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.
FLORIDA ANTHROPOLOGICAL SOCIETY
- I II I I I
VOLe XZII I
THE FLORIDA ANTHROPOLOGIST
Membership is oprn to all interested in the aims of the
SocioTy. The Florida Anthropologist ql publiqh$d by the Florida
Anthropological Society during the months of March, June, Sept-
ember, and December. Subscription 9i by membership; annual dues
are $3.00 ($1.50 for students),of which $1.25 is for subscription
to The Florida Anthropologipt and other publications as issued.
The Florida Anthropologist is published at The Department of An-
thropology, The Florida State University, Tallahassee, Florida.
Second-class mail privileges authorized at Tallahassee, Florida.
Applications and orders for back issues should be sent to
the Treasurer leach single number to members, S.50; each double
number, $1.00; to non-members, $.75, and $1.50 respectively;
Newsletters Noe. 1-35, S.15 each). General inquiries should be
sent to the Secretary, manuscripts to the Editor, and Newsletter
items to the President.
PUBLICATIONS OF THE SOCIETY
No. I. "Nwo Archaeological Sites in Brevard County, Florida,"
by Hale C. Smith. 32 pages, 4 plates............... 0.50
No. 2. "The Safety Harbor Site, Pluellas County, Florida,"
by John W. Griffin and Ripley P. Pullen.
43 pages, 4 plates.................................. 0.50
No. 3. "The Terra Ceia Site, Manatee County, Florida,"
by Ripley P. Bullet, 48 pages, 7 plates............. 0.50
No. 4. "The European and the Indian," by Hale C. Smith,
150 pages, frontispiece, 6 mape..................... 2.00
No. 5. "Florida Anthropology," C. H. Fairbanks, ed......... 1.25
President: John M. Goggin, 312 Peabody Hall, Univ.
of Fla. Gainesville Florida
let Vice President: Marvin J. Brooks, 806 N.1. 15th Ct.
Miami 33, Florida
2nd Vice President: Cliff E. Mattox, 209 Beverly Rd.,
Secretary: William C. Massey Dept. of Anthro-
pology Univ. of Fla. Gainesville,Fla.
Treasurer: Harry L. Coetz, 5012 suwannee,
Tampa 3, Florida
Editor: Charles H. Fairbanks, Box 3051 Fla.
State Univ., Tallahassee. Florida
Executive Committeemen: Irving Rouse, Yale University
E. Y. CuPrnsey,Box 426,Cocoa Beach.Fla.
W*. C. StZrtevant,Bureau of Am. Eth-
nology, Washington 25, D.C.
The Florida Anthropologist, Vol. XIII, No. 1, 1960
THE MONEY'S BEND SITE, Ce'3
Cherokee County, Alabama
Bennie C. Keel
The site is located between the Chattooga and Coosa Rivers in Northeastern Ala-
bama. Salvage excavations were conducted by means of a grant from the Alabama Power
Company to the Alabama Museum of Natural History. Three flexed burials and one
large circular storage pit were excavated along with what appeared to be a palisade
system. This consisted of a shallow open ditch adjacent to a row of large post
holes. Further toward the inside were two long wall trenches parallel to the larg-
er ditch. The bulk of ceramics from all features was Mulber Creek Pgin, very
minor amounts of O'eal Pain, BS Iand la, Besan Sle Stamped, Mulberr
CEBek grd Marked, and i= Ceqek Comlicated Stamped were found. The site was
evidently a pelisaded,fairly permanent village during the Middle Woodland Period.
Abstract ty C. H. F.
Archaeological investigations of the Weiss Dam Reservoir were
made possible through a grant made by the Alabama Power Com-
pany to the Alabama Museum of Natural History. The purpose of
this grant was to salvage such information as possible concerning
the aboriginal inhabitants of this area prior to the destruction
of this information by the construction and the ultimate flooding
of the area by the Weiss Dam. Much of the original purpose of
this grant has been achieved, however two other aspects of archae-
ological field work were products of the investigations, one be-
ing the training of students in archaeological field techniques
and the other being that these excavations will undoubtably lead
to a further understanding of specific problems in Southeastern
It must be said at the outset that the problems common to
all salvage programs were met in the undertaking of these excava-
tions. Due to limited time and funds it was decided that village
sites within the reservoir area would be given primary attention
for it was believed that they would give the most information con-
cerning the prehistoric cultures of the area.
The excavations were under the direction of Mr. David L. De-
Jarnette of the Alabama Museum of Natural History and the Univer-
sity of Alabama and Dr. Charles H. Fairbanks, Associate Profes-
sor of Anthropology and Archaeology at the Florida State Univer-
The Money's Bend Site was excavated by students of The Flo-
rida State University's Summer Field Session during the months of
June and July 1959. The excavation was under the supervision of
I wish to express my gratitude to Mr. DeJarnette for allow-
ing me to study the materials from this site and for permission
to publish my findings, to Ross Morrell and Robert Steinbach for
help in preparing the plates, to members of the Summer Field Ses-
sion 1959 for their labours and especially to Dr. Charles H.
Fairbanks for his guidance, suggestions and encouragement in pre-
paring this report.
This site (see Fig. 1) is located on the south bank of the
Chattooga River some 2.5 miles southwest of Cedar Bluff, Alabama
on property formerly owned by Mr. Austin T. Money prior to being
acquired by the Alabama Power Company as a part of the Weiss Dam
Reservoir. The site is situated some one hundred and fifty yards
south of the river which flows in a southwesternly direction un-
til it joins the Little River and changes its course to a more
southernly direction. The Little River converges with the Coosa
River some one and one half miles south of its confluence with
the Chattooga River.
