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.
Adelaide K. Bullen, Editor
l'orida AlMtropological Society
94.e 9[otdda c1ntdo/io'oyiit
Vol. VIII March, 1955 No. 1
STRATIGRAPHIC TESTS AT'BLUFFTON, VOLUSIA COUNTY,
FLORIDA ................... ............ Ripley P. Bullen 1
ARCHAEOLOGY ON ROCKY POINT, FLORIDA ..... William W. Plowden, Jr. 17
THE HORSESHOE ISLAND SITE, LAKE COUNTY, FLORIDA
................... ................ .. Paul and Grace Cabeen 23
RECENT TESTS AT THE BATTERY POINT SITE, BAYPORT,
HERNANDO COUNTY, FLORIDA ................. Gordon C. Coates 27
CONTRIBUTORS TO THIS ISSUE ............................... 31
STRATIGRAPHIC TESTS AT BLUFFTON,
VOLUSIA COUNTY, FLORIDA
Ripley P. Bullen
The Bluffton site, sometimes referred to as Dexter's Point, is one of
the famous fresh-water shell middens which used to border the St. Johns
River in Florida. These large middens have nearly disappeared with the use
of shells for road work. The mound and ridge at Bluffton are now demolish-
ed. Bluffton is located on the eastern side of the St. Johns River about
four miles southeast of Astor.
This paper is limited to the results of joint salvage excavations made
in the fall of 1951 by the Florida Park Service and the Florida Geological
Survey. Finds came from the earlier levels of the site the fiber-tempered
and preceramic zones. Evidence of later occupations was noted but is not
discussed here. Some unique specimens from this site are described else-
where (Neill, 1954).
The Bluffton site is mentioned by Bartram, Wyman, Le Baron, and
Moore (Goggin, 1952, p. 89). Wyman called the site Orange Bluff and de-
scribed it as a rounded shell-mound or midden which rose twenty feet above
the river with a long shell ridge to the southeast. A house stood on the
highest part of the midden, and two burial mounds adjoined the ridge. Both
the large shell-mound and the ridge were covered with a grove of orange
trees (Wyman, 1875, p. 37).
Our work was done in the large shell-midden mentioned by Wyman.
Fortunately, presence of the house had delayed shell removal in this area
until the summer of 1951. Then the house was moved to facilitate work.
After the dragline had begun operating on the highest part of the site,
William E. Edwards (then with the Florida Geological Survey) arranged
with John W. Griffin (then with the Florida Park Service) for salvage archae-
ology. Cooperation of the County Commissioners was secured by Griffin.
The main test, Trench A, was excavated by Bullen (then with the Florida
Park Service) while a smaller test, Test H, was dug by Edwards.
EXCAVATIONS TRENCH A
Trench A (consisting of Squares Al to A5) was located on the southern
side of the highest part of the remaining midden where the dragline had left
a fairly steep bank. This trench was twenty-five feet long with its major
axis east-west and approximately perpendicular to the river. Width of the
trench varied from five-and-a-half to seven feet, depending on the edge left
by the dragline scoop. This trench was carried downward to a depth of from
sixteen to seventeen-and-a-half feet.
The southern bank of the midden sloped outward with depth while the
inner wall of our trench had to slope slightly outward for safety. Therefore,
the excavation gradually became displaced southward as we dug deeper. At
a depth of twelve feet the trench was twenty-two-and-a-half feet long and
six to nine feet wide. Below thirteen feet, it became nearly impossible to
remove the excavated debris so we changed from an open to a closed trench.
At a depth of sixteen feet, the area being excavated was seventeen by four
When we started this trench, six stratigraphic zones or layers were
evident in the face of the midden. These were numbered with Roman
numerals for reference purposes. They sloped downward, a little, away
from the river. After we had dug into the midden, we found layers II, III,
and IV sloped steeply downward away from our original face.
Table 1 presents a stratigraphic analysis of the results from Trench A.
In this table we have tried, by making some allowance for the downward dip
mentioned above, to show a three dimensional picture in two dimensions.
Layer I was brownish in color and composed of crushed and whole
mussel shells mixed with ashes and dirt. Fired areas and small, shallow
pits were encountered. This layer was partially disturbed. To suggest the
extent of disturbance, we have shown in Table 1 depths at which glass or
iron objects were found. Usually, this disturbance did not exceed nine
inches but in Square A2 and part of Square A3 a deep ditch had been dug
for a drainage pipe. No doubt these disturbances reflected the construction
of the house and also the planting of orange trees mentioned by Wyman. Un-
doubtedly, some St. Johns sherds attained their relatively deep provenience
as a result of these disturbances.
Layer II appeared from the original profile to be formed of clean shells.
With excavation it was found to consist of light brown dirt, shells mixed
with ashes, and occasional lenses of charcoal. At a depth of forty-two
inches, in Square A4, was a fireplace which measured five feet across and
twelve inches vertically. It was filled with ashes and a few burnt shells.
At and near its base, heat had cemented together the surrounding mussel
Layer I, as suggested by the original profile, was composed of shells
mixed with ashes. A small shallow fireplace, two feet across and consist-
ing of ashes resting on heat-cemented mussel shells, was found in Square
Al at a depth of five-and-a-half feet at the base of Layer HI. A similar
fireplace, thirty inches across, was in Square A4 between depths of six and
Layer IV was formed of relatively clean, loose shells. Here large snail
shells (Pomacea) were more common than mussel shells.
