2000 Florida Anthropological Society Inc.
The Florida Anthropological Society Inc. holds
source text of the Florida Anthropologist
considered the copyright holder for the text
all rights to the
and shall be
and images of
The Florida Anthropological Society has made this publication
available to the University of Florida, for purposes of
digitization and Internet distribution.
The Florida Anthropological Society reserves all rights to this
publication. All uses, excluding those made under "fair use"
provisions of U.S. Code, Title 17, Section 107 are restricted.
Contact the Florida Anthropological Society for additional
information and permissions.
THE FLORIDA ANTHROPOLOGIST
Volume XXIII, No. 1
A New Concept for the Busycon Shell Receptacle
William J. Webster . . ... .. .. 1
Stratigraphic Tests at Stalling's Island, Georgia
Ripley P. Bullen and H. Bruce Greene . . 8
Salvage Archaeology at Fort Walton Beach, Florida
Yulee W. Lazarus . . . . 29
OFFICERS OF THE SOCIETY
President James W. Covington
Tampa University, Tampa, Fla. 33606
Ist Vice President Carl A. Benson
3400 East Grant Ave. Orlando, Fla. 32806
2nd Vice President William M. Goza
P. O. Box 246, Clearwater, Fla. 33515
Secretary-Treasurer Sarah B. Benson
3400 East Grant Ave. Orlando, Fla. 32806
Editor-Resident Agent Ripley P. Bullen
Florida State Museum, Gainesville, Fla.,
Three years: Thomas H. Gouchnour
Two years: Cliff E. Mattox
Cocoa Beach, Florida
One year: Don D. Laxson
At large, for one year:
Charles A. Hoffman, Jr.
Donald W. Sharon
Fort Walton Beach
A NEW CONCEPT FOR THE BUSYCON SHELL RECEPTACLE
William J. Webster
The many shell middens along Florida rivers and lakes have ever been
fascinating. Some of these shell heaps have an abundance of sherds while
others are completely lacking. This absence of sherds has been an enigma to
professional and amateur archaeologists for years. These questions may be
partially answered by the discovery of Busycon shells (Fig. 1) probably util-
ized as cooking vessels by the archaic shellfish food gatherers at Astor.
The site is located at Astor, a small community on State Highway 40,
at the St. Johns River, 30 miles west of Daytona Beach and 40 miles east of
Ocala. The surface terrain is composed of ceramic and pre-ceramic snail
shell middens. The original site probably occupied an area of some twenty-
five acres and maybe John M. Goggin's (1952: 40-43) Mt. Taylor type site.
The part investigated is located on the east bank of the St. Johns River, 300
yards south of S-40, and approximately 300 yards from the river's main
channel. There is a side channel between this site and the main channel but
it was dredged during extensive shell excavations during the past half century.
This area at present contains a marina and motel.
That portion of midden remaining unexcavated or undisturbed to great
depth is topped by a grove of fruit trees and may total two acres. It is also
the area farthest from the river. The most recent shell removals were made
adjacent to and south of this still existing midden, and all Busycon shell ves-
sels and tools referred to in this text were collected from this area. The
depth of midden material remaining is approximately 10 feet, rising slightly
away from the river to its apex, then declining to the original ground surface
at a distance of about 100 feet.
Quite clearly the occupants of this part of the site had an economy based
on shellfish as their primary food source. The freshwater snail Viviparus
georgianus wareanus Ktister is predominate throughout the midden. The fresh-
water snail Pomacea paludosa Say, and freshwater mussels Elliptio buckleyi
Lea and Elliptio strigosus Lea are noticeable as small patches or lenses in the
middle and upper portion of the midden. The freshwater fossil snail Gonio-
basis floridense, Reeve and freshwater snail Helisoma scalare Jay are pre-
sent in minute numbers (Pilsbry 1934: 29-66, with updating of nomenclature).
There is a sparse scattering of animal remains throughout the midden
indicating they supplemented their shellfish diet. Animals represented were
deer, alligator, turtle, raccoon, garfish, drumfish, and unidentifiable birds.
BUSYCON SHELL RECEPTABLE
Some human skeletal material was noted but without any visible signs of a
burial. Evidently the deceased were interred in the midden.
MANUFACTURE AND UTILIZATION
It appears that the aborigines at this site produced their shell cooking
vessels chiefly from the larger whelk, Busycon contrarium Conrad (Lightning
Whelk). Their shell working tools, on the other hand, were produced from
the smaller and thicker whelks, Busycon carica Gmelin (Knobbed Whelk) and
Busycon perversumLinne,formerly Busycon carica eliceans Montfort, (Kie-
ner' s Whelk). All are marine shells and indigenous to Florida (Abbott 1968).
They were obtainable by many routes: Three likely routes and approximate
distances are 120 miles north on the St. Johns River to its junction with the
Atlantic Ocean; 90 miles south on the St. Johns River to a point near Cocoa
where the St. Johns River parallels the Indian River at a distance of less than
5 miles; and the direct overland route of some 30 miles which terminates at
the Atlantic Ocean near Daytona Beach.
To manufacture these Busycon vessels, the columellas were removed,
probably by the use of a hammer and chisel as there are no signs of cutting
or sawing. The rims are slightly smoothed to produce a dipper-like recept-
acle (Figs. 1-2), not unlike those found in middens and burial mounds of more
recent cultures (Sears 1959: 28).
All specimens from this site described below are technically surface
finds as no archaeological tests have been made. However, they did not come
from the recent surface of the ground but were exposed by commercial shell
These Busycon vessels appear to be burnt mostly on their bases and
slightly on their sides (Figs. 1-2). The burnt area is dark blue-grey to black
and all vessels are broken on their base at the probable point of greatest heat
damage. Nine complete and broken specimen of these Busycon vessels were
collected. Eight are Busycon contrarium Conrad, and their approximate
widths are 15.5, 15.25, 14, 13.3, 12. 75, 11.5, 11.4, and 8.7 cm. The ninth
vessel was manufactured from the Busycon carica Gmelin and measures 11. 2
cm. The smallest B. contrarium vessel is the only one without discoloration
or any indication of having been fired, although it is broken at its base. The
rim is ground and polished extremely well and it apparently was utilized as a
drinking receptacle or dipper.
Nineteen specimens of shell tools were collected. Of these, five are
composed of whole shells, Busycon perversum Linne, and represent one plane,
two adzes and two picks with holes, possibly for hafting. Eight were scrapers
or gouges made from parts of the Busycon' s largest whorls, and six chisels
or drifts made from the columellas of Busycons. The miscellaneous tool as-
semblage comprises five Newnan Points, two Florida archaic stemmed points
(Bullen 1968), three unclassified stone points or knives, two small knives or
scrapers, eight polished bone points, and one stone drill,
Miscellaneous marine shells collected were three Crassostrea virginica
Gmelin (Common Eastern Oyster), one Mercenaria campechiensis Gmelin
(Southern Quahog), one Dinocardium robustum Solander (Giant Atlantic Cockle),
one Noetia ponderosa Say (Ponderous Ark), one Anadara notabilis Roding
(Eared Ark) and one broken piece of Charonia variegata Lamarck (Trumpet
Triton) (Abbott 1968).
SUMMARY AND CONCLUSIONS
As there was no discoloration or indication of fire damage present on
any of the shell tools that were extracted from the same shell midden as the
Busycon vessels, it has been discounted that the coloring on the base of the
vessels was formed from percolating surface water containing organic mater-
ial. Also it is on the outside, not the inside, of the specimens (Figs. 1-2).
These Busycon vessels were examined by Ripley P. Bullen, Florida
State Museum, and he acknowledged that they appeared burnt and possibly had
been utilized as cooking vessels.
Specimens of Busycon vessels and tools obtained by archaeological ex-
cavations were examined at the Florida State Museum in Gainesville, and the
following data was obtained. In agreement with the Astor site, none of the
shell working tools (ceramic or pre-ceramic) examined have discoloring or
any appearance of fire damage. They also appear to be manufactured from
the thicker and smaller Busycons as listed from the Astor site.
A total of eighteen Busycon contrarium vessels were examined. Eight
are from pre-ceramic or possibly early fiber-tempered horizons. All ex-
hibit fire damage (Fig. 2) and appear to have been utilized as cooking vessels
except the smallest. The lip of this small vessel is ground and polished and
the knobs on the exterior whorl have been removed by grinding. The remain-
ing ten Busycon vessels are from ceramic middens and burial mounds, and
none exhibit fire damage or discoloration. The following is a list of these
sites, and the specimens in the Museum collection that were examined by Mr.
Bullen and myself.
Archaic middens: Bluffton, 1 very large, 1 large, and 1 small Busycon
vessel (Cat. No. 95097), large ones show fire damage,small one does not;
Tick Island, 2 large, 2 medium sized Busycon vessels (96814/5/9, all with
fire damage; Mosquito Grove, 1 large (102532) with fire damage. Post- Ar-
BUSYCON SHELL RECEPTABLE
chaic middens and burial mounds: Castle Windy midden, 1 large, (95229),
no fire damage; Long Pond burial mound, 1 large (102422), no fire damage;
Pillsbury burial mound, 2 medium sized, (98124/7), no fire damage; Poca-
son Hammock, 1 large (102421) no fire damage; Crystal River, 1 medium
sized (29606), no fire damage; Palmer burial mound, 1 large (95614), no fire
damage. The correlation between pre-ceramic Archaic middens and fire-
damaged large Busycon vessels and between non-fire-damaged vessels and
later ceramic complexes is very high. The only exception seems to be three
vessels, 1 large and 2 small (102640), from Bluffton in the Simpson Collection.