The site, for the most part, is situated on a small hill
that overlooks the lower ground to the south and to the west.
Some three hundred yards northeast of the site there was at one
time a large clear spring, however, today the spring is silted up
and only a trickle of its clear water is to be observed. The
writer hesitates to point to this spring as a source of water for
the aboriginal village, although there remains such a possibility.
The site has been under cultivation for many years and at
the time of excavation was partially in crops of corn and cotton.
It may be pointed out that such cultivation has certainly pro-
duced its share of sheet erosion.
Prior to excavation a surface collection was made of the
site. Excavation of the site covered a period of fifteen work-
ing days. A series of five 5 feet by 5 feet test squares were
laid out on the northern edge of the site, these test squares
were excavated in arbitrary six inch levels, the bottom of Level
1, which was made up of the plow zone,rested on culturally sterile
yellow clay. This test area covered one hundred and twenty-five
square feet. From it eleven sherds were taken, ten of them of
sand-tempered type 2aa Henry Island Plainl, the remaining sherd a
limestone-tempered type 3a Mulberry Creek Plain. Lithic arti-
facts consisted of five fragmentary projectile points, one side
scraper and one end scraper.
It was noted while making the surface collection that on the
site were several concentrations of shell, these areas were mark-
ed for further investigation, they were staked out and excavated
in two units, the plow zone being one and the underlying midden
the second. Such a procedure was felt to be of value in deciding
where to excavate, since the prevailing conditions of the salvage
program prohibited complete excavation of the site. This pro-
cedure led to the discovery of five special features and three
Three vertical zones were noted on this site:
1. Plow zone light brownish sandy loam.
2. Fill brown to black mottled midden fill.
3. Subsoil a "hardpan" of culturally sterile
The depth and color of the plow zone and midden fill varied
from area to area, this is felt to be due to the depth reached by
the plow, erosion and the amount of organic material present. The
color of the midden fill in some cases was affected by fire.
1 Numbered pottery types follow the classification of Heimlich:
to Leesburg SCALE
to Cove SprIn 0 o 1.
FIGURE 1.- THE MONEY'S BEND SITE,CeV3, CHEROKEE
These zones were established by digging some thirty test holes,
each about a foot square, in each case the plow zone was followed
immediately by sterile subsoil or midden fill.
The most outstanding feature of this site was Feature 1 and
its two associated features 2 and 3 (see Fig. 2).
Feature 1 was discovered in the southwest corner of the site,
the surface above it was one of those areas of the site showing a
surface concentration of shell.
Feature 1 is a ditch with a maximum width of 9 feet, a mini-
mum width of 3 feet, the average depth of the ditch was 2.2 feet,
the maximumcbpth 3.5 feet below the present land surface. The
total excavated length of the ditch was sixty-three feet. A
second characteristic of this feature was an alignment of twenty-
six post molds along the western edge of the ditch which were
spaced on the average 2.4 feet apart and intruded into the sub-
soil to a depth of about one foot on the average, in size the
post molds averaged about four inches in diameter. On excavation
the ditch yielded 495 sherds which are classified in Table II.
The two features associated with Feature 1 lay 10 and 20
feet respectively to the east. Both of these features ran rough-
ly parallel to Feature 1 and are considered to be components of
Feature 1 although no stratigraphic relation was established be-
Feature 2 was a wall trench 0.4 feet wide characterized by
small round depressions at the bottom of the trench. These de-
pressions were evidently made by the poles that constituted the
vertical portion of a wall. The depressions averaged 1.5 inches
in diameter and were spaced on the average one foot apart. Some
forty-three of these small round depressions were found in the
forty-three feet that was excavated of this feature.
Cultural material found in the excavation of this feature in-
cluded three sherds of the limestone-tempered type 3a Mulberry
Creek Plain, one side scraper and one fragmentary side notched
Feature 3 is identical with feature 2 in appearance, the
size of the depressions and the distance between the poles that
constituted the vertical portion of this wall were on the average
the same as those observed in Feature 2. This feature was inter-
sected at three points by running trenched from west to east from
Feature 2 and the feature was excavated a total of nineteen feet.
Aboriginal material obtained from the excavation of this feature
consisted of three limestone-tempered 3a Mulberry Creek Plain
sherds, a medium sized flat based projectile point and a flint
Small amounts of animal bone, some identified as deer, were
found intermittently throughout the excavation of these three
From the general appearance of these three features and the
topography of the site today, they probably formed some sort of a
fortification structure, for. Feature 1 closely follows the wes-
tern edge of the hill and may have in aboriginal times continued
north to the river. The two other components of this structure
can be interpreted as secondary barriers, and since they are so
identical in form and their ends form an overlapping opening they
must be considered as being contemporary aside from the fact that
they both contained cultural material of the same types. A fur-
ther extension of the observable facts is that they tend also to
follow the contour of Feature 1 and may be considered as being
contemporary with the feature also.
From the tabulation of the ceramic material contained in
Table II the chronological position is clearly Middle Woodland,
most certainly belonging to the Gunterlands III phase of this
period in northern Alabama (Cebb and Wilder 1951:270).
Feature 4 was a barrel-shaped storage pit. The diameter at
the top was 2.9 feet, at the widest point it measured 3.0 feet
and constricted to 2.2 feet in diameter at a point some 4.1 feet
below the surface. The fill contained in the pit is especially
interesting, the upper portion was of mottled yellow clay, the
lower portion of soft dark bla6k humic soil.