Layer V in the face was formed of burnt and frequently cemented whole
mussel shells mixed with ashes. Excavation confirmed this but indicated
a much more complex situation. In Square Al, Layer V could be divided
into five zones as burnt shells were separated by two zones of brown dirt
and sand. The upper dirt zone was by far the more sandy of the two. Rem-
iniscent of a house floor, it was about five to six inches in thickness. At
one place at the base of this sandy-dirt zone was a pink area, eight inches
across, which contained nineteen lumps of red ochre-cemented sand. The
intermediate shell zone was formed of small snail shells (Vivipara) mixed
with a few cemented mussel shells. The lower dirt zone consisted of dark
brown soil mixed with small snail shells. The third or lowest shell zone
was composed of mussel shells mixed with some shells of the small snail.
To a large extent these shells were cemented together by heat.
The above situation was also found in Squares A2 and A3 and the first
foot of Square A4. Apparently, this complex represented a very large hearth
or area used for the cooking of shellfish, some sixteen feet across, which
had, at times, been covered with dirt or sand. The remaining area of Layer
V consisted of unburnt shells mixed with dark brown dirt.
Layer VI consisted of shells of the small snail (about 80%) plus those
of the large snail and the mussel, irregularly divided into zones by lenses
of brown dirt. In Square A5, near the top of this layer, was an oval charcoal
area, two by three feet. Mixed with the charcoal were a great number of
small broken rocks. This small deposit was only about two inches in thick-
ness. Between depths of twelve and thirteen feet, many more small broken
rocks were noted. They tended to concentrate in small areas of brown color.
Square A3 was taken down to a depth of seventeen-and-a-half feet but the
base of the midden was not reached.
VERTICAL DISTRIBUTION OF SHERDS, BANNERSTONE, AND SHELL TOOLS
IN TRENCH A AT BLUFFTON, VOLUSIA COUNTY, FLORIDA
SQUARES Al A2 A3 A4 AS
SPECIMENS 6 a 0
a || a I i
.'a' s s i s s i g a *
LAYERS A o 6 a . .
0 0 o
7 3 1 f
1 10 2 1 for 3 2 1 1 1 65 4 1 17 13 4 2
10 1 12 Drainage 1 3 1 24 4 1 24 3 2
1 5 8 1
14 7 1 1
20 6 1
3 7 6 4 1 1 8 3 24
6 6 4 B.G. 9 1 4
4 7 12 4 16 1 7
1 7 10 2
5 10 s
S trombus celt 1 8 S.C. 4
1 _1_ 9 S.C.
7 Strombus celt Busycon gouge 4
8 Bmsycon gouge
11 Btnetone V
Busycon gouge VI
1 1 _______ _______ --------------^- --
___Bus co gouge
N E X C A V A T E D
B. G. Busycon gouge
1 1 a
1 1 1
S. C. Strombus celt
DISCUSSION OF TRENCH A
Stratigraphic results of the work in Trench A are given in Table 1. As
has been mentioned, Layer I was disturbed to some extent. St. Johns sherds
and objects of glass and iron found deep in this layer presumedly gained
their relatively deep provenience as a result of these disturbances. It
seems only proper to omit them from our consideration.
Layer I, except for its first six inches, belonged to a decorated fiber-
tempered pottery period. Both Orange Incised and Tick Island Incised
sherds were present in significant quantities as well as fragments of Orange
Plain vessels. Layer II, on the other hand, clearly represented a plain fiber-
tempered period. Such a period has been found by Waring for coastal
Georgia (Griffin, 1952, p. 366), but this is the first definite evidence for
such a period in Florida.
The first appearance of plain fiber-tempered pottery in Trench A ap-
pears to be correlated with the junction between Layers H and m. It is not
certain whether the sherds we show in Table 1 for the upper part of Layer
III were trampled into it by the first pottery makers or whether pottery was
introduced before the completion of the deposition of Layer III.
There is little that can be said about the preceramic zones. Excava-
tion indicated these deposits to be rather complicated in structure with
various hearths or areas of cemented shells and intervening dirt deposits.
Whether these dirt deposits represented temporary abandonment of the site
or the spreading of dirt for sanitary reasons may be debatable.
The fragment of a steatite bannerstone (Fig. 3, E) in the upper part of
Layer VI should be noted. It was one foot below the cemented base of
Layer V. It is also important to note the presence of Busycon gouges re-
latively early in the preceramic portion of the Archaic period. Lack of
Strombus celts in the deeper preceramic deposits may be a sampling error
and not necessarily indicate a relatively late introduction of this tool.
Continuation of the manufacture of both Busycon gouges and Strombus
celts in the plain fiber-tempered period is clearly indicated in Table 1.
EXCAVATIONS TEST H
Test H was located about forty feet northwest of Trench A at a place
where the dragline had left a protruding corner about six by eight feet in
extent. Here many relatively thin layers of strata could be seen in the
exposed face of the midden. These strata sloped very steeply, sometimes
approximating a forty-five degree angle from the horizontal. Feeling the
large number of strata offered an opportunity to secure evidences of cultural
changes, Edwards removed this corner by layers or parts of layers (strata).
Work was difficult due to the steepness of the strata.
These strata lead upwards and were cut off by a thin disturbed superior
zone. Apparently, the top of the midden mound had been cut off or leveled,
probably at the time the house was built. Layer III of Trench A could be
traced along the face of the midden and could be shown to dip under these
strata. Hence, all zones of Test H should pertain to a ceramic period, as
proved to be the case. Otherwise, the relationship of these strata to layers
of Trench A could not be determined.
