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.
PUBUSHED BY THE
FLORIDA ANTHROPOLOGICAL SOCIETY
MUMMTElUIS SJIRA I ACI C CMJTRM
--- L --
THE FLORIDA ANTHROPOLOGIST
Vol. IV May, 1951 Nos. 1-2
Beaded Shoulder Pouches of The Florida Seminole
.... ...................... John M. Goggin 3
Experiments with Raw Materials Utilized by The Florida Indians
in Ceramic Construction
. . ..... Hale G. Smith and William Watson 18
The Gard Site, Homosassa Springs, Florida
................ .. .... . Ripley P. Bullen 27
Review Excavations at Kolomoki, Season I 1948 by William
.............................. Ripley P. Bullen 32
Contributors to This Issue
......... .......................... 35
published at the University of Florida
Fig. 1. Seminole Pouch and Sash, specimen no. 1.
BEADED SHOULDER POUCHES OF THE FLORIDA SEMINOLE
John M. Goggin
Although the Florida Seminole at the present time have developed a
unique style of clothing, they once shared many of the basic garments and
accourtrements of the Southeastern Indians. One particular item which will
be briefly considered is the shoulder pouch suspended from a broad sash.1
Shoulder pouches are undoubtedly of great antiquity among the Indians
but this particular style appears to have been adapted from the Europeans.
In one form or another it is sporadically widespread among the peoples of the
Eastern Woodland and is generally considered to have been derived from
pouches of 18th century English soldiers (Wissler, 1946: 20).
At the present time these pouches are no longer used and none are known
to exist among the Florida Seminole. It is possible, though, that one or two
may still remain and are brought out by the old men in occasional ceremonies
where most old articles of dress appear. Our study will be based on the few
specimens still surviving in museums and those illustrated in various old
drawings and photographs. All documented specimens date from the 19th
century, and two in particular are over 100 years old, each with an interest-
This study is part of a broader one covering art and material culture
of the Southeastern Indians. Preliminary work has resulted in a two-fold
eastern and western division of the region on the basis of certain art styles
(Goggin, n.d.) but more intensive study is expected to produce finer dis-
tinctions. It should be kept in mind that the group now called the Florida
Seminole is of very diverse origin, including a number of "Creek" units as
well as remnants of older Floridian peoples such as the Yamassee and Calusa.
At the present time a dual linguistic and political division is recognized in
Florida, the Cow Creek Indians to the north who speak Muskogee and the
Big Cypress Indians in the south who speak Mikasuki (these last have
1This paper is a contribution from the research program of the Department of
Sociology and Anthropology, University of Florida, aided by a grant from the
Viking Fund, Inc. The wrter is grateful to the authorities of various museums
where materials described here have been studied, especially, E. K. Burnett at
the Museum of the American Indian, Heye Foundation; Herbert Krieger at the
United States National Museum; and Alex Spoehr at the Chicago Museum of Na-
tural History. Photographs of specimens used in this paper were furnished by
DESCRIPTION OF SPECIMENS
The first of these (United States National Museum, cat. no. 380668),
which we will call specimen 1 (others will be numbered in order) is made of
red flannel or strouding backed with calico (Fig. 1). In form, the pouch is a
flat envelope, 8 inches square, with a triangular flap. Pendant from the
bottom are 10 short strings of large, faceted, blue beads, each terminating in
a cloth strip tassel.
The pouch is sewn to a shoulder strap, 4 inches wide, of similar mate-
.rial. Below the point where it is attached to the pouch the strap divides in-
to four tapering fringes, each with a cloth strip tassel.
Both the pouch and shoulder strap are decorated with an embroidered
design in blue, white, and green beads. The beading is very well done and
Fig. 2. Seminole Pouch, specimen no. 2.
is still quite tight. It is done in a two thread technique, one thread carrying
the beads, the other stitching over the first to hold it down.
The same motifs are repeated the length of the strap and on the pouch.
They are a simple diamond with recurved ends.
The specimen was collected by Anthony Breath at Charlotte Harbor,
Florida in 1849. It was made by the wife of Billy Bowlegs,2 a member of
Fig. 3. Seminole Pouch Sash, specimen no. 2; A, entire sash; B, detail of mid-
section showing two designs.
2Together with the pouch is a letter written by the collector, Anthony Breath, to
his brother Jim in 1849. Because of its historic interest this is reproduced in an
the original Alachua group of Seminoles.