In the humic portion of the fill were contained two sherds
of the limestone-tempered type 3a Mulberry Creek Plain and one
Type Number Percentage
Guntersville Basin Types
Steatite 1 .14
O'Neal Plain 2a 10 1.49
Henry Island Plain 2aa 43 6.43
Benson Simple Stamped 21 9 1.54
Rudder Comb Incised 2s 1 .14
Unclassified Sand-tempered 8 1.19
Mulberry Creek Plain 3a 565 84.57
Flint River Cord Marked 3h 2 .29
Prospect Red Filmed 31 1 .]4
Flint River Brushed 3n 6 .90
Mulberry Creek Cord Marked 4b 17 2.53
Plain Shell 5a 3 .44
Swift Creek Complicated
Stamped, Early 1 .14
Swift Creel Complicated
Stamped, Middle 1 .14
Total 668 100.00
TABLE 1. Money's Bend Site Pottery
Name Type Number Percentage
Guntersville Basin Types
Steatite 1 .20
O'Neal Plain 2a 4 .80
Henry Island Plain 2aa 16 3.23
Benson Simple Stamped 21 6 1.21
Unclassified Sand-tempered 5 1.01
Mulberry Creek Plain 3a 438 88.48
Prospect Red Filmed 31 1 .20
Flint River Brushed 3n 6 1.21
Mulberry Creek Cord Marked 4b 16 3.23
Plain Shell 4a 1 .20
Swift Creek Complicated
Stamped, Early 1 .20
Swift Creek Complicated
Stamped, Middle _1 .20
Total 496 100.00
Table 2. Money's Bend Site Pottery from Feature 1
small broken triangular projectile point. These sherds give a
definite clue as to the date of the pit for they were found with-
in the humic portion of the fill, this portion is most probably
made up of material stored as food within the pit. These lime-
stone-tempered sherds definitely place the pit as Middle Woodland
in time. This places the pit as being contemporary with Feature
1, 2 and 3.
Feature 5 was a midden pit some seven feet in diameter which
reached a maximum depth of 2.9 feet into the subsoil,it contained
five sherds, two sand-temnered type 21 Benson Simple Stamped and
three limestone-tempered type 3a Mkulberry Creek Plain.
B U IALS
There were three burials excavated at the Money's Bend Site.
Burial 1 was located in the central portion of the site. It was
contained in a midden nit which measured 4.9 feet in length and
3.8 feet in width with interiev depth of 1.9 feet. The pit con-
tained the characteristic midden materials of bone, shell and the
Type 2aa Henry Island Plain sherds 4
Type arerry reeck TIan slier s 14
Chipped disk scraper 1
Small straight based projectile point 1
The skeleton was identified as a mature female, it was
placed on its back in a partially flexed position with the skull
oriented toward the south.
Burials 2 and 3 were located approximately five feet south-
west of Burial 1. They were contained in a pit measuring 3.9
feet in length and 3 feet in width which reached a depth of 4.8
feet below the surface. Burial 2 was discovered at 2.1 feet below
the surface and Burial 3 at one half foot lower. Both of the skele-
tons were complete, however the condition of the bone was so poor
that neither ale nor sex of either could be ascertained. They
were both placed in tie pit on their right sides in a fully flexed
position with their heads oriented towards the southwest. Burial
2 had been disturbed by the interment of Burial 3.
Artifacts contained in the pit containing these two burials
consisted of one sand-tempered type 2aa Henry Island Plain sherd
and three limestone-tempered sherds of the type 3a Mulberry Creek
Plain. One concave based opposite beveled "spinner" point consti-
tuted the total lithic material from the pit.
It appears that these burials were placed in the midden pits
by the occupants of the village. There were no artifacts found in
association with the burials. Burials are illustrated in Figure 3.
The analysis follows that of Heimlich (1952) for the Gunters-
ville Basin. The whole range of northern Alabama temper types
were encountered at this site with the exception of the earliest
temper type, fiber temper. Table I is a tabulation of the total
sherds recovered from the Money's Bend Site, Table II is a tabula-
tion of those sherds found in Feature 1.
The chronological position of this site based on the analy-
sis of the pottery is clearly Middle Woodland (DeJarnette 1952:
277). The scarcity of other temper types other than limestone is
best explained by considering this site as being occupied only
after sand-tempering had replaced fiber-tempering and then occupi-
ed for any length of time only by the manufacturers of limestone-
tempered wares. The appearance of temper types that appear later
in this area can be explained by trade and temporary encampments.
In addition to the pottery presented in the tables three other ex-
amples of ceramics were found, these being sand-tempered pipe stem
fragments of Indian manufacture. Sherds and pipe stem fragments
are illustrated in Figure 4.
The following is a list of the lithic artifacts recovered
from the site with the exception of projectile points and blade
fragments and are illustrated in Figure 5.
Ground stone discoidal 1
Ground stone ball 1
Pecking stone 1
Boatstone fragment 1
Turtleback blank 1
Spoke shaver 1
Projectile points were classified after Foster (1952), only
13 of the 73 points recovered from the Money's Bend Site were com-
plete enough for classification. These 13 are relegated into the
following categories and are illustrated in Figure 6.
Projectile Point Types
Pine Tree Point 1
Temporally the Money's Bend Site is Middle Woodland. It shows
undoubtable relationships to other Middle Woodland sites in the
Guntersville Basin and other northern Alabama sites (Webb and
Wilder: 1951, Webb and DeJarnette: 1948a, 1948b).
Evidence found at the Money's Bend Site points to more than
a seasonal camp for Features 1, 2 and 3 certainly appear to be a
palisade system, a defensive structure that required no few man-
hours for construction. This certainly implies a social struc-
ture and organization above the archaic hunting and gathering
level. Life at the Money's Bend Site would appear to be at least
for a time one of sedentary existence.