It is stated that these were found inside each other. Apparently they were
never used and hence not subjected to fire damage. Of the great many shell
tools from Archaic middens in the Florida State Museum' s research collec-
tions, none exhibit burning damage.
To support the use of these Busycon shells as cooking vessels, a Busy-
con contrarium was obtained live at Cedar Key. The animal was removed and
the columella was chiseled away to produce a vessel similar to the ones found.
Water was boiled in this vessel over an open fire until approximately five gal-
lons was evaporated. There was no noticeable damage except discoloration.
Its properties as a cooking container are not unlike a modern type of ceramic
cooking vessel. This shell' s width is 10 cm. and its capacity is one mea-
sured cup. The only observable disadvantage of the Busycon vessel as com-
pared to a pottery vessel is its limited size. They do have a natural built-in
handle the shell's siphonal canal.
While inferring that the Astor site is a pre-ceramic midden, it is poss-
ible these Busycon vessels were coeval with early pottery containers in other
localities. Pending a radiocarbon test, this midden cannot safely be dated
although the presence of Newnan points suggests a date of about 5,400 years
Before Present (Bullen 1968: 30). I do not imply that all pre-ceramic mid-
dens contain these Busycon cooking vessels, but it is probable many did.
Grateful thanks are extended to Mrs. Ann Moulton of Astor, who so
generously allowed surface collecting during removal of shell from the midden;
to Mr. Ripley P. Bullen of the Florida State Museum for his assistance in
classifying the specimens from the site and in identifying the midden as pre-
ceramic; and to my personal friend and conchologist Mrs. Miriam K. Hicks
of Jacksonville for her help in identifying the marine and freshwater mollusks.
Dr. Fred G. Thompson, Florida State Museum, kindly supplied the latest no-
menclature for some of the freshwater mollusks.
Fig. 1. Busycon cooking vessel from Astor.
Fig. I. Busycon cooking vessel from Astor.
9 'F '
Fig. 2. Busycon
~P tB ,
Abbott, R. Tucker
1968 A- Guide to Field Identification SEASHELLS of North America.
Golden Press. New York.
Bullen, Ripley P.
1966 Burtine Island, Citrus County, Florida. Contributions of the
Florida State Museum, Social Sciences, No. 14. Gainesville.
1968 A Guide to the Identification of Florida Projectile Points.
Florida State Museum. Gainesville.
Clench, William J., and Ruth D. Turner
1956 Freshwater Mollusks of Alabama, Georgia and Florida from the
Escambia to the Suwannee River. Bulletin of the Florida State
Museum, Biological Sciences, Vol. 1, No. 3.
Goggin, John M.
1952 Space and Time Perspective in northern St. Johns Archeology.
Yale University Publications in Anthropology, No. 47. New
Pilsbry, Henry A.
1934 Review of the Planorbidae of Florida, With Notes on Other
Members of the Family. Proceedings of The Academy of Natural
Sciences of Philadelphia, Vol 86, pp. 29-66.
Sears, William H.
1959 Two Weeden Island Burial Mounds, Florida.
Contributions of the Florida State Museum, Social
Sciences, No. 5. Gainesville.
STRATIGRAPHIC TESTS AT STALLING'S ISLAND, GEORGIA
Ripley P. Bullen and H. Bruce Greene
In the spring and summer of 1961, Greene conducted archaeological
excavations at the well known Indian mound or shell midden located on
Stalling's Island in the Savannah River between Georgia and South Carolina
(Fig. 1, upper). The island is situated near the middle of the river, be-
tween an electric power dam and the Augusta City locks, shortly below the
mouth of Big Stevens River. Downstream from Stalling's Island are sev-
eral smaller islands, indicating the shallowness of the river at this point.
The location should have been an excellent one for the collection of fish
and freshwater shellfish, the remains of which form major components of
In 1929 the midden measured 512 by 300 feet (Claflin 1930: 1). Now,
due to floods and other natural agencies as well as works of man, its size
has decreased to about 480 by 285 feet. The site is formed of approximate-
ly 22 feet of natural clay left by the cutting of the Savannah River over the
centuries plus a superior layer of black loam, shell, and cultural refuse,
which forms the midden proper. Depth of cultural deposit varies from 54
to 76 inches due to depressions in the clay base of the midden. On the mid-
den's crest are three high tension towers (Fig. 1, lower, T-1/2/3) which
carry electric power lines across the river. Deciduous trees grow on both
sides of the midden and low bushes and briars grow across the midden.
The Stalling's Island midden has been investigated by Jones (1873) in
1870, Claflin (1931: 3-4) at various times between 1908 and 1925, the Cosgroves
(Claflin 1931: 4-6) in 1929, and-Fairbanks (1942) in 1940. Approximate loca-
tions of their excavations ( following Claflin 1931: Pl. 2) are shown in the low-
er part of Figure 1. Claflin's (1931) report of the Cosgroves' extensive ex-
cavations remains the classic as well as the first major report on a fiber-tem-
pered site in southeastern United States.
The area under the central high tension tower (Fig. 1, lower, G) was
selected for work for two reasons. First, it was the only centrally located
relatively undisturbed area; and second, the tower legs were ideal for bench
marks since field equipment was limited to tape, plumb bob, compass, string,
and line level.
BULLEN AND GREENE
Fig. 1. Sketch maps of Stalling's Island and of the surrounding region.
J, Jones' 1870 tests; C 1-4, Claflin's 1908-1925 tests; Tr 1-4, dug by the
Cosgroves' in 1929; F-F, areas of Fairbanks' 1940 investigations; G, area
of Greene's 1961 tests; T 1-3, corner supports of high tension towers.
In March 1961, two exploratory trenchs (Eig. 2, Tests 1-2) were dug
to see if further work would be productive. After several days, a pattern
began to form--particularly in Test 2--with plain fiber-tempered sherds
concentrating at the base of the ceramic zone, decorated pottery above, and
preceramic levels below.
In June and early July of 1961, six stratigraphic pits were excavated.
Their locations, size, and depths are indicated in Figure 2. A northwest-
southeast profile is presented as Figure 3. As indicated there, the midden
is -comprised of alternating layers of dirt, dirt and shell, shell, and burnt shell.
Shells were predominantly those of freshwater clams and no salt water varie-
ties were noted. Various animal and fish bones were present, both in the ce-
ramic and in the preceramic deposits. The corner of Claflin's C-1 Test was
clearly evident in the west corner of Pit 6, Otherwise evidence of disturbance
was limited to the highest foot of the midden.
As shown on the profile (Fig. 3) and indicated in Figure 2, pottery was
limited to upper zones which varied in thickness from 28 to 42 inches in total.
Predominantly, pottery was fiber-tempered--with or without substantial a-
mounts of additional sand temper--but a few shell-tempered and some purely
sand-tempered sherds were also found. These, particularly the shell-tempered
ones, concentrated at shallow depths (Table 3) and included decorated sherds
belonging to the Irene and Savannah ceramic complexes.
After excavation, sherds from Pits 3 and 5 were sent to the Florida State
Museum, Gainesville, Florida, where they form part of the Museum's Research
Collections and, later, were analyzed by Bullen. Sherds from the other four
pits remained in Georgia where they were studied by Greene. Results of these
analyses are presented in Tables2 and 3 and will be discussed later. Projectile
points were surprisingly abundant in the stratigraphic tests and are illustrated
by arbitrary 6-inch levels in Figure 4 where points from all six pits have been
combined. These points, also, will be discussed later (Table 1).
One interesting discovery was the finding of charred corn in Pit 6 at a
depth of 19 inches or well into the fiber-tempered ceramic zone. The cobs
lay in a small pocket or pit which, although covered by a shell cap, was pro-
bably the bottom of an intrusive storage, cooking, or smudge pit dug down from
higher Irene or Savannah levels. This corn has been identified by Richard A.
Yarnell of Emory University, Atlanta, Georgia, as Eastern complex or North-
ern Flint corn. Some of the corn cobs have been radiocarbon dated (Sample M-
1280) at 200 + 100 years B. P. dr around A. D. 1750.
Charcoal and charred plant remains were also found in three other places.
One from Pit 2 at a depth of 60 inches was about 18 inches below the lowest pot-
tery in that pit (Fig. 3). It consisted predominantly of pine but included red
BULLEN AND GREENE
P-4 P-2 Claflin's
70 54 5
70 Test 2
A Y II
Cosgrove Cosgrove Cosgrove Tr-2, Crogrove
grid line grid line a grid line main excavation
A -Tower legs 0 5 10 15 20 Feet
I I ii I
Fig. 2. Excavation plan locating Greene's exploratory Tests and stratigraph-
ically excavated Pits. Numbers in Tests and Pits indicate depths of lowest
pottery and of top of underlying sterile clay in inches.
oak and pecan hickery as well as nut shells (identified by R. A. Yarnell). The
date (Sample M-1277) is 4450' 150 B.P. or around 2500 B.C. The second,
consisting of charcoal and nut shells, was found in Pit 5. This sample came
from a small fire pit, 21 inches in diameter, which extended downward from
the base of the cultural deposits at 70 inches into the sterile clay a distance
of 12 inches. The date (Sample M-1279) of 4700 150 years B. P. or around
2750 B. C. is slightly earlier than that given above in agreement with their
12 STALLING'S ISLAND
Pit 3' Pit 2 Pit I Test 2
3- "-- S : S 3 2
S. S S -- S S S S
-3- .... ... .. s .....S .. .. 3..s ..... ..... .-.*.'-'
a..: -- .- b
45 S -S
3 Clay Cloy
6- s Clay
7- Cloy .)- 6 ft. -
Cross hatching-dirt s shell b burnt shell ..-lowest pottery
Fig. 3. Profile X Y from Figure 2.
relative proviences. These dates are both from the basal zone and sug-
gest habitation at Stalling's Island started around 2500-2750 B.C. dur-
ing a late phase of the preceramic Archaic period.