Specimen 2 (Museum of the American Indian, cat. no. 8/4299), is another
old pouch and like the former is made of red flannel with a calico lining
(Fig. 2). It is similar in shape, 8% inches wide and 7Y inches deep, with
a triangular flap which extends the whole front of the bag. Blue-green silk
ribbon binds the edge of the flap, and several dark blue and dark red wool
yarn tassels are pendant from the bottom of the pouch.
The carrying sash (cat. no. 8/4300), made of similar material, is 4'
inches wide and 5 feet, 1 inch long. The ends are divided into two taper-
ing fringes. The edges of the sash are bound with white silk ribbon. A
peculiar feature of this specimen is that it is formed of two separate pieces
of material joined together at the midpoint (Fig. 3). Each piece has a differ-
ent design formed in white,
yellow, and blue beads
Both pouch and sash
are embroidered with very
'small seed beads, in a two
thread technique. The
design on the pouch is an
elaborate geometric motif
S? (Fig. 2), while that on the
sash is simpler. On one-
/ half is a series of large
diamonds on a line of smal-
ler ones, while the other
half bears large X motifs
on a line of smalldiamonds-
According to catalog
data, the bag and pouch
Were collected by Abner
LFuller Lyon of Needham,
% Massachusetts from Chief
SCloud during the Seminole
Si i War in Florida.3
T Specimen 3 (Chicago
Museum of Natural History,
Fig. 4. Seminole Pouch and Sash, specimen No. 3. cat. no. 167917) is a pouch
of the same general shape
3There were two chiefs with this name, Little Cloud (Chatkwa Owluche) and Cloud
(Ye-how-lo-gee). Presumably this came from the latter. He was painted by Catlin
(1844, vol. 2, pl. 299) while imprisoned at Fort Moultrie, South Carolina.
with a triangular flap. Eight wool yar tassels line the bottom. It is made
of red cloth, the edge bound with ribbon (Fig. 4).
The shoulder sash is black cloth lined with tan colored cotton and the
edges are bound with ribbon. The ends of the sash are divided into five
sections, each terminating in a wool tassel.
Seed bead embroidery designs cover both the pouch and sash. On the
former it is composed of black, white, blue, and rose beads; while white,
light and dark blue, rose, and yellow beads form the design on the sash. The
technique is two thread.
The decorative motifs on the pouch are merely an overall pattern of
diamonds on both the bag and flap. The sash decoration consists of six
bold double curve motifs.
Neither the original owner nor maker are known but the specimen was
collected in Florida in 1894 by Charles B. Cory. Inasmuch as Dr. Cory had
most contacts with the Cow Creek group it is most probable that the bag was
made by them (Cory, 1896).
Two further examples can be mentioned, both in the Museum of the
American Indian. Specimen 4 (cat. no. 13/5086) is a fine beaded pouch and
bag of blue flannel, the sash 4 inches wide and the pouch 7 3/8 by 7 1/2
inches. The sash is decorated with white, light blue, and red beads forming
two distinct designs on each half of the sash. One is a series of elaborated
X motifs, the other a flower or star motif. The bag itself has eight stars on
the flap and a chevron-like pattern on the front.
Unfortunately, this is poorly documented. It is said to be Seminole and
to have been a "purchase." In many respects it is similar to documented
Specimen 5 (M.A.I. cat no. 2/5653) is only the pouch. It is made of red
flannel and measures 8 inches wide by 8 5/8 inches deep. The long tri-
angular flap and edges of the pouch are bound with ribbon. A simple four-
pointed star-like design is repeated in skimpy bead embroidery three times
on the flap. No sash is present.
This pouch was collected by M. R. Harrington from Seminoles in Okla-
homa. No other data are available.4 In form it is like other Seminole speci-
4Unless completely documented any ethnographic material collected in Oklahoma
must be evaluated with caution. Extensive contacts between the various tribes in
that state has resulted in considerable exchange of specimens.
