited

Bullen, Ripley P.
1971 The Sarasota County Mound, Englewood, Florida. The
Florida Anthropologist 24: 1-30.

Burger, Bill W., T. Ardren, T. Lee, B.K. Sullivan, and G. Vail
1986 Phase IInvestigations oftheMysteryRiverPointArchaeo-
logicalSite (8Soll) FranceEnvironmentalResearchArea,
Englewood, Florida. Report on file, Social Sciences
Division, New College, Sarasota.

Burger, Bill W., T. Ardren, and B. K. Sullivan
1992 Archaeological Investigations of the Mystery River Point
Site (8-So-11), Sarasota County, Englewood, Florida.

Manuscript on file, Social Sciences Division, New College,

Dickel, David N.
1991 Descriptive Analysis of the Skeletal Collection from the
prehistoric Manasota Key Cemetery, Sarasota County,
Florida (8So1292). Florida Archaeological Reports,
Number 22. Bureau of Archaeological Research, Division
of Historical Resources, Florida Department of State,

Fraser, Linda
1980 Faunal Analysis of the Venice Site. Florida Department of
State, Bureau of Historic Sites and Properties, Bulletin

Kozuch, Laura
1993 Sharks and Shark Products in Prehistoric South Florida.
Monograph 2, Institute of Archaeology and
Paleoenvironmental Studies. University of Florida,

Luer, George M.
1977 The Roberts Bay Site, Sarasota, Florida. The Florida
Anthropologist 30:121-133.

1999a AnIntroductionto the Archaeology ofthe LemonBay Area.
InMaritimeArchaeology ofLemon Bay, Florida, edited by
George M. Luer, pp. 1-23. FloridaAnthropological Society
Publication Number 14, Tampa.

1999b Cedar Point: A Late Archaic through Safety Harbor Occu-
pation on Lemon Bay, Charlotte County, Florida. In
Maritime Archaeology of Lemon Bay, Florida, edited by
George M. Luer, pp. 43-57. Florida Anthropological
Society Publication Number 14, Tampa.

Luer, George M., and Marion M. Almy
1982 A Definition of the Manasota Culture.
Anthropologist 35:34-58.

The Florida

Milanich, Jerald T.
1994 Archaeology ofPrecolumbian Florida. UniversityPress of
Florida, Gainesville.

Plowden, William
1953 University of Florida Site Survey Form for 8So 11. On file,
Florida Master Site File, Bureau of Archaeological Re-
search, Florida Division of Historical Resources, Tallahas-

Widmer, Randolph J.
1988 The Evolution ofthe Calusa: A Nonagricultural Chiefdom
on the Southwest Coast ofFlorida. University of Alabama
Press, Tuscaloosa.

Willey, Gordon R.
1949 Archeology of the Florida Gulf Coast. Smithsonian
Miscellaneous Collections, Volume 113, Washington D.C.