The majority of food remains found on this site was composed
of shellfish remains, but not to the extent found at earlier
sites. The amount of these remains was not impressive in the
light of the complexity of the village. The impression received
from excavation was that the periwinkle shell outnumbered the bi-
valve specimens. These small shellfish were whole when excavated
and were apparently prepared for consumption by boiling, veget-
able or animal supplements were probably added to thicken the
broth before it was eaten.
The evidence that hunting played no small part in the subsis-
tence economy is found in the amount of bone, much of which was
identified as deer, which was excavated on the site, however even
by addition of this source of food to the shellfish varieties pre-
viously mentioned they do not form an impressive quantity of food
for such a village. Some other food must have added a consider-
able amount to the diet of these people, undoubtably this was
The evidence of vegetable food is only hinted at by Feature
4, and admittedly this storage pit could have contained smoked or
dried meat or fish. It may be pointed out that no matter what
was stored in this pit, whether vegetable or animal food, that
other pits would have been found had excavation been complete.
Assuming that Feature 4 was used for the storage of veget-
able foods as most storage pits seem to have been then it can be
seen that this was the food that was important, for it was stor-
Settled villages are the result of man's control over food
production, not only in theory but in practice with few except -
ions in which they are the result of rich environments such as
the American Northwest Coast.
At this time period Caldwell (1958) places the inhabitants
of this general area in a rich ecology. They were, according to
his interpretation,dependent upon acorns, hickory nuts and other
forest seeds for their subsistence and possessed special techni-
ques for processing these foods. iJilley and Phillips (1958) how-
ever, do not place the peoples of this time as being dependent up-
on a forest economy but see the Middle 17oodland period as belong-
ing to the Formative Stage of their developmental sequence, for
even though they admit that this area is marginal to their con-
cept of the Formative Strge they grant this area rude agriculture.
Caldwell also suggests that the lack of rich mortuary prac-
tices that would be associated with such a level of culture may
be explained as suggested by DeJarnette (1952:278-279) that the
limestone-tempered pottery makers were also the Copena peoples,
this however awaits further proof.
In the final analysis this site was evidently a fortified
village and its people enjoyed a sedentary existence, an exist-
ence that would d-pend on exploiting the natural resources pre-
sent as suggested by Caldwell or on the other hand a. form of low
level agriculture as postualted by Willey and Phillips. Thus the
Money's Bend Site is more or less a typical example of Middle
WVoodland pattern for this area.
Caldwell, Jj R.
1958 Trend and Trpdition in the Prehistory of the Eastern
United States. Memoirs of the American Anthropological
Association, No. 88. Menasha.
DeJarnette, D. L.
1952 Alabama Archeology: A Sun'ary. In Archaeology of the
Eastern United States edited by James B. Griffin, Uni-
versity of Chicago Press, Chicago.
Heimlich, M. C.
1952 Guntersville Basin Pottery. Alabama Museum of Natural
History, Museum Paper 23. The Museum, University,
Webb, W. S. and DeJarnette, D. L.
1948a, The Flint River Site, Yao48.
Natural History, Museum Paper 23,
1948b The Whitesburg Bridge Site, MaPlO.
Natural History, Museum Paper 24.
ilebb, W. S. and l:ilder, C. G.
Alabama Museum of
The Museum, Univer-
Alabama Museum of
The Museum, Univer-
1951 An Archaeological Survey of the Guntersville Basin on
the Tennessee Viver in Northern Alabama. University
of Kentucky Press, Lexington.
Willey, G. R. and Phillips, P.
1958 Method and Theory in American Archaeology. University
of Chicago Press, Chicago.
Florida State University
L2a 2ee L5L L, 14 L4 L&I L o
S Q. /0
S i THE MONEYS BEND SITE, CCV3
SCheroke. Cu..1t, Alebee.
FIGURE 2.- FEATURES I, 2, AND 3
FIGURE 3.- BURIALS
Top- Burial I
Bottom- Burials 2 and 3
FIGURE 4.- SHERDS AND PIPE FRAGMENTS
a. Steatite; b. O'Neal Plain; c. Prospect Ped Filmed; d.
Rudder Comb Incised; e. Flint River Cord Marked; f. Henry
Island Plain; g. Swift Creek Complicated Stamped, Middle;
h. Benson Simple Stamped; i. Mulberry Creek Cord Harked;
j. Swift Creek Complicated Stamped, Early; k. Flint River
Brushed; 1. Mulberry Creek Plain; m. Pipe Stem Fragments
h j k I
FIGURE 5.- LITHIC MATERIAL
a. & b. Hammerstones; c. Whetstone; d. & e. Gamming Stones;
f. Bannerttone Fragment; g. Turtleback Blank; h. Side
Scraper; i. Disk Scraper; j. Spokeshaver Scraver; k. Drill;
FIGURE 6.- PROJECTILE POINTS
The Florida Anthropologist, Vol. XIII, No. i, 1960
SHFLI VOIUND, LEVY COUNTY, FLORIDA
Ripley P. Bullen and Edwarrd Dolan
The shell midden is located near Cedar Key on the Florida Gulf Coast in Levy
County. It occupies the southwestern end of an old series of dunes. A ten foot
square test was excavated in 6-inch arbitrary levels, consolidated for purposes
of presentation into 12-inch levels. The major pottery type was Pasco Plain, with
minor amounts of Dunn's Creek Red, St. Johns Red-on Buff, Sand Temere Plain, and
Deptford Check Stamped. Busycon and Melongena shell hammers were the major tool
types. St. Johns Plain is limited to the top four feet of the midden where more
hunting and less fishing or collecting were practiced. Abstract by C. H. F.