Edwards defined eleven layers between the surface and the bottom of
his test at a depth of sixteen-and-a-half feet. Each layer included one or
more strata. Material was bagged by him by layers and strata within layers.
It was analyzed by Bullen.
DISCUSSION OF TEST H
Four fragments of glass and a nail were found in the top layer. Except
for one decorated sherd (to be discussed later), the upper ten layers of Test
H pertained to a plain fiber-tempered pottery period. Orange Plain sherds
were found as follows: Layer 1, 18 sherds; Layer 2, 120 sherds; Layer 3,
81 sherds; Layer 4, 27 sherds; Layer 5, 37 sherds; Layer 6, 223 sherds;
Layer 7, 15 sherds; Layer 8, 49 sherds; Layer 9, 38 sherds; Layer 10, 41
sherds; Layer 11, none. Edwards assumed Layer 11 to have been pre-
These data suggest extensive deposits of a plain fiber-tempered period
which should equate temporally with Layer II of Trench A. The only diffi-
culty is that Edwards found one Orange Incised sherd (referred to above) on
top of his spoil while he was shovelling material which he had looked
through and which came from Layer 10. His notes state that, two minutes
earlier, one shovelful had been removed from Layer 9 (presumedly to straight-
en the face; also presumedly this one shovelful had not been looked through).
As Edwards was at this time working at the base of a fifteen foot verti-
cal face and did not find the incised sherd in the material through which he
had looked but on top of his spoil pile, I am inclined to think it fell from
the top of the midden where such sherds were known to have been present.
In any event, he looked through a great deal of midden material in a forty-
square-foot area, sixteen feet deep, without finding another decorated sherd.
Some should have been present if he was not working in deposits of an un-
decorated pottery period.
Orange Plain (fiber-tempered) sherds from our excavations exhibit well-
smoothed inner and outer surfaces which may have received a clay wash.
Vessel walls are relatively thin, usually measuring a quarter of an inch in
this respect although extreme examples as thin as one-eighth and as thick
as one-half of an inch can be found. Rims have either simple rounded or
slightly flattened lips. They are as thick as, or slightly thinner than,
vessel walls. Flat bases vary from one-quarter to one-half of an inch in
Apparently, vessels were made with both circular and square (or
rectangular) mouths. The latter had one side about eight inches in length.
In the middle of this side the lip was pulled outward to form an ear (Fig. 1,
B-D). Characteristically, a pit or depression is found on the inside of
these ears (Fig. 1, C). These straight-sided containers vary from two to
four inches in depth (three inches, most common). Walls are straight and
Less information is available for the circular-mouthed vessels. One
reconstructed, flat, circular base measures seven-and-a-half inches in
diameter. A few sherds indicate an inturned rim (Fig. 1, A). A unique
"pitted" ear is illustrated (Fig. 1, E). It did not come from our excava-
tions but was found at the site by Mr. H. James Gut of Sanford, Florida.
Orange Incised vessels were made of the same paste as plain fiber-
tempered containers. Walls average a little thicker, about three-eighths of
an inch. Fragments of two Orange Incised vessels from Trench A indicate
a round mouth, twelve inches in diameter, and a vessel depth of about
four-and-a-half inches. Except for one variant (Fig. 2, D), designs consist
of vertically placed, nested diamonds bordered at the rim by straight lines
paralleling the lip (Fig. 2, E-G).
Tick Island Incised vessels were made of the same paste as plain
fiber-tempered containers and had the same relatively thin walls. One sherd
suggests a straight-sided container while the others pertain to circular-
mouthed vessels. Incision and punctation is usually bold and bordered by
raised edges. Of the illustrated sherds (Fig. 2, A-C), the first was the most
common; the second, a minor variant; and the third, unique.
One broad, stemmed spear point or knife was found in Square A4 at a
depth of fifteen to eighteen inches (Fig. 3, D). It appeared to be in an un-
disturbed deposit and should belong to the decorated fiber-tempered period.
0 I 2
Fig. 1. Top and side views of Orange Plain Sherds.
A, circular vessel with inturned rim; B-D, straight-sided vessels with
ears; E, unique ear, from circular vessel (not from excavations).
0 I 2
Fig. 2. Decorated fiber-tempered sherds.
A-B, Tick Island Incised; C, unique Tick Island Incised; D, unique
Orange Incised; E-G, Orange Incised.
0 I 2
Fig. 3. Bone, stone, and shell tools.
A, splinter awl; B, fragment of bone pin; C, fragment of bone awl;
D, stemmed point; E, fragment of steatite bannerstone; F-H, Busy-
con gouges; I, Strombus celt.
Half of a steatite bannerstone was found in the upper part of Layer
VI of Trench A (Fig. 3, E).
Two hammerstones were found in Test H, both in the lowest excavated
pottery layer, Layer 10.
A lump of red ochre (sandy in composition) was found in the highest
six-inch level of Layer V, Trench A. Nineteen similar lumps were found in
an eight-inch-diameter area a foot deeper in the same layer.
Other "stone" objects include a few chert chips (Layer I, Trench A),
various rocks, fragments of marl, lumps of clay, and fossils. These were
irregularly scattered throughout the deposits from top to bottom. Pieces of
marl and clayey concretions were, however, heavily concentrated in lower
levels, Layers V and VI, of Trench A. A piece of coral rock also came from
Layer VI of this trench.
Fragments of fossilized bones were found in all layers of Trench A.