Patricia Crandon Randell

Pat Randell, a long-time resident ofPineland on Pine Island
in Lee County, southwestern Florida, died on October 19,
2002, at the age of 82.
As much as anyone else, Pat was responsible for preserving
a large portion of Pineland. It was her fervent wish that
Pineland be saved in as natural a state as possible. Her late
husband, Don Randell, shared her hope and took steps to carry
it out.
Growing up in Miami, Pat was heart-broken to see so many
ofDade County's rural, wild areas become urbanized. As a
young person, the natural beauty and wildlife of the
Everglades greatly impressed her. She loved birds. She often
recalled the abundance of wading birds in the "Glades," and
was saddenedby their disappearance. Later, the spread of land
development in southwestern Florida sickened her. She saw
the 1960s dredging and filling of Cape Coral, across the river
from Fort Myers, as a disaster for natural habitats and wildlife.
For the Randells, Pineland was a retreat to the tranquility
and beauty of natural Florida. They retired there in 1968,
moving from New Jersey, where Don had commuted to work
in Manhattan. Pat expressed her love for Pineland and Pine
Island in many ways. An example is a series of post cards, or
"Cracker Cards," that she created in the 1970s to benefit the
Pine Island Public Library (Figures 1 and 2).
From 1968 through the Spring of 1980, the Randells lived
in an old frame house at the southern end of Pineland's
waterfront. Enchanted by the ambiance of shell mounds, sky,
water, and nature, the Randells decided to do what they could
to preserve the beauty of the area. In 1970, they purchased
nearby Josslyn Island, with its shell mounds and beautiful
gumbo limbo trees. Soon after, they encouraged the listing of
Pineland in the National Register of Historic Places (Penton
1972) and began purchasing more of Pineland to save it from
development. They also contacted the State of Florida to
assess Josslyn Island, leading to avisitby archaeologist Carlos
Martinez (1976) and to the listing of the site in the National
Register of Historic Places (Scarry 1978).
In 1980, Pat and Don went two doors northward on Water-
front Drive, moving into a new house near the western
entrance of the Pine Island Canal (Figure 3). I stayed with
them that summer, and a number of times after, to study
Pineland, the Pine Island Canal, Josslyn Island, and other
sites. Pat showed me a small collection of sherds, shell

artifacts, and glass beads that she and Don had picked up
around Pineland, which I photographed and sketched. Pat
lamented that, as a young girl, she had not been taught
anything about Florida Indians. As an adult, however, she
enjoyed reading about other cultures, including books by
anthropologist Loren Eiseley.
Pat and Don became long-time members of the Florida
Anthropological Society (FAS). Theymadesurethatmembers
of FAS and the Southwest Florida Archaeological Society
(SWFAS) assisted with archaeological mapping of Josslyn
Island, which they funded in 1983 through the Florida State
Museum (Marquardt 1984). Pat's curiosity aboutPineland led
to archaeological field work in her yard in 1988, when she
helped sort excavated materials in a field lab set up under her
house by zooarchaeologist Karen Walker (Marquardt 1989:2,
Both Pat and Don were pleased with articles in The Florida
Anthropologist about Pineland and the Pine Island Canal
(Luer 1986, 1989, 1991). After first reading my study of the
canal, Pat wrote to me:

Surely it must have taken generations of Indians to accomplish
the canal system. (I keep thinking of the chain gangs with
picks and shovels working on roads when I was a child!) Did
the Indians manage with nothing but shell tools?! Never mind
answering that. When I next see you I'll probably have a list of
questions. First, however, I must re-read the article. I sort of
swallowed it whole the first time round. Donald echoes my
enthusiasm .... [Randell 1989a]

In a follow-up letter, Pat thoughtfully observed that clay
could have been an additional item of exchange that the
Indians transported via canals:

It surely is possible that some clay was transported? You have
written that sherds at Pineland could represent vessels and
perhaps some oftheirmakers, transported there via canals. Clay
can be transported. At least, some was once "transported" to me
in an old bag. And it certainly didn't come from an art store.
Since I was not at that time connected to the Calusa, I didn't
inquire about its source. [Randell 1989b]

In 1989, Pat and Don were honoredby SWFAS with the first
"Dr. Frank C. Craighead Award" at an elegant banquet in
Bonita Bay (Lee 1989a, 1989b). The award was especially


MARCH 2003

VOL. 56(1)



2003 VOL. 56(1)

Figure 1. Drawings by Pat Randell for a series of post cards to benefit the Pine Island Public Library.