A fourth radiocarbon dated sample came from Pit 4 at a depth of
30 inches. This charcoal, while predominantly pine, included sycamore
as well as red oak fragments. The date (Sample M-1278) is 3730 150
B.P. or about 1780 B.C. As this sample came from the base of the
pottery producing zone in this pit (Fig. 2) it should date an early phase
of fiber-tempered ceramics, perhaps the plain period noted during exca-
vation. That such a period should be present at Stalling's Island is indi-
cated by the increase in the percentage of plain to total fiber-temered
sherds in nearby Pit 5 where they increase to 54 per cent in the fourth
and 71 per cent in the fifth or last pottery level (Table 2).
In Claflin's (1931: 40-42) report, he cites the locations of eight o-
ther sites of the Stalling's culture. Greene visited these locations in the
spring of 1961 and found that only two were left. There was the possibil-
ity of a third, but it could not be located according to Claflin's descrip-
As Claflin stated, Sites 1 and 2 were destroyed, Site 3 was still
there but Site 4 was destroyed in 1961 in the process of widening a road.
Greene was unable to locate Site 5, and believes it was destroyed in the
flood of 1938. Site 6 was still present although it was dwindling away
BULLPEN AND GREENE
fast. Site 7 had been destroyed the winter of 1960 by bulldozers in mak-
ing a logging trail, and Site 8 had been destroyed by a farmer over a per-
iod of yearsin cultivating. The actual location of the last site is .5 miles
down stream from where Claflin located it unless he is referring to a site
destroyed when the water works were built in the mid-twenties and which he
might have previously visited.
Perhaps the most important result of the 1961 work at Stalling's Is-
land is data regarding changes in projectile points with time. Not only
are there typological and size changes, as indicated in Figure 4, but there
was also a high correlation between typology and material (Table 1). The
following section covers pointed stone tools--projectile points or knives--
irrespective of their functional usage. Drills and a few obviously asymme-
tric forms will be covered under "other stone tools."
We have defined four descriptive types of points for the 1961 pits at
Stalling's Island. Type 1, similar to Claflin's (1931: 34) "square stem," is
illustrated in the lowest line on Figure 4 (dd-ii). These are fairly good-
sized, broad-bladed points with a stem which has parallel or approximately
parallel sides. The base of the stem is usually straight but may be concave
(Fig. 4, ff, ii). These points are usually made of porphyritic rhyolite and
in the stratigraphic pits concentrated in preceramic deposits (Table 1).
Type 2 averages a little shorter than Type 1 but differs chiefly in that
the stem is expanding and the blade sometimes barbed (Fig. 4, cc, jj) so
that the point might be called corner-notched. Type 2 points are fairly rare
at Stalling's Islhnd, usually made of quartz, rarely of rhyolite or chert, and,
like Type 1 points, concentrate in preceramic deposits. However, they were
found at shallower depths on the average.
Type 3 points are shorter and relatively narrower than Type 1 points.
They are also thicker and without the parallel faces frequently found on Type
1 points. Their stems are relatively longer, have contracting sides, and us-
ually rounded bases. Blade shoulders are rounded or pointed and but little wi-
der than the tops of their stems (Fig. 4, p-r, v-w). These points are usual-
ly made of quartz and resemble oblong diamonds with sharp corners. When
made of rhyolite or chert, the base of the stem tends to be concave. Type 3
points concentrated in the fiber-tempered pottery zones (Table 1). In
this zone were also some small editions of Type 1 and Type 2 points.
The "extinction of Type 3 points as the digging penetrated the base of fi-
ber-tempered ceramic deposits was quite noticeable during excavation.
Type 4 points are hard to differentiate. They are made of quart-
zite and chert, and concentrate in the highest 18 inches of the cultural
debris. They have small straight-sided and straight-based stems (Fig.
4,. j-o). Blade and stem corners may be sharp or rounded. There are
two varients. In the first, always made of chert, the stem shape is
similar to those of Type 1 but the blade is shorter and its sides converge
more rapidly. In the second, always made of quartzite, the base is
rounded or converging but the sides of the stem are more parallel than
for Type 3.
PROJECTILE POINTS BY DEPTHS AND BY MATERIAL, PITS 1-6
Typology Depths in the ground, inches Totals
0-18 19-36 37-54 55-72
Type 4 29 11 40
Type 3 10 36 10 56
Types 1-2 8 10 39 24 81
Totals 47 57 49 24 177
Typology Chert Quartzite Quartz Rhyolite Totals
Type 4 31 8 1 40
Type 3 1 48 7 56
Types 1-2 1 80 81
Totals 32 8 49 88 177
Available photographs of these points have been arranged in Figure
4 by depths in the ground. Additional data will be found above in Table 1
where points have been tabulated by depths and by materials. The 177
BULLPEN AND GREENE
points listed in this table include only fairly complete specimens and iden-
tifiable basal portions. Tips and broken mid-sections have not been includ-
ed. The only small triangular point found (Fig. 4, a) has been reproduced
as "q" in Figure 5 to show chipping details. Also illustrated in Figure 5
is a large quartz knife (Fig. 5, r) similar to, and found at the same depth,
18 inches, as "h" in Figure 4. One has a contracting and the other a
square stem. Apparently, these older Archaic forms persisted as large haft-
ed knives into the later, post-fiber-tempered, ceramic periods.
Other Stone Tools
Other than projectile points, the most common chipped stone tools
found at Stalling's Island were drills. We have illustrated unfinished as
well as complete and broken examples (Fig. 5, f-l). Typically they have
expanded, sometimes, cruciform, hafts. Most commonly hafts resemble
those of Type 1 points. Such drills in our pits occurred between depths of
12 and 24 inches or in the fiber-tempered ceramic zone but undoubtedly they
are holdovers from the previous preceramic times. One (Fig. 5, f) has
been enlarged in a lower illustration (Fig. 5, s) to show chipping scars.
The next more frequent stone tools are markedly asymmetric hafted
knives (Fig. 5, t-u). These were found in fair quantities in the lower or
fiber-tempered part of the ceramic deposits. A few exhausted cores (Fig.
5, b), many worked fragments or broken tools (Fig. 5, d), uniface scra-
pers or knives (Fig. 5, c, p), and worn out specimens (Fig. 5, n) were
found in all levels. One broken point had been resharpened to form a
hafted knife with a sharply angled edge (Fig. 5, a). Three specimens with
sharp points suggest engraving tools (Fig. 5, e, m, o) but may be small
hafted knive s.
As listed in Table 3, two broken steatite weight fragments were found
in Pit 3 (Fig. 6, a-b). These have parallel sides and are not fragments of
steatite vessels. Except that they are broken across the drilled hole, they
resemble those illustrated by Claflin (1931: P1. 52) exactly. Their rela-
tively shallow prevenience should be noted ( Table 3).
Ceramic description and discussion will be divided into two sections.
The first covers the sherds from Pits 1, 2, 4, and 6 which were kept in
Georgia and analyzed by Greene. The second covers sherds from Pits 3
and 5 which were deposited. at the Florida State Museum where they were
analyzed by Billen. Examination of the profiles (Fig. 3) suggests the pos-
sibility that pottery from Test-2n-might be earlier, on the average, than
that from Pit 3. Similar differences between the ceramic chronologies of
Pits 3 and 5 (Table 3) may be suggestive of the differences in the thickness
of their deposits. Although the two groupings and analyses do not give very
significantly different pictures, it seems destreable for the above reasons
not to combine the data.
Greene, from his examination of the pottery from Pits 1, 2, 4, and
6, notes that two major types were found at Stalling's Island, fiber-tem-
pered sherds and those of the later Savannah and Irene wares. The fiber-
tempered pottery, which contains a variable amount of sand, could be bro-
ken down into three decorative varieties: plain, punctated, and stab-and-
drag. The plain ware, at least in the early period, was made by modeling
rather than coiling. This opinion is based on the large number of sherds
which lacked parallel-edged breaks in comparison with the larger percentage
with parallel-edged breaks found in the later horizons, when pottery was made
In general, the fiber-tempered ware at Stalling's Island had an average
thickness of .375 inches, a rim diameter of 16 inches, and a vessel depth of
7 inches. The maximum indicated diameter was 25 inches and the shallowest
depth 1.25 inches. Vessel sides sloped slightly outward from the bottom,
which was flat. Vessels with punctated decoration tended to have a slight in-
ward lip extension. Color ranged from black to buff with darker colors pre-
The early pottery development at Stalling's Island is not readily apparent.