Identifiable fragments include manatee ribs in Layers II and V, deer in
Layers II and III, and the cuboid of a camel in Layer II. Three fragments
of fossilized turtle bone, apparently Trachemys sculpa (0. P. Hay), were
found in Layer VI. In Test H, a scoot of an extinct glyptodont (Holmesina,
sp.) was found in Layer 2 and a tooth of an extinct horse in Layer 3.
These remains of an early fauna differed from food bones found in the
midden in that they were darkish brown in color, moderately mineralized,
broken instead of split, and slightly rounded at sharp edges. They were
similar in appearance to similar fragments found in many fossil localities
in Florida. Undoubtedly, they were picked up by Indians in the bed of the
St. Johns River and taken by them to the midden along with fresh-water
Six fragments of worked bone were found in the lower zones of Test H.
In Trench A, fragments of bone pins or awls and pieces of worked bone
were found in all layers except Layer IV. Illustrated are a bone pin from
Layer I (Fig. 3, B), a splinter awl from Layer V (Fig. 3, A), and a bone
awl from Layer VI (Fig. 3, C).
A sprinkling of fragmentary marine shells was found throughout the
deposits. The quantity is so small, fifty-one for Trench A and twenty-four
for Test H including shell tools, it seems likely all marine shells found at
the site were brought there either as tools or ornaments or as raw material
for the manufacture of such objects. Included are Strombus gigas, Busycon
carica, Busycon perversum, Ostrea virginica, Venus mercenaria, Noetia
ponderosa, Dinocardium robustum, Arca campechiensis, Arca transversa,
and Tonna galea shells. Noetia and Dinocardium shells had worn edges.
The provenience of shell tools, Busycon gouges and Strombus celts,
in Trench A is indicated in Table 1. Three Busycon gouges and two
Strombus celts were found in Test H. Of illustrated specimens the Strom-
bus celt (Fig. 3, I) and Busycon caricaa) gouge (Fig. 3, H) came from Test
H. The other two Busycon gouges (Fig. 3, F-G) are those indicated in
Table 1 for Layer VI of Trench A. They are the oldest tools we found.
The large midden at Bluffton was predominantly composed of varying
amounts of large (Pomacea paludosa, Say) and small (Viviparus contectoides
goodrichi, Archer) snail and mussel (Elliptio crassidens, Lamark, and
Lampsilis avadontoides floridensis, Lea) shells. While admixture frequently
occurred, there was a tendency for these shells to be roughly stratified,
possibly reflecting changing seasons, ecological conditions, or appetites.
A third mussel (Microma viber, Lea) was present in small amounts. The
land snail (Euglandina rose, Ferussac) was found in appreciable quantities.
These fresh-water shellfish represented the staple food of the commun-
ity. This diet was substantially supplemented with mammals, birds, reptiles,
and a few fish. Deer bones were by far the most common bones. They out-
numbered turtle bones, the next most common, by a ratio of three to one.
Turtles included the common box turtle (Tarrapens carolina), the gopher
(Gopharus polyphemus), the soft shell turtle (Amyda farox), the snapping
turtle (Chelydra sarpantina), and the Florida pond turtle (Pseudemys
floridans). Of these the box turtle, probably because it is small and does
not produce much food, was the least common.
The next most common animal was the raccoon. Other, relatively
rare, animal bones include those of the alligator, rabbit, and dog. The
only dog bones we found were in Layers III, V, and VI of Trench A. Bones
of both the large and small dog were found in Layer VI so that these two
races of dogs were present at least as early as the time of that deposit.
Bird bones include those of the turkey (Meleagris gallopavo), various
ducks (lesser scaup, Aythya affinis; ring-necked, Aythya collaris; and mal-
lard, Anas platyrhynchos), the pied-billed grebe (Podilymbus podiceps),
and the American coot (Fulica americana americana). While the duck,
grebe, and coot bones were limited to preceramic levels, only recent forms
Vertical distribution of food bones did not seem to produce significant
data. All modern species found were certainly present as early as Layer
VI of Trench A or, for a guess, around 4000 B.C. No extinct forms were
represented among food bones. It was noted, however, that the quantity
of food bones per unit of midden debris was less after the introduction of
pottery. This fact may represent better utilization of animal food after
stews could be prepared but other factors could also produce this result.
Bird bones were identified by Dr. Pierce Brodkorb of the University
of Florida. Other food remains and fossil bones were identified with the
assistance of Walter Auffenberg of the University of Florida, Edmay V.
Flowers of the Florida State Museum, H. James Gut of Sanford, Florida,
and Dr. Wilfred T. Neill of Ross Allen's Reptile Institute, Silver Springs,
COMPARISONS AND CONCLUSIONS
The main excavation at Bluffton, as shown in Table 1, uncovered a
zone of plain fiber-tempered pottery between a decorated fiber-tempered
zone and extensive preceramic deposits. Presence of a plain fiber-tem-
pered period is strongly supported by Edwards' work in Test H.
The preceramic Archaic people who lived at Bluffton consumed vast
quantities of river mussels and fresh-water snails which they collected
from Lake Dexter and the St. Johns River. This repast they amply supple-
mented with deer, turtle, and other animal food.
They imported from the coast a few marine shells either in the form
of tools and ornaments or as raw material for the manufacture of such pro-
ducts. Busycon gouges, Strombus celts, and bone awls were common
items in their tool kits. A bannerstone implies they used the spear and
spear-thrower. This bannerstone, being made of non-Floridian material,
represents an exotic specimen.
There seems to be no change in way of life with the introduction of
pottery with the possible exception that, as a result, there may have been
more efficient utilization of animal food. Certainly the same shell gouges
and celts were used after the introduction of pottery.