Figure 2. More of Pat Randell's drawings for post cards for the Pine Island Public Library. On the "Holiday Greetings"
card, note the red mangrove "Christmas tree" with crowning starfish. It is festooned with a hummingbird, anhinga,
kingfisher, owl, woodpecker, cardinal, great white heron, snowy egret, and baby raccoon, plus a blue crab, horseshoe crab,
horse conch, and great blue heron around its base. Behind are a jumping mullet, soaring bald eagle, and flying brown


~~h~bi. -~~ t



Figure 3. The new house that the Randells had built on Waterfront Drive in Pineland, from their 1980 Christmas letter.

meaningful because the late Frank Craighead, a retired
entomologist, had done ecological work helping to understand
and preserve the Everglades. In 1990-1993, Pat and Don
generously made donations to support The FloridaAnthropol-
ogist (Anonymous 2002:Table 4; Lee 1991). After their first
donation, Pat modestly wrote:

We thank all of you on the Board of Directors for awarding us a
Family Life Membership in the Florida Anthropological Society.
It is a lovely and much appreciated surprise. We are delighted
if we have made some worthwhile contributions toward Florida
archaeology. We have certainly learned a great deal. (I, an
elderly Florida native, never heard of a Calusa Indian until we
moved here!) [Randell 1991]

In 1990-1992, Pat and Don encouraged broad public
education and further archaeological work at Pineland as part
of two "Year of the Indian" grant projects, which included
participationby SWFAS and Time Sifters Archaeology Society
members. In 1992, a book about archaeological research in
the Lee County area, in part funded by the Randells, was
dedicated to them (Marquardt 1992). In 1993, Pat and Don
were given a "Stewards of Heritage Preservation Award" by
the Florida Archaeological Council.
Pat remained insistent that Pineland be preserved, leading to
a gift agreement in 1994 deeding 53 acres of their Pineland
property to a private entity, the University of Florida Founda-
tion. It was to be sold to the State of Florida (underway in
2002-2003) to help fund an endowment for an archaeological
and interpretive center at Pineland, with the land "used solely
for charitable, conservation, educational, or scientific pur-

poses" (Randell et al. 1994).
Don passed away in July 1995, and Pat continued to live in
the house on Waterfront Drive. Her health declined, but she
stayed in Pineland. She enjoyed her upstairs screened porch,
where she could look out into gumbo limbo trees, often with
birds in their branches. Today, her spirit lives on in the sunny
trees and pasture of Pineland.

References Cited

2002 FAS: A Real Treasure ofFloridaArchaeology. The Florida
Anthropologist 55:38-46.

Lee, Arthur R.
1989a Randells Honored for Archaeological Contribution. The
Pine Island Eagle, April 19, page 20.

1989b Randells to Recieve Award: Honor to be Conferred at April
20 Dinner Meet. Newsletter, SouthwestFlorida Archaeo-
logical Society. April. VoL 4, no. 12, pp. 1-2.

1991 The Randells of Lee County's Pineland: Florida Archaeol-
ogy Owes Them Much. The FloridaAnthropologist44:76.

Luer, George M.
1986 Ceramic Faces and a Pipe Fragment from South Florida,
with notes on the Pineland Site, Lee County. TheFlorida
Anthropologist 39:281-286.

1989 Calusa Canals in Southern Florida: Routes of Tribute and
Exchange. The Florida Anthropologist 42:89-130.


2003 VOI 56(l)


1991 Historic Resources at the Pineland Site, Lee County,
Florida, The Florida Anthropologist 44:59-75.

Martinez, Carlos A.
1976 Trip Report Josslyn Island Field Inspection. Report
dated July. Pp. 11. Prepared for the Florida Division of
Archives, History, and Records Management. On file,
Florida Division of Historical Resources, Tallahassee.

Marquardt, William H.
1984 The Josslyn Island Mound and Its Role in the Investigation
of Southwest Florida's Past. Florida State Museum,
Miscellaneous Project Report Series Number 22,

1989 Return to Battey's Landing. Calusa News 3:1-3, 8-9.

1992 Culture and Environment in the Domain of the Calusa,
edited by William H. Marquardt. Monograph Number 1,
Institute of Archaeology and Paleoenvironmental Studies,
University of Florida, Gainesville.

Penton, Daniel T.
1972 "Pineland Site" nomination form for the National Register
of Historic Places. On file, Florida Division of Historic
Resources, Tallahassee.