However, with care it can be traced. After the initial plain period, simple
punctating was introduced and vessels boldly marked with half moons, circles,
and slight curves. Circles were probably made by a hollow reed and the other
marks by bone tools. Both random and straight line patterns were found but
punctations were not placed extremely close to each other. A few sherds-with
slash-like incising were also found but not enough to justify a separate cate-
gory. In the third stage, linear punctation or the stab-and-drag method was
used and individual punctations are very close to each other. Decoration was
applied by pushing the tool in at an angle, as before, except that it was "drag-
ged" back and then pushed in again almost directly behind the previous mark.
This formed the trough or depressive which is very typical of Stalling's Island
decoration. There is a tendency to use a smaller tool for more intricate de-
signs. Sometimes two or more tools have been used to vary the design on the
same vessel. Sometimes blank spaces were left, during the application of de-
coration, to form panels as well as geometric patterns.
BULLEN AND GREENE
Savannah pottery is predominantly grit-tempered and decorated with
complicated stamping. It ranges in color from dark brown to black. Irene
pottery, on the other hand, is primarily shell-tempered, sometimes with
sand admixture and, rarely, with sand but no shell tempering. It tends to
be gray in color. When decoration is present it is done by incision. Irene
sherds, especially when they contain sand as temper, are by far the hard-
est found at Stalling's Island. The average thickness of both Savannah and
Irene ceramics is .25 inches. No vessel size was calculated but it was noted
that the vessels seemed larger and deeper than the earlier ones. Both Sa-
vannah and Irene vessels have sloping sides converging to a rounded base
with, in some cases, flaring rims.
VERTICAL DISTRIBUTION OF POTTERY, PITS 1, 2, 4, 6, BY LEVELS
Typology Levels 1 2 3 4 5 6 7 Totals
Irene, shell, some sand temper 31 23 15 5 74
Savannah, grit temper 33 12 18 4 67
Stalling's stab-and-drag 50 72 71 71 33 4 301
Stalling's punctated 18 26 38 97 38 18 10 245
Stalling's plain 17 45 60 87 65 42 26 342
Totals 149 178 202 264 136 64 36 1029
Per cent, Stalling's plain of 20.0 25.5 47.8 72. 2
total Stalling's ware 31.4 34.0 65.7 38.5
Results of Greene's analysis are presented by 6-inch levels in Table 2.
The dramatic increase in the percentage of Stalling's plain pottery with depth
is to be noted. The priority of simple punctation over linear stab-and-drag
is also evident. A comparison of Table 1 and 2 shows a correlation between
Type 4 points and Irene-Savannah ceramics, Type 3 points with fiber-temper-
ed pottery, and Types 1-2 points with preceramic levels.
Model analyses by Bullen of the ceramic materials from Pits 3 and 5 are
presented in Table 3. These analyses are probably unnecessarily detailed but
have been done in that manner in an attempt to present as closely as possible
a complete description, to indicate the variation involved, and to find modes
which changed with time and, hence, might be used to divide the long Stalling's
period into subperiods.
Fiber-tempered sherds from Pits 3 and 5 have a contorted paste but are
well-fired and rather hard, 2.5-3 on Mohr's scale. This hardness is probably
due to the inclusion in the paste of appreciable amounts of sand. Practically
all include bits of micaceous material or tiny quartz grains which produce a
surface "sparkle. While occasional bits of grit, small quartz pebbles, and,
extremely rarely, bits of gold are found, none of these seem to have been
added intentionally as temper. Fiber-tempering molds are usually abundant
but there is much variation and sometimes it takes careful examination before
definite evidence of fiber-tempering is found. In these cases sand is extremely
abundant. The paste, in general, consists of an admixture of clay with fi-
brous materials and sand, with the amounts of sand and fiber inversely pro-
portional to each other. As shown for both pits in Table 3 under "Stalling's
Plain, sand is more abundant at higher than in lower levels.
Color is usually dark tan, grading into reddish and blackish brown. No
evidence of coiling was found. Walls range from 5 to 11 mm. in thickness but
the great majority are between 8 and 10 mm. Rim sherds are incurving, some
very slightly, others more so, and a few pronouncedly. Lips are us-
ually flat-rounded and inward sloping, sometimes simple rounded (Figs. 6,
e-h; 7, a-d). Occasionally, rim edges are thinned (pinched) and pulled up-
wards with a vertical lip (Figs. 7, j; 8, t) or rolled inward and thickened (Fig.
8, u). In one case the top of the rim had been widened to form a "T" with an
inslanting flat lip. This unillustrated sherd has a little fine incising on its
side. Another has a scratched "L-shaped" rim or lip.
A few vessels had inturned casuela-like upper sides (Fig. 6, i). Other-
wise, sherds suggest fair-sized, round- or oval-mouthed, vessels with flat
bottoms. The only measurable bases were slightly oval with diameters of 3
and 3.5 inches (7.5 and 9.0 cm.). Parts of 6 flat bottoms were found in Pit 3
and 5 in Pit 5. Three from the 20-to 27-inch depths of Pit 3 bore punctations.
One sherd came from a vessel which was also punctated on the side. In that
instance, punctations on the bottom were more squashed than were those on
the side (Fig. 8, m). One base in the fourth level of Pit 5 had a scratched bot-
tom. No rounded bases were found.
Sherds from Pits 3 and 5 are listed in Table 3 and typical examples illus-
trated in Figures 6-8. The complexity of variation in a narrow decorative
scheme makes division into ceramic subtypes difficult. The majority may be
called linear punctated with closely spaced indentations pressed deeply into
the surface to produce a groove and subsumed under Stalling's Punctate (Grif-
fin and Sears 1950: 6-1). This stab-and-drag decoration is well illustrated in
Figures 7, a-j; 8, a-f). Occasionally, there is more emphasis on the "drag"
than on the "jab" as shown in Figure 8, i. Individual punctations may be semi-
lunar, round, pointed, or bifid. One common tool is diamond-shaped, perhaps
from wear, like a screwdriver with the corners ground off. While straight
BULLEN AND GREENE
parallel rows (Fig. 7, f-j) are common, complicated designs as shown in the
upper part of Figure 8 are not uncommon. Bifid (Fig. 6, q-s) and different
sized tools on the same vessel (Fig. 8, f-g) produce variation. Paneling is
also a common decorative feature (Figs7 6, r-s; 7, g). Often, but less fre-
quently, punctations are not close together (Figs, 7, m; 8, o, s). In only
one case (Fig. 7, c) did a punctated lip edge occur.
Others methods of pottery decoration used at Stalling's Island on fiber-
tempered vessels include grooving, scratching, and incision plus, in one in-
stance, rocker stamping. Except for the last, these decorative styles occur
in combination with linear punctattion as summarized in Table 3.
What is meant by grooving has been referred to by others as "simple
stamping. However, it differs so much from what is usually meant by the
concept "simple stamping" that it seems best to use a different term. Groov-
ing seems to be of two types. First, there are long, shallow, and rather
broad parallel lines impressed on the sides, flattened lips, and, also, on the
inner edge of lips (Fig. 6, f-g; Claflin 1931: Pl. 19. i). This usually re-
quires three separate operations. Such imprints could easily have been made
with the sides of styluses. The same treatment, impressions on the flatish
top and, also, on the outer and inner edges of lips occurs on linear punctated
vessels (Fig. 7, b, f). On flat or flat-rounded lips (but not on the edges)
these indentation occur rather commonly (Fig. 7, e, g-l, n). They also oc-
cur, but perhaps less frequently, on incurving, pinched, or inturned lips.
This attention to the lips is a feature of Stalling's Island ceramics and seems
to foretell similar treatment found later on Deptford and Swift Creek vessels.
The other type of grooving is similar but composed of rather narrow
lines which are close together and relatively deeper (Fig. 7, a, d). This
type of grooving sometimes is applied to the outer surfaces of vessels before
the application of drag-and-jab decoration (Fig. 6, d). Imprints of the wider
tool also occurs in two instances before punctation (Figs. 7, i; 8, i, u; Claf-
lin 1931: Pl. 14, a). The last may have been applied after punctation.
A third form of surface treatment may be called scratching. This oc-
curs less commonly and may be part of the finishing process. It is promi-
nant on one inturned rim (Fig. 6, i), on our unillustrated "L"-lip vessel, and
on some sherds illustrated by Claflin (1931: Pl. 14, d-f).