Shell Mound is located on the shore in the western part of
Levy County, Florida, about five miles north of Cedar Key. The
site, which consists of a large shell midden, is situated at the
southwestern end of a long finger of old, connected, sand dunes
which extend some two miles northeasterly from the site but which
do not exhibit any evidence of aboriginal occupation except in the
vicinity of the midden.
This long narrow series of dunes averages about ten feet in
elevation except at the midden where the highest point is about 20
feet above mean sea level. Apparently, Indians dumped shells at
the southwestern end of this series of dunes, extending the high
land further. towards the Gulf. The highest part of the midden is
about 28 feet above mean sea level.
To the west of the site are shallow, oyster producing waters.
To the north and south are salt marshes. Dennis Creek borders the
dunes to the south and east. Across the marshes and across the
neighboring waters various marshy islands may be seen. Some of
them are partially wooded and some were occupied in Indian times
The whole region is one of extremely low relief. To the west
are the shallow waters of the Gulf, to the east extensive wooded
swamns. This is a region which can produce vast quantities of
fish, shellfish, and shore birds but one which is not suitable for
agriculture under present conditions.
During the summer of 1959 construction of a road to Shell
Mound permitted ready access. The junior author, while employed
by the Florida State Museum, made a stratigraphic test in the high-
est part of the midden. This report, written by the senior author,
presents the results of his excavation. We appreciate the permis-
sion, cheerfully given us by Mr. Dennis E. Andrews of Chiefland,
Florida, which made this work possible.
A 10-foot square was excavated by arbitrary 6-inch levels.
The results were analyzed by these levels but are presented here
(Fig. 3) by 12-inch levels to save space. It was necessary to per-
mit the walls of the test pit to slope a little inward as a safety
measure. By a depth of 7 feet, the size of the pit was thus re-
duced to 7 feet on a side. At this depth, the test was narrowed
to 2 feet on a side. These changes in the excavated area are in-
dicated on the profiles presented as Figure 1. They should be
taken into account in any consideration of the quantities given in
Figure 3 for levels below a depth of 7 feet. Fortunately, the low-
est zones penetrated by our test were relatively rich.
Specimens from Shell Mound need little description. Sherds
are listed by conventional pottery types in Figure 3. Only two
merit further comment. One is a Dunns Creek Red sherd which was
painted on the inside and not, apparently, on the outside.
The other is a St. Johns Red-on-Buff sherd (Fig. 2) made of
a very hard St. Johns paste. All surfaces are extremely well
smoothed. Walls of this vessel 1/4 of an inch thick and curve in-
ward just below the lip which extends outward a little. The lip
is well flattened and 3/8 of an inch wide. A band of red paint,
5/8 of an inch in width, parallels the rim and is separated by an
inch from the outer edge of the lip. This band is painted over
the buff background which may be seen both above and below. Un-
fortunately, the sherd in too small to determine the design form-
ed by the red-painted area.
While differing in naste, color, and technique of decoration
this sherd came from a Vessel which must have been very similar in
shape and design of decoration to those called Crystal River Nega-
tive Painted. It is hard to believe the veqqel represented by our
sherd from Shell Mound could be separated by any great period of
time from the stylistically similar Crystal River Negative Painted
IN HUMIC ZONE.
LARGE. OYSTER SHELLS
WITH .EPTH, SHELLS
SLIGHTLY .SMALLER AND MORE
SHELLS, BLACK SOIL
WITH DEPTH, SHELLS SLIGHTLY
SMALLER AND MORE BLACK SOIL
SHELLS,3 LACI SOIL -
0 1 z 3 4
Fig. 1 West and north profiles, Shell Mound, Levy County.
Chert chips, a small core, and two utilized flakes are listed
in Figure 3. Made of the same gray chert as the basal portion of
a projectile point having a slightly notched, expanding stem (Fig.
2). It came from a depth of 6 to 6 1/2 feet.
Shell specimens include the Busycon and Melongena hammers
listed in Figure 3, an Oliva shell bead (Fig. 2), and several frag-
ments of unclassifiable Busycon tools. Hafting holes of the three
illustrated Melongena hammers may be seen in figure 2. The shell
bead was found between bpths of 1 1/2 and 2 feet.
Deposits showed the usual minor variations with depth which
are to be expected in coastal shell middens (Fig. 1). Near the
top oyster shells, mixed with a few Melongena corona shells, repre-
sented 90 percent of the deposit. At a depth of 18 inches some
scallop shells were found and at 42 inches an increase in clam
shells was noted. Clams seemed to be more common around a depth
of 7 feet than elsewhere. Otherwise, the relative quantities of
different kinds of shells did not vary greatly. Oysters were the
most common shell from top to bottom of our test while clams were
more plentiful near the bottom than the top. Some Melongena coro-
na shells were always present as were, more rarely, other shells
of the region.
Between depths of 50 and 60 inches a noticeable change occur-
red in the deposits. Shells became pulverized and intermixed with
black dirt. At 58 inches was a lens of mortar-like, clayey,burn-
ed shell 18 inches in diameter. Other areas of burned shells were
noted at depths of 60 and 62 inches. These deposits seemed to re-
present an occupation zone. Below 60 inches the percentage of
sand decreased but was still considerably greater than in the up-
per part of the test.