Randell, Donald H., Patricia C. Randell, John V. Lombardi, T. Peter
Bennett, and Robert R. Lindgren
1994 Gift Agreement to Establish the Randell Archaeological
Research Center. Dated May and June. Copy on file,
Randell Research Center, Pineland, Florida.

Randell, Pat
1989a Letter to George Luer, dated March. On file with G. Luer.

1989b Letter to George Luer, dated April. On file with G. Luer.

1991 Letterto-Terry Simpson, FAS Membership Secretary, dated
April 5. Copy on file with G. Luer.

Scarry, John F.
1978 "JosslynIsland" (8LL32) nomination form for the National
Register of Historic Places. On file, Division of Historic
Resources, Tallahassee.


... yes, but what was the

The exact, full wording of that reference is as close
as your phone:

Back issues of The Flodida Ant74bpologst -- going
back close to a half century are available at the

Graves Museum of Archaeology
and Natural History

481 South Federal Highway
Dania, FL 33004

Phone (954) 925-7770


(954) 925-7064

Sole agents for back issues of The Florida Anthropologist


Thinking About Significance: Papers and Proceedings,
Florida Archaeological Council, Inc. Professional Develop-
ment Workshop, St. Augustine, Florida. Robert J. Austin,
Kathleen S. Hoffman, and George R. Ballo, eds. Special
Publication No. 1, Florida Archaeological Council, Inc., P.O.
Box 2818, Riverview, Florida 33568-2818. 2002. 242 pages,
$15.00 (paper).

Florida Bureau of Archaeological Research, Division of
HistoricalResources, 500S. Bronough Street, Tallahassee, FL
E-mail: ltesar@mail.

This important publication is the result of a May 11, 2001
professional development workshop hosted by the Florida
Archaeological Council, Inc., in St. Augustine, Florida. The
conference subject, the concept of "significance" as it pertains
to cultural resources, particularly archaeological ones, is
critical. The fate of cultural resources, most of which have
lasted for hundreds and thousands of years, often depends
upon whether or not they are determined to be significant.
Determinations of significance have generally been in terms
of whether or not the site or property is considered to be
eligible for listing in the National Register of Historic Places.
The quality and strength of such decisions often depends on
the experience and research (or personal) interests of the
persons) preparing the initial site assessment work. Native
Americans and others have introduced broader cultural
considerations into the decision making equation.
Workshop participants included representatives of federal
and state agencies, academia, private consulting firms and
Native American tribes. The subject of the workshop has been
of personal interest to me for more than 25 years. For personal
reasons I was unable to attend the workshop, although I
responded to subsequent information requests from Ken
Hardin and Bryan Yates with respect to the preparation of
their contributions to the publication under review.
The publication begins with "Thinking About Significance:
An Introduction" by Robert J. Austin and Kathleen S.
Hoffman. They describe the workshop format: "we asked
[each participant] the question, 'from your perspective, what
is significant and why?' Under this rather broad topic, we
listed a number of issues and asked each participant to
consider any or all of them when preparing their papers" for
presentation at the workshop and subsequently in this publica-
tion (Austin and Hoffman 2002:5).
The presentations and audience discussions consider three
perspectives: agency issues, Native American issues and
archaeological issues. The various papers in this important

volume address those issues from a historical perspective (how
the process has evolved in Florida over the past 35 years), a
statutory and regulatory perspective (how the process routinely
works) and a humanistic perspective (how the process is
perceived by, affects, and is affected by the cultural beliefs of
Native Americans and other living people. It provides an
insight into the past, the present and the future of the process
of determining the cultural significance of properties and the
decisions that affect the treatment of those properties.
As Kenneth W. Hardin (2002:31) concludes in his presenta-

It is important to remember that no one person has the answers
to the problems that confront our discipline. This was true in
the early years of CRM and remains so today. Answers need to
be developed through consensus, which calls for an honest and
open exchange of opinions. It also is incumbent upon all of the
participants in the process to seriously educate themselves
regarding the varied perspectives held by others. Without this
there can be no consensus and without consensus, we can have
no "product" that everyone can buy into. I believe that unless
we discuss our ideas together in a respectful manner, people will
increasingly rely on or retreat into policy and procedure instead
of working together to find creative solutions to prevent the loss
of our heritage.