Incision, while present, is usually poorly done, especially when it oc-
curs alone (Fig. 6, j-m). When applied in combination with linear punctation
to form a complicated design, incising may be very fine and well done (Fig. 6,
n; Claflin 1931: Pl. 13, e) but this is rare. One sherd with a sharply incised
side has a grooved flat lip (Fig. 6, 1). On this lip, narrow grooving appears
VERTICAL DISTRIBUTION OF POTTERY, STALLINGS ISLAND
Pit 3 Pit 5
Typology 3- 9- 16- 20- 23- 30- 37- To- 0- 6- 12- 18- 24- To-
9 16 20 23 30 37 42 talks 6 12 18 24 30 talks
3 7 3
1 4 1
4 3 9
Steatite weight fragments
Sand plus fiber
Fiber plus some sand
Decorated on lips only
Fine inc., nicked lip edges
Scratched, wide "L'Llip
Scratched, pinched lip
Grooved on sides and lips
Grooved on sides only
Grooved over punctations
Grooved over incision
Scratched, flat lip and side
Scratched on sides only
Scratched over punctations
Unique linear punctated
Irregular, crossed lines
Ditto plus wavy lines
Punct. above, inc. below
Inc. lip, cross-hatched side
Punctated, rim herdsf
Linear with plain lip
Linear with grooved lip
Linear with scratched lip
Linear with nicked lip
Ovate punct., grooved lip
Ditto, parallel slant, lines
Ditto plus grooved lip
Ditto plus incised lip
Punctated body sherdsg
Ditto with bifed tool
Ditto, hollow reed
Ditto, complex designs
1 1 1
3 2 8
2 1 4
12 29 19 76 54
17 236 2 4 16
4 2 4 3 13
Totals 45 62 67 119 72 49 27 441 19 27 54 86 17 203
BULLEN AND GREENE
aIncludes 3 Irene Incised, 2 Irene Complicated Stamped, 3 Savannah Fine Cord
Marked, and 3 Savannah Check Stamped for Pit 3; and 2 Irene Complicated and
1 Savannah Check Stamped for Pit 5--all in first three levels.
bAlso contains some sand. cHas flat, widened, inward slanting, undecorated
lip. dHas flat, widened, "T"-shaped lip and some sand temper. eContains
some sand temper. Usually fiber-tempered, occasionally contains sand.
gWhile sand is occasionally observable, great majority exhibit only fiber tem-
pering grooves and holes.
superimposed on wide grooving.
A few unique fiber-tempered sherds need to be-mentioned-- One, defi-
nitely fiber-tempered but containing a lot of very fine sand, bears parts of
two rows of rocker stampings (Fig. 6, o). Another (Fig. 6, c) exhibits curv-
ing,, perhaps spiral, indentations reminiscent of the Florida type "Tick Is-
land Incised." Close examination, however, discloses it is not incised but
that the curved lines are formed by closely spaced, elongated, linear puncta-
tions. This sherds contains very little sand and is light in weight.
No attempt has been made in this section to describe the Savannah and
Irene ceramics found in higher levels of Pits 3 and 5. The interested reader
is referred to footnote "a" for Table 3, at the head of this page.
This paper has been written to record the results of Greene's excava-
tions at Stalling's Island and to present a detailed description of the fiber-
tempered pottery found there. Sears and Griffin's 1950 (6-1) type descrip-
tion, while excellent for gross comparisons, does not give enough detail to
permit comparisons of a type variety nature. One of the details noted in this
report is the great attention given to vessel rims and lips which are impress-
ed on their tops and edges with simple indentations. This trait is rot men-
tioned by either Claflin (1931: 13-17) nor Fairbanks (1942: 24-25) in their o-
therwise excellent descriptions of Stalling's Island pottery. It is clearly shown
in at least one of Claflin's (1931: Pl. 17, i) illustrations.
Miller (1949: 41-42) for the Lake Spring site, also in Georgia, mentions
decoration on flattened lips which "usually appears as straight-line incising,
made with a sharp-pointed instrument placed at right angles to the body of the
vessel. The striations are spaced at irregular intervals, ranging from 2 mm.
to 5 mm. The strokes appear to have been applied from the outside in, .as they
are deeper on the outside. Most of the lines become indistinct towards the in-
ner edge of the lip. Our shallow indentations vary considerably but may be
generalized as usually 5 mm. wide with 5 mm. to 1 cm. spacing. On one
sherd with incurving rim (Fig. 7, i) indentations approach 1 cm. in width
with 1.5 cm. spacing. Our wide shallow grooving does not conform to Miller's
description but our second type (Fig. 7, a) might meet his description adequate-
ly. There is a considerable similarity in Stalling's rim treatment and that des-
cribed by Phelps (1969: 25) for the Thom's Creek phase of eastern Georgia, par-
ticularly his types 2 and 6. Concentration of this decoration in higher Stalling's
period deposits support his and others contention that Thom's Creek ceramics
developed out of those of the Stalling's Island phase.
Examination of Table 5 shows very few typological changes with depth
which might be used for chronological implications. The ratio of plain to
decorated ware is much higher at the base of the ceramic deposits than high-
er up. Also there seems to be a concentration of decoration where puncta-
tions do not touch each other in lower levels. These two trends support
Greene's observations made during excavation and in his analysis of sherds
from Pits 1, 2, 4, and 6 (Table 3). There is also an increase in the use of
sand as tempering material, and probably in the presence of scratching, as
depths become less great. Of course, Savannah and Irene potteries are
limited to shallow depths.
Changes in projectile point typology are interesting and indicate a shift
during the fiber-tempered ceramic period from points with fairly large straight-
or parallel-sided stems to smaller points with contracting stems. There also
seems to be a surprising emphasis on the manufacture of hafted drills or per-
forators. These changes must reflect environmental or cultural changes coin-
cidental with the development of clay vessels.
A comparison of the pottery from Stalling's Island shows very little simi-
larity with that from the Orange period of Florida except for the fact of fiber-
tempering. Bullen would suggest from his examination that the same fibrous
materials were used but, of course, there is a great deal more sand in the
paste of Stalling's Island pottery. As is well known, the major difference is
the use in Georgia of closely spaced, deeply impressed, linear punctations
and in Florida of incising. In both cases, within the decorative method chosen,
there is considerable design variation. While punctated fiber-tempered pot-
tery is known for Florida, only three sherds of Stallings Punctate--trom
three widly separated places--have been found in Florida (Goggin 1952:
75). Incision on the other hand is found at Stalling's Island. In our sam-
ple it falls into two groups. The firdt (Fig. 6, j-m) where it is used as the on-
ly decoration. In such cases it is poorly done resembling carelessly decorated
Ornage Incised. In the other case, incision has been used in connection with
linear punctation (Fig. 6, n) where it is very neatly and expertly applied.
The lack of similaritiy between decorative modes in Georgia and Florida
do not indicate frequent periods of communication between the two areas during
fiber-tempered times. The Wheeler series, to iudge from available illustra-
BULLEN AND GREENE
tions (Sears and Griffin 1950: 101 to 41.), exhibit several of the traits found
at Stalling's Island and are closer in many respects to Georgia than to Florida
fiber-tempered ceramics as has been noted by other writers on various occa-
Steatite vessels are found in Florida during the later part of the Orange
period. The lack of great similarity in decoration or of trade vessels between
eastern Georgia and eastern Florida seems to confirm the theory that steatite
was brought into Florida via the Chattahoochee -Apalachicola River valley and
not via eastern Georgia. The lack of steatite fishing weights in Florida would
argue similarly. Both Georgia and Florida looked westward during the fiber-
tempered ceramic period and seem to have respected each other's territori-
Claflin, William H., Jr.
1931 The Stalling's Island Mound, Columbia County, Georgia. Papers
of the Peabody Museum of American Archaeology and Ethnology,
Harvard University, Vol. 14, No. 1. Cambridge.
Fairbanks, Charles H.
1942 The Taxonomic Position of Stalling's Island, Georgia. American
Antiquity, Vol. 7, No. 3, pp. 223-31. Menasha.
Goggin, John M.
1952 Time and Space Perspective in Northern St. Johns Archeology.
Yale University Publications in Anthropology, No. 40. New Haven.
Jones, Charles C., Jr.
1873 Antiquities of the Southern Indians, Particularly of the Georgian
Tribes. Appleton. New York.
Sears, William H., and James E Griffin
1950 Prehistoric Pottery Eastern United States. Ceramic Repository
for Eastern United States. Ann Arbor. Mimeographed.
Miller, Carl F.
1949 The Lake Spring Site, Columbia County, Georgia. American
Antiquity, Vol. 15, No. 1, pp. 38-51. Menasha.
Phelps, David S.
1968 Thom's Creek Ceramics in the Central Savannah River Locality.
Florida Anthropologist, Vol. 21, No. 1, pp. 17-30.
February 20, 1970.
0 12 in.
12 18 in.
0 I 2 3 4 5 in.
a i I I
Fig. 4. Projectile points from all six pits by depths in the midden.
BULLEN AND GREENE
2 4 i
- 24 in.
n 0 I 2 3 4 5
I, I I s i
On \ tempered
Fig. 5. Miscellaneous stone tools from all six pits by midden depths.
a, resharpened knife; b, small core; c-d, worked fragments; e, small point;
f-1, drills; m, small point; n-p, wor-n knives; q, enlargement of Fig. 4, a;
r, hafted knife, 18 in. depth; s, enlargement of f; t-u, asymmetric hafted knives.
- kl '-
Fig. 6. Steatite fragments and decorated sherds from Stallings Island.
a-b, drilled steatite weights; c, unique linear punctated;d, grooved over
linear punctations; e, wide, expanded, inward sloping lip; f-g, wide grooves
on lips and sides; h, faintly grooved lip, micaceous paste; i, scratched,
bent inward rim; j, incised, inward curving rim; k, incised; 1, scratched
lip, incised sides; m, incised with straight and wavy lines; n, curved linear
punctations above incised cross hatching; o, rocker stamped; p, grooved
lip, oval punctations, q-s, decorated with bifed tools, paneled designs.