Plain fiber-tempered pottery seems to appear at Bluffton as a "full
blown" trait with both round and square (or rectangular) shapes. Frequent-
ly, the latter had eared appendages. In spite of this "finished" appearance,
this plain fiber-tempered ware should represent the first ceramic period of
Decorated fiber-tempered pottery from Layer I in Trench A was limited
to Tick Island Incised and one motif of Orange Incised. In this respect, as
well as others, our results differ substantially from those secured by
Ferguson at South Indian Field and by Griffin and Smith at the Cotton site.
South Indian Field is located on a low "island" surrounded by swampy
land, two miles east of the upper St. Johns River and about eighty-five
miles from Bluffton. The Cotton site is located on the west bank of the
Halifax River about thirty miles from Bluffton. The following tabulation
compares fiber-tempered period traits for these three sites. In the case of
South Indian Field, the data are taken from the report on that site (Fer-
guson, 1951). A report on the Cotton site is forthcoming but not available
at the time of writing (Griffin and Smith, n.d.). Comparisons in the case
of the Cotton site are the results of the author's examination of pottery
from that site which is at the Florida State Museum.
South Indian Field
Vessel walls, .7 to 1.8 cm.,
usually 1.0 to 1.2 cm. thick
Vessel bases, .8 to 1.3 cm.
Lips tend to be a little
thicker than walls
No plain fiber-tempered zone
No Tick Island Incised
Many motifs of Orange Incised
Some very wide lips
Some decorated lips
Steatite sherds relatively late
Dog bones relatively common
Walls, .6 to 1.6 cm.
Bases, 1.2 to 1.9 cm.
Lips tend to be a little
thicker than walls
No plain fiber-tempered zone
No Tick Island Incised
Many motifs, Orange Incised
Some very wide lips
Some decorated lips
Steatite sherds present
Dog bones relatively common
Walls, .3 to 1.2 cm.,
usually .6 to .7 cm.
Bases, .6 to 1.2 cm.
Lips tend to be a little
thinner than walls
Some eared appendages
Plain fiber-tempered zone
Tick Island Incised present
Only one motif Orange Incised
No wide lips
No decorated lips
No steatite sherds found
Dog bones absent or rare
As shown above, South Indian Field and the Cotton site share many
similar traits. The excavated portion of the Bluffton site is rather diver-
gent from the other two. Major differences are a plain fiber-tempered period,
relatively thin vessel walls, eared appendages, presence of Tick Island In-
cised, the limitation of Orange Incised to one motif, and the absence of
wide, decorated lips. These differences would seem to have chronological
As Bluffton produced a plain fiber-tempered period, not present at the
other sites, and as the decorated pottery seems to be an added feature
without a change in lips, it would seem the excavated portion of the dec-
orated fiber-tempered zone at Bluffton was earlier than both South Indian
Field and the Cotton site. Presence at South Indian Field and the Cotton
site of Orange Incised with design motifs which were carried over onto
chalky paste to form St. Johns Incised vessels and the absence of such
designs in the excavated portion of the Bluffton site strongly support the
thesis that the latter deposits were the earlier.
If this is correct, the fiber-tempered period of Florida may be divided
into five stages, which might be called Orange 1 to 5. Orange 1 would be
a plain fiber-tempered period similar to that found at Bluffton. Orange 2,
like Layer I of Trench A at Bluffton, would have Tick Island Incised and
Orange Incised but no wide, decorated lips. Orange 3, as at South Indian
Field and the Cotton site, would have a wider range of design motifs (in-
cluding those also found on St. Johns Incised vessels) and some wide,
decorated lips. Orange 4 would be the same as Orange 3 except for the
addition of steatite sherds (Bullen, 1954). Orange 5 might, logically, refer
to undisturbed deposits in which sand-tempered or chalky pottery, as well
as fiber-tempered sherds, were found. It would represent the closing stage
of the Orange Period when other pottery containers were superceding those
made with fiber as tempering material.
Bullen, Ripley P.
1954. "Culture Changes during the Fiber-Tempered Period of Flor-
ida." Southern Indian Studies, Vol. VI, pp. 45-48. Chapel
Ferguson, Vera M.
1951. "Chronology at South Indian Field, Florida." Yale Univer-
sity Publications in Anthropology, No. 45. New Haven.
Goggin, John M.
1952. "Space and Time Perspective in Northern St. Johns Arche-
ology, Florida." Yale University Publications in Anthro-
pology, No. 47. New Haven.
Griffin, James B.
1952. "Radiocarbon Dates for the Eastern United States." In Arche-
ology of the Eastern United States (J. B. Griffin, ed.), pp.
Griffin, John W.
1952. "Prehistoric Florida: A Review." In Archeology of the
Eastern United States (J. B. Griffin, ed.), pp. 322-334.
W., and Hale G. Smith
"The Cotton Site." To be published by the Department of
Anthropology and Archaeology of Florida State University.
Neill, Wilfred T.
1954. "Artifacts from the Bluffton Midden, Volusia County, Flor-
ida." The Florida Anthropologist, Vol. VII, No. 1, pp. 11-17.
1875. "Fresh-water Shell Heaps of the St. John's River, Florida."
Peabody Academy of Science, Fourth Memoir. Salem.
FLORIDA STATE MUSEUM
ARCHAEOLOGY ON ROCKY POINT, FLORIDA
William W. Plowden, Jr.