While the individual presentations in this volume are
important, I find the transcript of discussion comments
between each session's participants and the audience provides
a broader perspective to the issues discussed.
When I began this review of this 242-page compendium,
nearly every presentation and the audience comments at the
end of each session included comments that made important
points that I felt merited citation. However, that resulted in an
earlier review version of many, many pages. This workshop
publication is too important to be highlighted by such brief
excerpts, although it will be widely cited for years to come. It
is well worth reading in its entirety. It offers a
historic/cultural preservationfiameworkby which the past can
serve the present, and by extension future generations.


VOL. 56(1)

MARCH 2003



About the Authors:

TraciArdren is Assistant Professor of Anthropology at the University of Miami. She is Co-Director of the Pakbeh Regional
Economy Project centered at the ancient Maya trading center of Chunchucmil, in the western Yucatan peninsula. She
believes the ancient cultures of the Gulf of Mexico should be studied from a regional perspective, and plans to begin
excavations in south Florida again soon.

Bill Burger has been an independent archaeological consultant since 1982, working chiefly in west central Florida. He has
acted as archaeologist for Sarasota County's Department of Historical Resources and as Project Archaeologist for the
Southwest Florida Water Management District's Emerson Point C.A.R.L. project in Manatee County. He is currently an
archaeological consultant for Scheda Ecological Associates.

George Luer is an archaeologist from Sarasota, Florida. He has studied shell tools, ceramics, shell middens and mounds,
and canoe canals.

Ken Sassaman received his Ph.D. in anthropology in 1991 from the University of Massachusetts, Amherst. His research
centers on the social and technological developments on prehistoric hunter-gatherers of the American Southeast. He is the
author of Early Pottery in the Southeast: Tradition and Innovation in Cooking Technology (1993), co-editor with M.S.
Nassaney ofNative American Interactions: MultiscalarAnalyses and Interpretations in the Eastern Woodlands (1995), co-
editor with David Anderson of The Paleoindian and Early Archaic Southeast (1996), and others. Current projects include
fieldschools at Hontoon Island and Blue Springs, a GIS project focusing on the St. Johns River basin, dating of soot found
on soapstone vessels, and research on the Stallings Culture of South Carolina and Georgia. He is assistant professor of
anthropology at the University of Florida.

Brian Keith Sullivan is a graduate of New College of Florida and has excavated in Kentucky, Florida, and at a number of
Maya sites in Belize and Yucatan. He currently resides in Dalonega, Georgia and is part of the Celtic folk-rock band,
Emerald Rose.

Louis D. Tesar presently serves as a Senior Archaeologist in the Bureau of Archaeological Research, Division of Historical
Resources. From July 1977-March 1991 he was involved in the Division's compliance review program and from March 1991
until January 2000 as the Supervisor of the Archaeological Research Section. Prior to that, with the completion of his
master's degree in anthropology in 1973 at Florida State University, he worked on a series of archaeological survey and
excavation projects, many of which were conducted to meet the requirements of Section 106 of the National Historic
Preservation Act. The issue of archaeological site significance and factors concerning determinations of significance have
been of interest to him throughout that work. He is a strong proponent of public involvement in the process. He served in
various capacities in the Florida Archaeological Council and is a past Editor of The Florida Anthropologist (1983-1992).

Nancy Marie White is a professor of anthropology at the University of South Florida in Tampa. For many years she has been
investigating the prehistory and early history ofthe Apalachicola River Valley region. Other research interests include gender
in archaeology, archaeological theory, and public archaeology/cultural resources management.


2003 Voi. 56(l)

Full Text