STALLING'S ISLAND 27
IN. i I 3
-y/ (y~ EC l C C UCC* aal
~~ ~d m
Fig. 7. Stallings Punctated sherds with decorated lips.
a, d, narrow, relatively deep, grooves; b, e-l, n, shallow grooves;
c, diagonal grooves at front edge; i, n, have markedly inturned rims;
j, has pinched lip, m, scratched; o, flat, undecorated lip.
BULLEN AND GREENE
Fig. 8. Variations in Stallings Punctated decoration.
a-e, adjacent areas with different row directions; f-h, different tools used
on same sherd; i, punctations in panels; j-m, curved rows; n, fine punctations,
curved lines; o, shallow, ovate punctations; p, bifed tool; q-r, single row de-
signs; s, split reed imprints; t, deep grooves made with closely spaced marks
of curved edge tool; u, shallowly grooved surface, done after punctating.
a,e, inturned rims; a, p, r, flat lips; f, rounded lip; g, inward rolled rim,
flattened on inside with shallow.grooves; j, m, from flat bottoms; p, shallow
grooved lip; t, pinched rim; u, double row of shallow grooves, oneon top,
other on outer edge of lip.
SALVAGE ARCHAEOLOGY AT FORT WALTON BEACH, FLORIDA
Yulee W. Lazarus
The fact that Fort Walton Beach began to grow substantially only in the
past 25 years (population of 96 in the 1940 census) has afforded an opportunity
for archaeological study not possible in many older towns. Several different
types of archaeological activities have been conducted in the past ten years in
the downtown area. Information from these projects has been collected and
presented here to supplement that previously known. Indications are that the
earliest villagers, the Deptford people, were followed chronologically by sub-
sequent cultures in the same general area, but with differing characteristics
in settlement pattern and in the specific selection of living areas.
The present town is situated in the Florida panhandle on the west side of
Choctawhatchee Bay and on the north shore of Santa Rosa Sound where the sound
enters the bay. The sound is formed by the presence of Santa Rosa Island,
stretching for 40 miles east to west from Choctawhatchee Bay to Pensacola
Bay. In general, the area is rather heavily wooded, with fresh water streams
emptying into the bay and sound. Aboriginal materials show there was ex-
tensive occupation for the past several thousand years.
Although early settlers in Fort Walton Beach farmed the area, the plow
zone did not entirely penetrate nor completely disrupt Indian village debris.
Recent developments are presented here in chronological order and by archaeo-
logical project names. These projects are discussed as the Brooks Street Pro-
file 1957, the ADSO area 1960, the Ramp Dig 1963, the Brooks Bridge Survey
1964, and the Highway 98 Salvage 1966 (Fig. 1). These are sites within a square
half mile area downtown where in Deptford times people lived just behind the
dune ridge of the shoreline. Later people seem to have lived somewhat closer
to the shore and the last villagers erected a temple mound over the earliest
middens. The ramp of the temple mound faces south and the area between the
mound and the shore was evidently their plaza area though their village areas
appear to have been remote from their ceremonial center.
For background information, there are several early reports in varying
detail on the Indian sites in present downtown Fort Walton Beach. The first
good account is found in Walker' s 1883 report on mounds and shell heaps on the
west coast of Florida. He reports two Indian towns on Choctawhatchee Bay,
one on the south side of the bay and a larger town where the sound enters the
bay, the subject area of this paper. Although he admits to dense undergrowth
in the vicinity of the Fort Walton temple mound as a deterrent to careful survey,
he makes definite report of the numerous shell piles along the sound and the
FORT WALTON BEACH
shell midden and pottery exposed by farming. Moore (1902: 454) adds further
detail, besides his digging in the Temple Mound itself, of stamped ware and
borrow areas for building the mound. The most careful report, Archaeology
of the Florida Gulf Coast (Willey 1949: 84) records Deptford, Santa Rosa -
Swift Creek, and Weeden Island material as present in the areas around the
Temple Mound, which, however, was built by the people of the Fort Walton
culture. Fairbanks' (1965: 253) report on the mound points to a predomi-
nance of Deptford and, to a lesser extent, Santa Rosa Swift Creek material
in mound fill. These reports are supported by the following projects.
SANTA ROSA SOUO/ND
Fig. 1. Sketch map of downtown Fort Walton Beach locating various projects.
a, Brooks Street profile; b, ADSO area; c, ramp dig; d, Brooks Bridge sur-
vey; e, Highway 98 salvage; f, Willey and Woodbury's 1940 tests.
The Brooks Street Profile 1957
In 1957 the gas company dug an east-west four-foot deep trench for 1100
feet along the north side of Brooks Street, parallel to, and south of, Highway
98 between the sound and the mound (Fig. 1, a). Profile drawings (Fig. 2)
reveal the following points of interest. Four areas had residual shell from
aboriginal shell heaps varying from 25 to 85 feet across. Black midden hu-
mus was found almost the entire length (70 per cent) with evidence of this ma-
terial having been removed in the recent past from the remaining distance.
The four foot depth penetrated the occupation level for all but 200 feet and ex-
posed either the yellow sub-soil, common to this area (for 495 feet), and/or a
white dune sand, also common to this area (for 1000 feet) and usually under-
lying the yellow soil in sites near the Gulf shore. No artifacts were salvaged
from this work.
7YELLO WA//TE SAMIP
HUMUS ///qGRAY DiRfT \\\SHFLL t ++CL4AY / NtM
Fig. 2. Brooks Street profile.
One section for 200 feet was black midden soil throughout with no under-
lying deposit reached. The measurable midden depts ranged from 6 inches to
48 inches. One section had a 4 to 12 inch underlying layer of clay some 6 to 12
inches below the upper surface. Admittedly, this may remain from an early
settler' s clay road. On the other hand, it is directly in front of the Temple
Mound and could have been a plaza floor packed down by the Temple Mound
builders. This possibility is supported by the fact that the same type of clay
layer has also been observed in three other places in the vicinity. Only further
testing can clarify this point.
The ADSO Area 1960
In 1960 along this same street one parcel of land was sold to be developed
for a bowling alley (Fig. 1, b). This is the area just east of the old Indianola
Inn on the shore of Santa Rosa Sound and it contained the shell heap tested by
Willey and Woodbury in 1940 (Willey 1949: 74-75). The then present building
was removed and the bulldozer moved in to level the land. The site of the OK-6
excavations of 1940 was completely destroyed (Fig. 1, f). All seven of Willey
and Woodbury' s pits were in the bull-dozed area. By following the bulldozer
and grader it was possible to pick up a large collection of sherds. Interest-
ingly enough, some of the sherds thrown out by the blade fitted together and a
plain Weeden Island bowl was reassembled.
When the grading reached the yellow sterile sand there remained two
circular shell areas identified as pits which the Indians had probably dug for
refuse (Fig. 1, b). The bulldozer operator agreed to stop in this area to per-
mit systematic excavation. There was no humus layer within the pits, almost
all solid shell with some bone, charcoal and sherds. The east pit was the
larger and was dug to a five-foot depth. The west pit, four feet in diameter,
was dug only to a depth of three and a half feet and the bottom of this may not
have been cleaned out to its final depth. Ninety per cent of the total sherds
(1799) were of Weeden Island period types. Ninety per cent of the other ma-
terial reveal a diet of oyster, clam, scallop, fish, turtle, mammal, deer and
bird. Both pits were 300 feet from the present shore of the sound and along
the dune ridge.
32 FORT WALTON BEACH
The Ramp Dig 1963
During the summer of 1963 further excavations into the Temple Mound
were conducted as a follow-up to the Fairbanks dig of 1960. His excavations
(Fairbanks 1965) had firmly established the character and origin of the mound
as having been built by people of the Fort Walton culture using midden material
of earlier periods as mound fill.
a a z m U
Fig. 3. Trench in ramp, Fort Walton temple m
S* Two-thirds of the way
Sup the ramp, just short of
where the ramp joined the
summit of the mound, a
trench was begun (Fig. 1,
c). It cut from the west side
side into the incline from
ground level. By the end
of the season this cut had
penetrated half way through
the width of the ramp and
S. down to a depth of 9.2 feet
measured from the high-
est wall. It reached below
original ground level by 2
S feet (Fig. 3). In the dig
S various levels of mound
building were recognized
in the first 6.5 feet and a
S pure Deptford period mid-
S den was reached at the o-
riginal ground level. At
Level 11 in Pit 5, when
it became apparent that all
the material being proces-
sed through the screens
a was pure Deptford, a halt
was called to revamp the
details used in excavat-
ing. At this point all
material was processed
W through window screen of
14 wires to the inch with
S a water wash to clear the
artifacts for collection.
This project was set up
found. to be 7.5 by 5 feet but by
the time a 3-by 3-foot area, .7 feet in depth, had been processed through the
fine screen, the excavators were completely saturated with materials. They
elected to square up the work and to consider the faunal remains as one quan-
titative sample'for analysis. After that, they returned to processing through
a 1/4 by 1/4 inch screen. The same people, who collected the quantitative
sample, dug the rest of this level. Based on their knowledge of what had gone
into the quantitative sample, they picked from the screen any bones which they
suspected might-be different from those in that sample. These became a quali-
This Deptford layer was a nearly foot thick black midden capped with
shells. This cap was the reason for the excellent preservation of food bones.