Rocky Point is a small peninsula on the northeastern shore of Tampa
Bay, in Sections 11, 12, and 14, Township 29 south, Range 18 east. The
peninsula is bisected by State Highway 60, also known as Courtney Camp-
bell Causeway. The peninsula is primarily composed of salt marsh and
mangrove flats, of which the vegetation is mainly Rhizophora mangle, with
a scattering of Avicennia nitida, the black mangrove. Along the shoreline
are to be found a number of sand ridges which apparently indicates an old
beach line. Aboriginal occupation with but two exceptions, Sites IV and V
on Figure 1, is found on this ridge. Flora on all of these shell mounds is
similar, consisting of Cabbage Palm (Sabal palmetto), Red Cedar (Juniperus
silicicola), and Oak (Quercus myrtifolia).
Earliest archaeological accounts of the area are Stearns (1872) and
Shepard (1886). Stearns reports shell heaps, large and small, with only one
of the heaps of sufficient size to be dignified by the name of mound. Stearns
states that this mound was 15' high and covered an area of half an acre,
located one mile from the end of Rocky Point. It was composed entirely of
shell and produced no pottery. However Stearns reports finding flint arrow
heads. This mound is probably the site marked IV on Figure 1.
Shepard reports pottery with shell imprints, shell celts, beads from
columellae, pendants, bone implements, and arrow heads.
Rocky Point after 1885 missed archaeological attention until 1949 when
Willey mentioned the area and gave a classification of a collection of sherds
in the United States National Museum (Nos. 32466632-3246663). The seven-
teen sherds in Willey's collection apparently were carefully picked, for ten
types are represented. It is interesting to note that Willey's classification
lists one shell-stamped sherd, bearing out Shepard. I am inclined to think
that the sites of Willey and Shepard are the same. There is also a slight
suggestion that this site may be that on the southern tip of Rocky Point
marked I, in Figure 1.
During the summer of 1953, in connection with a site survey of the
Fig. 1. Location of sites on Rocky Point, Florida.
Tampa Bay region, I was struck with the number of sites on this small
peninsula and the lack of clarity in the existing references.
The sites on Rocky Point, with the exception of Rocky Point I, have
as yet no designation either in Florida Park Service files or in files of the
Rocky Point I (Hi-7)*
This site on the extreme southern point of the peninsula varies from
20' to 100' in width and is approximately 3,000' long in a crescent shape.
It has been in the past a source of road building material, and in places
the mound has been removed to bedrock. However, the extreme northwest-
ern point of the crescent is in original condition and has a height of 3'6".
This mound or ridge is composed of oyster shell with a mixture of Melon-
gena sp. with lenses of beach sand scattered throughout. Cultural material
is scarce. The collection is composed of:
Glades Plain ..................3
Sherd tempered ................. 1
Spanish Olive Jar . . . 1
Several unworked flint chips were also picked up.
Rocky Point II
This site is located midway up the peninsula on the eastern shore.
The site is on an old sand dune but in two places widens into a cabbage
palm hammock. The site is 30' to 40' wide with an average depth of 3'
and length of 800'. It is in a relatively undisturbed state. Cultural mate-
rial is scarce. The collection is composed of:
St. Johns Plain ................. 1
Weeden Island Plain .............. 2
Glades Plain .................. 1
Unclassified check stamped ......... 1
Flint work was represented by one scraper.
*Hi-7 is the site number in the joint archaeological site survey of the University
of Florida and Florida State University.
Rocky Point III
Here a number of sites apparently have been joined together in the
making of an early road. In places shell can be observed across the ditch
from the present road. Material was collected only from these outlying por-
tions as the material on the road could have come from any number of sites
in the region. Rocky Point III, 750' long and from 40' to 100' wide with an
approximate height of 2', is at the extreme northwestern corer of the small
bay which separates the peninsula from the mainland on the east.
Sherd tempered ................ 2
Belle Glade Plain ............. 3
Weeden Island Plain ........... 1
Glades Plain ................ 22
Pasco Plain ......... ....... 1
Unclassified check stamped ...... 1
Worked shell is represented by one Melongena hammer.
Rocky Point V
This site is located on a small key 100 yards to the west of Rocky
Point IV. It is low, around 1' high, 150' long and 20' wide. No sherds
were found at this site.
From the material collected and without stratigraphic work, one would
hesitate to draw positive conclusions concerning the dates of these sites.
Two of them, sites I and III, could be late if the hypothesis drawn by Bullen
and Griffin on their work at Amelia Island on sherd-tempered material can be
extended to this area. Willey's collection gives a Weeden Island II dating
to one of the sites. Site III may be of the Perico Period.
The olive jar sherd can probably be ignored because of the fact that
Cuban fishermen were camping in these waters up into the nineteenth
century. Stratigraphic testing of the Rocky Point area is planned by the
Tampa Bay Chapter of the Florida Anthropological Society.
Bullen, Ripley P.
1952. "Eleven Archaeological Sites in Hillsborough County, Flor-
ida." Report of Investigations No. 8, Florida Geological
Bullen, Ripley P., and John W. Griffin
1952. "An Archaeological Survey of Amelia Island, Florida." The
Florida Anthropologist, Vol. V, Nos. 3-4, pp. 37-64.
1886. "Shell Heaps and Mounds in Florida." Annual Report of the
Smithsonian Institution for 1885, pp. 902-6. Washington.
Stearns, R. E. C.
1872. "Remarks on Mounds and Shell Heaps of Tampa Bay, Florida."
Proceedings of the California Academy of Science, Ser. 1,
Vol. 4, pp. 214-5.