This change in the physical stratigraphy is particularly in-
teresting because a cultural division occurs at about the same
time. Examination of the vertical distribution table for sherds,
tools, and food bones (Fig. 3) shows a definite shift at a depth
of about 4 feet. While Pasco Plain sherds are found throughout
the test, St. Johns Plain sherds are limited to the highest 4 feet
z S Foo0l
5 BSON tES
0- 0 ^ Cj a U < 1 -
Sz i T I -j
R R A
0-i 41. 2 11 53 27 1 1
1- 18 3 1 96 14 1 1
S-35 5 4 1 1 15 Z 250 47 2 3
3-4 21 2 3 3. 110 14 z
4-5 89 1 : Zb 5 7 1 65 C 4 9 9 1
5- 44 3 4 9 195 31 15 4
6-7 323 1 4 3 141 21 3 4
7-8 43 1 1 Zf 43 7 3 5
8-10 9 1 2o I 2.
ToTALS S552 11 1 8 1 4 41 Z l3o73 ZZ8 37 5 2 5
Fig. 3. Vertical distribution of sherds, shell hammers, and food bones
at Shell Iound, Levy County.
aIncludes 1 burnished, 1 cord-marked fIncludes 1 utilized flake
bIncludes 1 Smooth Plain gIncludes 1 shark vertebra
CIncludes 2 Belle Glade Plain-like hIncludes 1 shark tooth
dIncludes 1 scored 1Includes 1 dog, 1 rabbit bone
eIncludes 1 core, 1 utilized flake JIncludes 1 raccoon bone
while sand-tempered sherds concentrate below a depth of 4 feet.
Similarly, chert chips (and a projectile point) concentrate in
lower levels as do deer and bird bones while crab claws appear in
These data are suggestive of a shift in the economy of the in-
habitants of Shell Point away from an em-hasis on hunting (chert
chips, projectile point, deer bones) to a greater dependence on
fishing and shellfish collecting. Could a rise in sea level have
resulted in an increased oyster production relative to clams while,
at the same time, land to the east of the site became more wet ten-
ding to drive deer further from the site so that their capture be-
came more difficult?
For comparison with the excavated materials from Shell Point
we have two surface collections. One, made by the junior author
during excavation, came chiefly from the shore edge of the midden.
It includes 3 Pasco Plain, 3 sand-tempered plain, 1 St. Johns
Plain, 3 St. Johns Check Stamped, 2 Wakulla Check Stamped, 5 Deut-
ford Check Stamped, 1 Jefferson tyne rim, and 3 Chattahoochee
Brushed-like sherds. It will be noted that nearly 50 per cent of
the sherds in this collection are check-stamped while 5 of them
(Wakulla. and St. Johns Check Stamped) are relatively late check-
The other collection,made some time prior to excavation and
taken chiefly from the sandy field immediately east of the shell
mound proper, includes 5 Pasco Plain, 44 sand-tempered plain, 5
St. Johns Plain, 9 St. Johns Check Stamped, 3 Wakulla Check Stamp
ed, 1 smoothed-over check-stamped, 1 cord-marked, and 4 miscel-
laneous punctated sherds. Again, it will be noted that a relative-
ly large number (18 percent) are check-stamned and that relatively
late check-stamped types are represented.
Large quantities of St. Johns and 7akullla Check Stamped
sherds in these surface collections indicate people were living at
Shell Mound during the late 'feeden Island II period. Such zones,
however, were not penetrated by our teat which starts during a
Deptford period and continues upward in time through the Weeden
Island I period (Fig. 3). lVe would correlate zones which produc-
ed the St. Johns Red on Buff, the Dunns Creek Red, and the St.
Johns Plain sherds with this Weeden Island I time period. Whether
or not zones between the Deptford and St. Johns sherds, i. e. be-
tween depths of 5 and 8 feet, represent a Deptford through Santa
Rosa-Swift Creek continuum or a Perico Island-like period of un-
decorated pottery may be argued.
Indians lived at Shell Mound from the Deptford through the
Weeden Island II periods. This represents a period of time great-
er than a thousand years. Probably the site was the home of a
very small group who, from generation to generation, tended their
oyster bars and made occasional forays further afield to supple-
ment their diet of fish, shellfish, and turtle. During this time
there seems to have been some ecological changes in the region
which were reflected in the debris left by these inhabitants.
After a period of abandonment, the new road assures Shell
Mound new inhabitants those who come by motor car and go by
motor boat instead of by a dougout propelled by paddles or poles.
Another ecological change has begun.
Florida State Museum
University of North Carolina
Chapel Hill, North Carolina
Fi3s 2. 011va shell bead, St. Johns Red on Bluff sherd,
projectile point, and three elonga hammers from Shell
Mound, Levy County.
The Florida Anthropologist, Vol. XIII, No. 1, 1960
THE HARRIS CREEK SITE, TICK ISLAND, VOLUSIA COUNTY
Francis F. Bushnell
Bushnell reports his salvage of the large midden during commercial shell re-
moval. Sherd counts of Orange, Tick Island, St. Johns, and Swift Creek Complexes
are given for various areas of the midden. A burial "trench" containing 15 20
burials was destroyed before adequate observations could be made. Various areas
show chronological differences in sherd counts. Abstract by C. H. F.
The Harris Creek Midden is located on the eastern side of Tick
Island, some two hundred yards up the right bank of the creek from
Lake Woodruff, and approximately six miles from the town of DeLeon
Springs, Volusia County, Florida.
Several professional groups have visited and tested this mid-
den, but due to its large size, much remained intact until recent
commercial excavations for the shell began. It was the author's de-
sire to salvage as much material as possible from the surface and
from the excavated areas in my six visits.
Many thanks go to Mr. Wester Branton, of DeLeon, for permission
to work over the area, and to those who offered much needed infor-
mation concerning the recent history of the total area.
The map (Fig. 1) represents the immediate geographic location
of the midden with other features.