Although the bones were largely fish bones there was also fish cartilage,
numerous fish scales, and bones of mammals and birds. All of the bones
of this 3 by 3 foot section were separated for analysis which was done by
Dr. Elizabeth S. Wing, Florida State Museum. Twenty species of fish and
six other vertebrates were represented. Of note is the fact that the most
popular fish of these early people were jack, catfish, and shad. All were
shallow water fish. Besides fish and shellfish, their diet includes box
turtle, terrapin, gopher tortoise, rabbit, white-tailed deer, and bird (Wing
Shells were collected in the same manner. These were analyzed by
Brandy Siebenaler, manager of the local Gulfarium, who reported there
were seven different types of shell, all but two coming from brackish water.
From the brackish waters there were the Bulla occidentalis Adams, Strombus
pugilis Linne, Oliva sayana Ravene', oyster, and barnacles on oyster shells.
The Cardium sp. shells came from salty water and the Southern Quahog (Venus
mercenariumLinne) from Gulf waters. It was felt that Bay Scallops should
have been present but there were none. Variations in the salinity of the Bay
water from one period of time to another may be the reason.
In the concentration of midden from which the faunal samples were taken,
there were 109 sherds. Classification of these is included in Table 1. Com-
ments on these sherds as quoted from the field notes are: "Only one of the 16
sherds classified Santa Rosa Stamped is classic. The other 15, which appear
to come from one vessel, are on a Deptford-like paste and the rocker stamp-
ing is bold and deep with no dentate characteristics. It is suspected it is a
type with which we are unfamiliar possibly in the Tchefuncte series. The
Basin Bayou sherds are two very small ones but have the characteristic round
bottom broad line of this type. The St. Andrews Complicated Stamped (Early)
sherd is not a good specimen and is as close as it can be identified. It is a
complicated stamp geometric in design. The sherd is small. The Swift
Creek Complicated Stamped sherds are small, also. The identification is
tenuous. The Santa Rosa Punctated are on a fine Weeden Island-like paste and
FORT WALTON BEACH
SHERDS FROM PIT 5, RAMP DIG 1963
associated with faunal sample
St. Andrews Com. St., Early Var. 1 St. Marks Plain, tetrapod 1
Swift Creek Com. St., Early Var. 2 Deptford Check Stamped 12
Basin Bayou Incised 2 Deptford Bold Check Stamped 31
Santa Rosa Punctated 5 Deptford Linear Check Std. 20
Santa Rosa Stamped 16 Plain, Deptford paste 13
Polished plain 6 Total 109
from general diggings
Lake Jackson Plain 7 Alligator Bayou Stamped 4
Pensacola Incised 4 Mossy Oak Simple Stamped 11
Pensacola Plain 2 Smithsonia Zone Stamped 4
Fort Walton Incised 1 St. Andrews Com. Stamped 4
Moundville Engraved 1 Crooked River Com. Stamped 3
miscellaneous brushed 6 unidentified stamped 82
Franklin Plain 21
Weeden Island Incised 3
Weeden Island Punctated 2 Deptford Linear Check Stamped 188
Weeden Island Plain 2 Deptford Bold Check Stamped 202
Carrabelle Punctated 6 Deptford Simple Stamped 129
Carrabelle Incised 4 Deptford "punctated" 7
Mound Field Net Marked 1 Dunlap Fabric Impressed 12
Keith Incised 1 Bayou La Batre Plain 49
Ruskin Dentate 1 Alexander Plain 5
Wakulla Check Stamped 1 St. Marks Plain, tetrapod 2
West Florida Cord Marked 4 micaceous plain 15
Swift Creek Com. Stp., Late Var. 8 clay-tempered plain 1
Mossy Creek Stamped 5 sand-tempered plain 349
plain red 3 Smooth Plain 122
Swift Creek Com. Stp., Ear. Var. 145 Fiber-tempered plain 2
West Florida Cord Marked, E.V. 13
Santa Rosa Stamped 46 Total 1495
Basin Bayou Incised 17
well polished. Mica is in the paste. The punctates are not round bottomed as
in normal Santa Rosa Punctated and may have been done with a shell tool. The
fine thin polished plain has considerable mica in it. Cylindrical vessel shape
is indicated. They may be part of the same Santa Rosa Pupctated vessel. In
other words, all fine polished ware could be from one vessel. All sherds are
small." Other significant artifacts from Level 11 are listed in the field notes
as Bayou La Batre, simple stamping, base section of a quartzite point
(.beveled) a fragment of sheet copper with two edges folded, and two small
fragments of mica plus some Florida hickory nuts (pig nuts). Using Sears
(1963: 21) ceramic seriation at the Tucker Site, most of the pottery fits be-
tween Tucker I and II or the middle of the Deptford Period.
Sherds from the rest of this mound ramp test, i. e. from above the
midden, are listed in the lower part of Table 1. While 72. 9 per cent of the
sherds are from the Deptford period, more recent pottery is, of course, also
present. Three projectile points (Fig. 5, a-c) were Jackson or Jackson-
like (Bullen 1968: 20) and hence tend to confirm the Deptford period dating.
The extent of the Deptford midden is apparently large and extends under the
mound itself. Further analysis must await further excavation. Comments on
this work are not complete without noting the complete absence of fish
hooks. The abundance of very small fish bones leads one to assume that fish
traps, nets, and weirs were used by these early people and boats were, per-
haps, not essential. Their diet was preserved by the ground acid neutralizing
effect of shells.
The Brooks Bridge Survey 1964
For over 1000 feet a storm drainage system was installed in conjunction
with building a new Brooks Bridge connecting the mainland to Santa Rosa Is-
land on the east side of town and about a half mile east of the mound (Fig. 1, d).
Survey of the cuts made for this and for new foundations for the bridge revealed
the natural ground sloped upward to a bank about 200 feet back from the sound
shore. The cut penetrated the yellow subsoil overlying white beach sand and
profiles along the walls were made where possible at selected areas (Fig. 4) .
s_ s $ SH hs ,
"1 ~ r Wrr A -; 7
Fig. 4. Brooks Bridge area profile.
FORT WALTON BEACH
Only 15 sherds were recovered and only 5 of these were in situ finds.
The sample from disturbed area were 6 plain (one with fiber temper) 2 shell-
tempered Pensacola series (1 incised) and 2 Weeden Island (1 incised)
sherds. In situ sherds were all in the midden layer and were 1 Lake Jackson
Plain, 1 micaceous plain, and 3 Pensacola Plain sherds. Fort Walton culture
ceramics predominates in the black top layer and was more plentiful, in ratio,
than along the sound to the west. Elevation at the starting point was 7 feet
above sea level. After deep cuts were made and foundations built, the dirt
was graded in to the present elevation of 5.36 feet.
Highway 98 Salvage 1966
Upon completion of the bridge, construction crews then moved into the
job of four-laning the highway through town. Removal of old paving and side-
walks revealed the residual shell deposits previously reported by Walker
(Fig. 1, e). Cooperation of the highway construction company and its crews
afforded the opportunity to recover materials. Despite the haste of such an
operation, some information-was gained. Field work of this type is hardly
complete but data from future developments may be combined with this report
to fill out the picture. Test pits were staked off and screened by levels in
order to obtain the most information possible under the circumstances.
Some 2200 sherds were recovered, predominantly Weeden Island
but with Santa Rosa-Swift Creek and Deptford materials nearly equal
in quantity (Table 2). Though some parts of the shell heap areas appeared
to be in original condition, upper areas were very disturbed and some wall
profiles showed that refuse pits and grading had mixed aboriginal and his-
toric materials to a considerable degree. Besides fragments of glass, metal,
and crockery, and an old fence post, which might be expected, the historic
debris was highlighted by a 1936 penny (Denver mint), two bullets (projec-
tiles, old fashioned 45 caliber), a token for the Indianapolis Railway (!) and a
button from someone's blue jeans. A respectable number of diagnostic sherds
and stone tools were recovered to add to the limited collection of the early
cultures in this area (Table 2 Column 2).
Ceramic fragments recovered in Highway 98 salvage appear to be large
in number but it must be admitted many of the sherds are very tiny. Most of
the pottery was sand or sand and grit temper, much of it with the character-
istic sheen of mica in the sand temper. Tetrapods were both large and small,
pointed and rounded. A number of sherds were rather hard to classify. Ten
sherds scattered among five tests seem to be from the same vessel. Sand-
tempered paste is hard, almost black, smoothed on the inside, and with clear
paddle impressions. The dominant paddled decoration seems to be linear
check stamping but impressed or ovenstamped in several directions and once
with a doughnut shape object (Fig. 6, a-e). Also illustrated are a Deptford
Fig. 5. Miscellaneous specimens from Fort Walton Beach salvage work.
a-c, Jackson points from Ramp Dig; d, perforated shark's tooth; e, Pi-
nellas point; f, base of well-chipped point of non-Florida stone; g, Westo
point; h, Archaic stemmed point, broken base; i-j, pottery scrapers of
fossil bone and of shell; k, unique complicated stamped sherd. All High-
way 98 Salvage except a-c. k is 1.625 inches wide,, others by scale.
FORT WALTON BEACH
'*** **y 'M
0 I 2 3
I inche I
Fig. 6. Sherds from Highway 98 Salvage.
a-d, Deptford Bold Check Stamped; e, Deptford Linear Check Stamped;
f, Deptford Simple Stamped; g, West Florida Cord Marked; h, Santa
Rosa Stamped; i, unidentified cord-marked.