Willey, Gordon R.
1949. "Archeology of the Florida Gulf Coast." Smithsonian Mis-
cellaneous Collections, Vol. 113. Washington.
UNIVERSITY OF NEW MEXICO
ALBUQUERQUE, NEW MEXICO
THE HORSESHOE ISLAND SITE,
LAKE COUNTY, FLORIDA
Paul and Grace Cabeen
The Horseshoe Island site is located near the southern end of Lake
Harris. The backbone of the island is about ten feet above the lake. This
elevation grades down to about four feet on the crescent-shaped arms. We
have visited this site at least once every year from 1950 to 1954. As shown
on the map (Fig. 1), occupation seems to be limited to one restricted area.
While most of our collection from Horseshoe Island was found by search-
ing the surface, we did dig a little in the higher part of the eroded area. Our
finds include eight complete points (Fig. 2, A-H), fifteen broken points, a
drill (Fig. 2, I), a sandstone grindstone (Fig. 2, L), a large scraper (Fig. 2,
N), and several medium-sized scrapers.
Of particular interest are a stone club-head (?) and a stone plummet-
like object. The latter (Fig. 2, K) is egg-shaped, circular in cross section,
and made of a white stone. In one end, the maker started to drill a hole.
The club-head (Fig. 2, J) is grooved on one side (presumedly for a shaft)
while the more-or-less circular edge has four smaller grooves (marked a, b,
c, and d in the illustration of this object). It seems likely these grooves
were made for lashings. This object may have been used as a club-head or
as a balance weight for a spear thrower.
Three pieces of conch shells probably represented shell tools.
Our pottery collection from this site includes 450 plain, sand-tempered
sherds, one similar sherd with incision on the interior, twenty-three St.
Johns Plain, and fourteen St. Johns Check Stamped sherds. Especially
noteworthy is a pottery gorget (Fig. 2, M).
Half of a lower jaw, with teeth intact, and eleven fragments of human
skull bones were found a short distance from the main part of the site (Fig.
1, human bones).
Except for one medium-sized scraper and a half dozen St. Johns
Check Stamped sherds, found on the beach near the stone outcroppings,
all our specimens came from the fairly small area marked "SITE" on the
map (Fig. 1). For this reason we feel the Horseshoe Island site was the
residence of one family. It is possible, however, the site was used spas-
modically over a long time and that our collections represent two or more
HOLLY HILL, FLORIDA
1- cypress trees
Fig. 1. Map of Horseshoe Island, Lake Harris.
0 1 2 3
Fig. 2. Artifacts from the Horseshoe Island site.
A-H, projectile points; I, drill; J, multi-grooved stone; K, plummet-like
stone object; L, sandstone grindstone; M, pottery gorget; N, large stone
RECENT TESTS AT THE BATTERY POINT SITE,
BAYPORT, HERNANDO COUNTY, FLORIDA
Gordon C. Coates
Stimulated by reports of collections from the Battery Point site pub-
lished in The Florida Anthropologist (Bullen and Bullen, 1953; 1954), a
survey was made to ascertain if anything remained which might add to our
knowledge of this important site. The digging of over thirty test holes
showed that productive excavation could only be made in the beach between
high- and low-tide lines.
A small test, about six feet long and two feet wide, was dug there.
Specimens were removed in such a way that records could be kept. Due to
constant water and tides, digging was very difficult and not always
Digging through the pumped-in debris and the underlying sawgrass sod
and muck to a total depth of about eighteen inches, we reached the top of a
layer of chips, stone artifacts, and pottery. This layer, which we will refer
to as Layer A, was about five inches thick. Below it was another zone, to
be referred to as Layer B, where we found burned bones, ashes, and other
debris. Layer B also contained pottery and a few stone artifacts. It was
about six to eight inches in thickness. Digging further, we found dark
yellow, sterile sand.
In this test, basally-notched points (Fig. 1, C) concentrated near the
top of Layer A. A stemmed point (Fig. 1, D) was found deeper. Adze-like
tools (Fig. 1, K) were more common near the base than near the top of Layer
A. Only one small adze-like tool and a scraper were found in Layer B.
Over a hundred fragments of pottery were uncovered in each layer. In
Layer A were found St. Johns Incised, St. Johns Plain, Perico Punctated
Type A, Perico Plain, Pasco Plain, Franklin Plain, and sand-tempered
plain sherds. Layer B produced Pasco or Perico Plain, sand-tempered
plain, semi-fiber-tempered plain, and simple-stamped semi-fiber-tempered
sherds. It will be noted that semi-fiber-tempered sherds were limited to
Layer B and decorated St. Johns or Perico sherds to Layer A.
Fig. 1. Artifacts from the Battery Point site.
A-E, projectile points; F, drill; G, plummet; H-J, flake knives; K, adze-like
tools; L, large knife or adze.
The above suggests a tentative sequence for the excavated area. First
we have a time when semi-fiber-tempered, frequently simple stamped, pottery
was used as well as plain sand-tempered and plain limestone-tempered
vessels. Next chalky pottery including St. Johns Incised and heavy, adze-
like tools were introduced. The various Perico decorated types may belong
to this period or may be slightly later. The third change is the noticeable
decrease in the quantity of large stone tools and the introduction of basally-
notched projectile points.
One of the test holes in the beach, about twenty feet south of the trench
just discussed, encountered part of a human skeleton. This test was en-
larged disclosing a semi-flexed burial at a depth of about twenty-four inches.