In rambling over the site, the writer mentally conceived sepa-
rate areas, each area of the midden proper receiving equal time and
effort. The areas are indicated in Fig. I by capital letters.
The findings by the given areas are as follows:
Area A This area was the most productive portion of the midden.
A sherd count of the whole area would have been too time consuming
because of the tremendous amount of material present. With great
care a grid system was used, and when the results of six grids were
found to be approximately the same, the count was stopped. Each
grid measured 6 feet by 6 feet.
The concentration of sherds appeared to be the same over the
Fi. I IMlADIaTE GEUORaPHIC Rhhn OF 'iHLE 1-1hb CR1.rK bilTE
Ft Z SHilID CUUl.T BY GRID, aiAk "A"
2 2 4
71 4 7 12 4 5 4 il
S 2 4 0 2 0 3 L4
S 8 10 6 13 16 15 6 4
_________ 1 ___________1
- 26 -
L'riLL b'U iAT
entire "A" area, and the majority of all other artifacts were also
found here, with the exception of projectile points.
Sherds of all of the St. Johns period occurred to approximate-
ly the three-foot line, with many fine examples of Orange Incised
showing up at the three, to three-and-one-half foot level. In many
cases this was below the present water-line, the shell being remov-
ed to the base by dragline.
Fig. II represents the results of the grids used.
Bone, antler, flint artifacts and ornaments were numerous in
this area, although these items were usually found in a fragmentary
condition. Only one example of a polished greenstone polled celt
was seen. This celt, measuring 10" in length, was found by a local
fisherman's son, who would not Dart with it.
Area B. This area showed numerous Orange Incised sherds below the
one-foot mark. This whole area had been surface stripped and clear-
ed by bulldozers. Orange Plain ran into the Incised at varying
shallow depths. Very little material was found in this area other
than the pot sherds. Some chalky ware occurred directly on the sur-
face at rare intervals. This area had been very badly mixed.
Area C. This area was also relatively barren, with the exception of
three small shell mounds, each containing one burial. Only one of
these burials contained artifacts: one small scalloped shell gor-
get, and one small triangular flint point. All burials were decay-
ed beyond salvage, or recognition as to position in burial.
Area D. It is interesting to note that in this area there is a
striking "hill" or rise which might at first be mistaken for a
temple mound structure. In this rise, sherds of the St. Johns
period are almost absent, being found only on the direct surface .
Orange Plain appears at a very shallow depth, with a possible pre-
ceramic region showing up at about two feet.
When the northern edge of this rise was removed, a white sand
strip appeared at approximately five feet that was surrounded by
dense shells. This strip contained fifteen to twenty burials, all
badly decayed, with many showihg indications of violent death; the
chest cavity of one burial containing sev-ral projectile points.
In several other instances, areas of the cranium were crushed in-
ward. This isolated burial trench was destroyed by the commercial
operations before a thorough study could be made.
Most of the projectile points discovered came from this area
and were found on the direct surface. These were all of exception-
ly fine workmanship. Flint chips and turtle-backs were present in
Area E. From the rear of the Area D, a shell trail traverses the
swamp and extends across the island. This trail terminates on Tick
Island Run, and at one time supported a tram track that was used to
haul shell to barges. Information gathered from many old-timers in-
dicated that this trail and others on the island were present be-
fore the commercial operations started.
Half-way between the midden and Tick Island Run another trail
branches to the right, passes through an old homestead area, a low
hammock, and finally reaches the Tick Island Burial Mound (Area G
on the map). This same trail then extends from the mound to a long
excavated midden site (Area H).
While conversing with several local commercial fishermen, it
was learned that a sand mound lies in the saw-grass area south of
Harris Creek. This mound is reputed to be in an untouched condit-
ion, but is not accessible during most of the year, either by boat
or on foot due to high brush and swampy conditions. It has not
been seen by the author.
At the same time, the location of a very large midden on Kim-
ball Island, Alexander Spring Creek, Lake County was confirmed. It
is interesting that even in this isolated location, orange trees
are found in some number. These trees have been present for a long
period of time, manv of the old time residents of Astor having jour-
nied into the area in their childhood to procure strong sour-orange
stock. It could be that these trees represent a carry-over of
growth from either seeds distributed during the Spanish occupation,
or an old Spanish Grant grove-homesite.
Summary and Conclqsion:
The midden at Harris Creek, while exceptionally rich in mater-
ial, was much too large to work on small scale with any manner of
efficiency. While surface collecting cannot solve definite prob-
lems, it certainly results in a rather good picture of the site,
its uses, and its areas of industry.
Two trails of interest were noted and surveyed along with the
well-known Tick Island Burial Mound. Two new sites were discovered.
In the Midden, Area A was obviously more recently occupied
with strength while the "black" areas were of great age. Area A
probably extended through St. Johns II, possibly into St. Johns
IIb or later. Artifacts were numerous and of fine quality. The
age and use of Area I is, at the present, questionable and open to
One other small site was covered, as far as surface collecting
goes. This area, known as "Hardscrabble" is located on the north-
ern shores of Tick Island on Lake Dexter. One sherd of Tick Island
Incised was found along with numerous flint chips and more recent
pottery types. The finding of the Tick Island Incised is of inte-
rest, as this type is more commonly found locally around the Blue
Creek area of Lake George, and is seemingly quite rare on Tick Is-
land itself. Most of "Hardscrabble" was removed long ago for use
as road building material.
No trade items of European origin were found on or near the
surface of the sites on Tick Island during the author's visits.
Such items may occur however, for European material had been found
in the majority of the sites surrounding this area, particularly
the burial mound at DeLeon Springs.