SHERDS FROM HIGHWAY 98 SALVAGE 1966
Pensacola Incised 3 St. Andrews Com. Std. 11
Pensacola Plain 3 Crooked River Com. Std. 4
Fort Walton Incised 1 unidentified stamped 16
miscellaneous brushed 1 Franklin Plain 21
Weeden Island Punctated 7 Deptford Linear Check Std. 29
Weeden Island Plain 90 Deptford Bold Check Stamped 136
Carrabelle Punctated 7 Deptford Simple Stamped 20
Wakulla Check Stamped 16 St. Marks Plain, tetrapod 12
West Florida Cord Marked 10
Swift Creek Com. Std., Late Var. 100 residual plain 1497
plain red 124
Swift Creek Com.Std., Ear. Var. 106
Santa Rosa Stamped 15
West Florida Cd. Mkd., Ear. V. 17
Basin Bayou Incised 9
Alligator Bayou Stamped 13
Simple Stamped, West Florida Cord Marked, Santa Rosa Stamped and an ex-
ample of relatively late cordmarking (Fig. 6, f-i). One sherd is a unique
complicated stamp but used in roulette fashion (Fig. 5, k). Half a cup was
recovered, two and a half inches in diameter and one inch deep. It is sand.,
tempered with a black core, red slip, rough textured inside and outside sur-
faces, and a rounded lip.
The variety of shell types is greater than in the ramp materials but,
significantly, also represent brackish waters in general. Bones were of
turtle, deer, and fish. There was an assemblage of tools and tool-making
materials. These included the following: 1 shell and 3 stone tools; 13 stones
of limonite, pumice, or sandstone used for sharpening; 5 pieces of ochre and
one of hematite for paint; 1 hammerstone, 2 scrapers; and 1 drilled shark
tooth (Fig. 5, d). The shell tool and a stone [fossil bone] tool are considered
to be pottery smoothing tools as both are identical in configuration, improb-
able to have been shaped by nature, and fit the hand conveniently for scraping
or smoothing the outside of a ceramic bowl, and have scraping facets (Fig. 5,
i-j). Quartz chips appear in relative large quantity (28) with 11 more of crys-
tal quartz. A total of 8 whole or broken points were represented 2 of chert,
5 of quartzite, and one of flint (Fig. 5, e-h). In general,- their time span fits
the apparent ceramic time span. One (Fig. 5, f) is the base of a well-chipped
knife of non-Florida streaked gray and brown-gray flinty rock.
FORT WALTON BEACH
Stratigraphy followed a pattern of gray surface layer for the top 12 to 20
inches underlain by black humus. In some places the shell deposit was in the
midden dirt, in others on top of it. For a distance of at least 85 feet a layer
of clay 8 to 12 inches deep was visible under the gray layer. Here again we
may have part of the plaza floor mentioned earlier. Profiles from some 462
feet of trench and test walls can be combined with notes from general obser-
vation of the 2500 feet as the dozers and graders dug a deep trench for new
storm sewers and prepared for new road surfacing. An overall picture of
this area is reasonably possible to assume. Much of the total trench length
produced no evidence of aboriginal materials but several shell heaps were
exposed, black humus of living areas were adjacent, another area of packed
clay strata is similar to the one mentioned earlier was found.
A word should be inserted here to explain that the shell heap where the
Indianola Inn once stood has been referred to as (probably) Walker' s largest
shell heap (Willey 1949: 75). Excavation of this area was begun in 1966 but has
not been completed. When it has been, another report will be pertinent. Any
report on it at this time would be preliminary to a final one. It is carried in
the records as 80K6sm (shell mound) and will eventually be treated as a sepa-
rate entity. Also, shell middens are exposed to a considerable extent along
the dune ridge throughout town but only surface collections have been made.
In general, the Weeden Island period is strongly represented.
A comprehensive look at the various reports produces these points to be
considered. Walker reports shell deposits of 40 to 60 feet diameter between
the mound and the sound in 1883. Moore refers to various small mounds of
midden refuse and also to cultivated fields in the same area in 1901. He notes
an abundance of check and complicated stamped sherds in mound fill and in the
surrounding cultivated areas. Willey' s shell heap near the water' s edge is
composed of a significant amount of Deptford period pottery though it did con-
tain more Santa Rosa-Swift Creek and, higher, Weeden Island material. Fair-
banks observes the Temple Mound fill came from beyond the northeast corner
of the mound, directly north of the mound and northwest of the mound. This
area is more remote from the shore. He (Fairbanks 1965: 253) further states
"There is a strong suggestion that the Deptford midden extended in these di-
rections. Santa Rosa-Swift Creek is not well represented in mound fill,
Weeden Island less so. The ramp dig did produce a bona fide Deptford de-
posit on original ground level. The possibility of a Deptford burial mound
under the center of the Temple Mound is not remote. It would be quite feasible
for a group such as the last, those of the Fort Walton culture, to select an
existing small mound to use as base for building their temple mound remote
from their own villages and to use existing nearby village debris of an earlier
The compatibility of all these data, major and minor, blends well into
a summary of occupation in the area as suggested in Figure 7. The proximity
of remains of the Santa Rosa-Swift Creek culture to the Deptford, contiguous
or adjacent, in most of the lower levels supports the conclusion that Santa
Rosa-Swift Creek people moved in and lived among the Deptford people and
eventually imposed their material culture, at least ceramically. Milk quartz
from Mobile and mica from central Alabama are evidence of short distance
trade while the copper fragment must have come from a more distant location.
The Swift Creek paddle must have arrived with early Georgians, yet some of
the designs among the complicated stamp sherds of Highway 98 are not fami-
liar among designs east of the Appalachian chain (Broyles, Bettye J., per-
sonal communication). The Deptford and Santa Rosa-Swift Creek peoples
seem to have lived a little more inland while the later Weeden Island people
lived closer to the shore. This is the reverse of what would appear to be
/// SHELL AREAS Lo0 Fr
Fig. 7. Sketch map of downtown Fort Walton Beach showing reconstructed
FORT WALTON BEACH
logical unless the later people, developing agricultural practises, cultivated
the areas more inland from their homesites (Fig. 7).
Bullen, Ripley P.
1968 A Guide to the Identification of Florida Projectile Points.
Florida State Museum. Gainesville.
Fairbanks, Charles H.
1965 Excavations at the Fort Walton Temple Mound, 1960. Florida
Anthropologist, Vol. 10, No. 4, pp. 239-44.
Moore, Clarence B.
1902 Certain Aboriginal Remains of the Northwest Florida Coast, Part II.
Journal of the Academy of Natural Sciences of Philadelphia,
Vol. 12, pp. 127-358. Philadelphia.
Sears, William H.
1963 The Tucker Site on Alligator Harbor, Franklin County, Florida.
Contributions of the Florida State Museum, Social Sciences,
No. 9. Gainesville.
1885 Mounds and Shell Heaps on the West Coast of Florida.
Smithsonian Institution, Annual Report for 1883, pp. 845-68.
Willey, Gordon R.
1949 Archeology of the Florida Gulf Coast. Smithsonian Miscellaneous
Collections, Vol. 113. Washington.
Wing, Elizabeth S.
1967 Animal Remains from a Midden at Fort Walton Beach.
Quarterly Journal of the Florida Academy of Sciences, 30: 1,
Fort Walton Beach
THE FLORIDA ANTHROPOLOGIST is a quarterly publication of the
Florida Anthropological Society, Inc. Subscription is by membership in
the Society for individuals or institutions. Annual dues are $4.00; stu-
dent membership $2.00.
Copies of Volumes I through XIV may be purchased from Johnson
Reprint Corporation,111 Fifth Avenue, New York, N. Y. 10003 at $8.00
per volume or $2.00 for single and $4.00 for double numbers if available.
Copies of Volumes XV through XXIII may be purchased from the Treasurer
of the Society at $4.00 per volume or $1.00 for single and $2.00 for double
numbers. The Society has no resale discount for book dealers.
INFORMATION FOR AUTHORS
THE FLORIDA ANTHROPOLOGIST publishes original papers in all
subfields of anthropology with an emphasis on archaeology. Contributions
from allied disciplines are acceptable when concerned with anthropological
problems. The journal's geographical scope, in order of importance, in.-
cludes Florida, Southeastern United States, Caribbean, and related regions.
Manuscripts should be double spaced and typed on one side only of an
8 1/2 by 11 inch sheet leaving at least a 1 inch margin on all sides. Contri-
butors are referred to American Antiquity and the Chicago Manual of Style
for matters of style and reference. Textual references should be similar to
"Smith 1964: 78-80; Jones 1969: Pl. II, b). No footnotes. In the "Refer-
ence Cited" section, at the end of an article, do not abbreviate names of ser-
ies or journals. Books need full title, publisher's name, and place.
All illustrative material--line drawings (in India ink) and photographs--
should be included in one numbered series of "Figures" suitable for a maxi-
mum reduction to 6 1/2 by 8 1/2 inches. Parts of figures are designated by
lower case letters. Figures must be accompanied by brief descriptive cap-
tions. All tabular material must be included in separately numbered "Tables. "
Twenty-five reprints without covers of each article are furnished free
to each author or group of authors. Additional copies, with or without covers,
may be purchased but the editor must be informed of the number desired when
the article is offered for publication.