Burial was on the side with head to the southeast and back to the west. The
back was bent and the legs drawn up somewhat. Pottery was found while
uncovering this skeleton. Decorated pottery included two St. Johns Incised
sherds, one just above and the other immediately below the bones.
Our collection from the Battery Point site includes: twenty-two large
and medium-sized adze-like tools (Fig. 1, K), seventeen small adze-like
tools, a large well-flaked knife or adze (Fig. 1, L), twenty-five projectile
points (Fig. 1, A-E) of which the majority are basally-notched (Fig. 1, C),
two drills (Fig. 1, F), a plummet (Fig. 1, G), four hammerstones, many
small scrapers, over a hundred utilized flakes or flake knives (Fig. 1,
H-J), and several hundred worked fragments.
Decorated pottery types in our collection from the Battery Point site
are Perico Incised, Perico Punctated A and B, Perico Linear Punctated,
Pasco Incised, St. Johns Incised, Deptford Simple Stamped, Deptford Linear
Check Stamped, Deptford Bold Check Stamped, and simple stamped semi-
fiber-tempered sherds. Plain sherds of the same paste as the above types
are more common than decorated sherds except for semi-fiber-tempered
sherds. These usually are simple stamped.
Perico punctations are sometimes round as if made with a pointed
tool, sometimes long and curved as if made with a fingernail. One Perico
Punctated variant has three rows of parallel, rounded punctations a half
inch below the rim. One limestone-tempered sherd is decorated with rough-
ly parallel rows of very small punctations. It may have been shell-marked.
Another limestone-tempered sherd has an incised line parallel to the rim
similar to Weeden Island rim treatment. One "Perico Incised" type of
decoration is on a sand-tempered, Deptford-like paste.
We also found three broken bone tools, a piece of galena, a fragment
of a greenstone celt, a piece of a steatite vessel, and a piece of mica.
The last is about three-quarters of an inch across.
This collection, when added to those described by the Bullens (1953
and 1954), shows the importance of this interesting site. The number of
stone tools found at the Battery Point site is very large. These Indians
used a great many scraping and cutting tools. We have suggested from our
test that some things may be relatively early and others relatively late.
We hope we have added to the picture of an early settlement on the shore
of the Gulf of Mexico.
Bullen, Adelaide K., and Ripley P.
1953. "The Battery Point Site, Bayport, Hernando County, Florida."
The Florida Anthropologist, Vol. VI, No. 3, pp. 85-92.
1954. "Further Notes on the Battery Point Site, Bayport, Hernando
County, Florida." The Florida Anthropologist, Vol. VII, No.
3, pp. 103-108. Gainesville.
ST. PETERSBURG, FLORIDA
CONTRIBUTORS TO THIS ISSUE
Ripley P. Bullen is curator of social sciences at the Florida State
Museum and treasurer of The Florida Anthropological Society.
William W. Plowden, Jr. helped to organize the Tampa Bay Chapter of
the Florida Anthropological Society. He is now studying at the University
of New Mexico.
Paul and Grace Cabeen present their first article on Florida archae-
ology in this issue of The Florida Anthropologist. However, the Cabeens
have recorded their findings for many years.
Gordon C. Coates adds more data to the Bayport findings. He joined
our society in 1954.
FLORIDA ANTHROPOLOGICAL SOCIETY
Wilfred T. Neill, Silver Springs
First Vice President:
Second Vice President:
Charles L. Knight, Tampa
D. D. Laxson, Hialeah
Charles H. Fairbanks, Jr., Tallahassee
Ripley P. Bullen, Gainesville
Adelaide K. Bullen, Gainesville
Frederick W. Sleight, Mount Dora
H. James Gut, Sanford
Charlton W. Tebeau, Miami
Membership in the Florida Anthropological Society is open to everyone
interested in its aims. Dues are $3.00 per year. Student Membership $1.50.
Members receive the Florida Anthropologist, the Newsletter, and other pub-
lications of the Society. Applications should be sent to the Treasurer, who
should be addressed also concerning receipt of publications. His address
is 103 Seagle Building, Gainesville.
General inquiries concerning the Society should be addressed to the
Manuscripts and publications for review should be sent to the Editor at
the Florida State Museum, Gainesville.
Address items for the Newsletter to the President, Research Division,
Ross Allen Reptile Institute, Silver Springs.
LIST OF PUBLICATIONS
FLORIDA ANTHROPOLOGICAL SOCIETY
The Florida Anthropologist
Vol. I, Nos. 1-2, 3-4 Vol. IV, Nos. 1-2, 3-4
Vol. II, Nos. 1-2, 3-4 Vol. V, Nos. 1-2, 3-4
Vol. III, Nos. 1-2, 3-4 Vol. VI, Nos. 1, 2,3,4
Vol. VII, Nos. 1, 2, 3, 4
Each Single No. Each Double No.
Members $ .50 $1.00
Non-Members .75 1.50
Newsletter, Florida Anthropological Society
Nos. 1-28 . . . .. . 15 cents each
Publications, Florida Anthropological Society
No. 1 "Two Archaeological Sites in Brevard County, Florida,"
by Hale G. Smith. 32 pages, 4 plates . .50 cents
No. 2 "The Safety Harbor Site, Pinellas County, Florida,"
by John W. Griffin and Ripley P. Bullen. 42 pages, 4
plates ...........................50 cents
No. 3 "The Terra Ceia Site, Manatee County, Florida," by
Ripley P. Bullen. 48 pages, 7 plates . 50 cents
All publications may be ordered from the Treasurer:
Ripley P. Bullen
103 Seagle Building