The most thorough work on this mound north of DeLeon was done
by a youth who subsequently was killed at the site in a cave-in.
All material recovered from this site has long been lost.
A good portion of the Harris Creek site still exists, but is
rapidly being removed; the material being hauled by barge to DeLeon
and sold for driveway and septic tank fill.
St. Petersburg, Fla.
- 29 -
A, Base of St. Johns Plain Vessel. B, Side and handle of chalky
ware cup. C, Swift Creek Comp. Stamped (Chalked in desing). D,
Orange Incised. E, F, St. Johns Incised. G, St. Johns Check
Stamped Modified. H, I, Orange Incised. J, Base of a textile
impressed vessel, St. Johns Plain. K, Orange Plain. L,St.Johns
Plain cup. M, Plain chalky ware.
A, Bone Dagger (note faint incisions on handle). B. Strombus
celt. C, Sandstone projectile base. D, Chipped flint celt.
Remainder Projectile points of flint and bone from area D.
A, Effigy clay "Blunt". B, Serpentine plummet (drilled and groov-
ed) C, Sherd bar-gorget. D, Scalloped shell gorget. E, Cup handle.
F, Spindle shaped clay disks. Remainder bone pins of varying type
and decoration, plummets, bone tubes and fired clay lump.
A, Antler dagger handle. B, Sawed deer bone. C, Sandstone bone.
D, Polished deer ulna. E, Conch dipper. F, Polished greenstone
adze edge. G, Grindstone. H, Antler flakers (note one on far
right bears visible tool edge marks at the base cut) I, Conch
scraper. J, Conch cup.
Spindler, George and Louise, eds. Case studies in Cultural Anthro-
pology. Holt-Dryden Books, Henry Holt & Co., N. Y.$1.25 each
A new series of inexpensive, paper-bound ethnographic des-
criptions of various cultures. The series so far consists of
the Cheyennes, Bunyoro (Africa), Twi (Australia), Tepoztlan
(Mexico), Palau (Pacific). These are not reprints but newly
prepared and edited descriptions of the cultures. They do
not follow a formula but follow the author's interests as
well as the people being described. Each is close to 100
pages and gives a fresh, readable, sound description of a
people and their culture.
Webb, Clarence H. The Belcher Mound A Stratified Caddoan Site in
Caddo Parish, Louisiana. Society for American Archaeology,
Memoir No. 16, XIV, 213 pp., 142 figs., 4 tables. $3.00,
Salt Lake City, 1959.
A meticulous description and analysis of the excavation of
a large temple mound over a period of nearly twenty years.
The phases present are Haley, Alto, Bossier, and Belcher foci,
all representing a late and elaborate manifestation. The cul-
ture may have existed from about A. D. 858 to just about the
close of the prehistoric period.
The report is a beautiful example of the achievement of a
dedicated non-professional archeologist.
Haag, William G. The Archeology of Coastal North Carolina, Louisi-
ana State University Studies, Coastal Studies Series No. 2,
xii + 136 pp., 17 figs., 5 tables. Louisiana State Univ.
Press, Baton Rouge, 1958.
Haag reports the site survey he has conducted along the
Cape Hatteras shores. Very small surface collections and
test excavations are used with great skill in producing a cul-
tura.l framework for the area. For North Carolina, the report
fills in a gap. For most of us its importance lies in showing
just how much of real value can be gleaned from a pitiful
handful of sherds.
- 32 -
Ols0n, Stanley J. Foqail !piamnlo of Florida. Floridn Geological
Survey, Special Publication No. 6, iv + 74 pp., 14 plates, 13
figs., Tallahassee, 1959.
This is a replacement and amplification of G. G. Simpson's
1928 "The Extinct Land Mammals of Florida." The numerous il-
lustrations by Andrew Janson verv adequately supplement the
text. The bulk of the text and illustrations cover the Ceno-
ajoic Era, the Paleocene to recent times. It will prove as
popular as the earlier book.
Below are listed "paper back" books dealing with archeology
or the American Indian. All are significant for all anthro-
pologists, professional or non-professional.
Firth, Raymond. Human Types. New American Library. $.50
Lintcn, Ralph. The Tree of Culture. Vintage Books. $1.25
White, Leslie A. The Science of Culture. Evergreen Books.
Collier, John. Indians of the Americas. New American Lib-
Crane, Verner W. The Southern Frontier. Ann Arbor. Pap-
Clark, J. Desmond. Prehistory of Southern Africa. Pen-
Oakley, Kenneth P. Man The Tool Maker. Phoenix. $1.25
Clark, Ella E. Indian Legends of the Pacific Northwest.
[Jniversitv of California Press. $1.95
Kluckhohn, Clyde. Mirror for Man. Premier Books. $.50
Colden, Caldwallader. The History of the Five Indian
Nations. Cornell University Press. $1.75
Benedict, Ruth. Patterns of Culture. New American Lib-
Tylor, Edward B. The Origins of Culture. Torch. $1.75
Radin, Paul. Primitive Religion. Dover. $1.85
Mason, J. Alden. The Ancient Civilizations of Peru. Pen-
THE FLOPIDA ANTHROPOLOGIST
VOL. XIII MARCH, 1960 NO. 1
The Money's Bend Site, CeV3
Keel, Bennie C. ................................. 1
Shell Mound, Levy County, Florida
Bullen, Ripley P. and Edward M. Dolan............17
The Harris Creek Site, Tick Island, Volusia County
Bushnell, Francis F..............................25
C. H. F. ................... 33 & Inside back cover
Members must advise the editor of change of ad-
dress. Current postal regulations make it prohibitive for
the Society to provide forwarding service. If you do not
advise us of your new address, you will not receive The