SPECIMENS FOUND IN PICTURES
To add more complete data to our study an attempt has been made to lo-
cate all available illustrations, prints, and photographs of Seminoles wear-
ing pouches. Many pictures have been examined but only a few have yielded
usable data, either because pouches were lacking in most cases or because
the reproduction was too generalized, obscure, or limited in area of the per-
Out of the total pictures examined only eight sashes and pouches or
pouch sashes can be distinguished well enough for recognition and in few
cases can the pouch itself be seen well enough to discuss. However, the
sashes and their designs stand out.6
A very distinctive sash is specimen 6, worn for several years in the
1850's by Billy Bowlegs (the same mentioned previously).7 The design is
apparently bead embroidery in an elaborate running diamond design (Fig. 5,
A pouch sash, number 7, worn by another member of the same party in the
New Orleans visit, No Kush Adjo, is bead embroidered with a row of diamonds
with recurved ends (Fig. 5, B).8 This is very similar to the specimen 1(Fig.
1). No Kush Adjo is presumably a member of the same group as Billy Bow-
legs, the Alachua Seminoles.
Specimen Number 8 was worn by Chief Tallahassee in late 19th century
5Major sources examined include the 3 volume Indian Tribes of North America
by Thomas L. McKenney and James Hall. First published in 1836, it appeared in
several editions, each with varying numbers of plates. All portraits, though, are
reproduced in a modem edition edited by Frederick Hodge. In this paper references
are to this edition (McKenney and Hall, 1933-34). However, it is advisable to go to
earlier editions for the greater detail and clarity of the plates. George Catlin's vari-
ous books have been examined, especially his major work (Catlin, 1844; also see
6Every effort was made not to select one of the typical shoulder sashes, usually
fingerwoven wool yarn, which are worn without a pouch.
7This sash first appears in pictures made when he visited New York in 1852.
Drawings made from photographs appeared in The Illustrated London News for May
2, 1853 (reproduced in Porter, 1946) and in Gleasons Pictorial (Anon. 1852). When
in New Orleans in 1858 he was apparently wearing the same sash, judging from a
drawing in Harpers Weekly (New Orleans Correspondent, 1858; reproduced in Gif-
ford, 1925). A photographic portrait of Billy Bowlegs made on one of these occasions,
perhaps the last, still exists (Bureau of American Ethnology, Smithsonian Institution).
This photograph, may have been the inspiration for a portrait appearing in one of the
later editions of McKenney and Hall's book (McKenney and Hall, 1933-34).
8Taken from an illustration in Harpers Weekly (New Orleans Correspondent, 1858)
which is also reproduced in Gifford (1925).
Specimen 10 is worn by Coe
Hadjo in the picture painted by
George Catlin at Fort Moultrie in
1838 (Catlin, 1844, col. 2, pl. 301).
Only a sash is shown but this is
c believed to be the support for a
pouch rather than the simple decora-
Fig. 5.Designs taken from illustrated pouch tive sash. Details are hard to deter-
sashes. A, specimen no. 11; B, mine due to Catlin's drawing tech-
specimen no. 7; C, specimen no. 6;
D, specimen no. 10; E, specimen no. nique but the sash appears to be
8. bead embroidered cloth with a series
of diamonds forming the design (Fig. 5, D). Coe Hadjo was one of the St.
Johns River Seminoles.
Catlin (1844, vol. 2, pl. 302) also depicts another Seminole, The Licker
or Lashee, wearing a sash which probably held a pouch. This, specimen 11,
also is probably bead embroidered cloth with a simple design of circular and
rectangular motifs in line (Fig. 5, A).
There also can be mentioned as specimen No. 12, the pouch worn by
Osceola in a portrait painted by Catlin (1844, vol. 2, pl. 298) at Fort Moul-
trie in 1838. A plain narrow strap holds a square pouch with a very long
photographs (Wilson, 1896, illus-
trations opposite pp. 42 and 52.)
The sash differs from others in its
detailed design, apparently in bead
embroidery (Fig. 5, E). The sash
end is divided into two sections.
Tallahassee was a member of the
Cow Creek group.
Tukosee Mathla's sash and
pouch may be called no. 9. In his
full length portrait (McKenny and
Hall, 1934, vol. 3, pl. opposite p.
82) he is wearing a sash, with a
chevron design, which holds a
pouch. Details are not clear but
both sash and bag appear to be
finger woven, presumably from
wool yar and glass beads. This
dates about 1820-40. Tukosee
Mathla, known also as John Hicks,
son of the former chief of that
name, was a Mikasuki.
ar~ P~Bar ~a~s, ~
~-~.T~F~'~m ~LP~ ~D"P~
triangular flap. It does not seem to be decorated but may be made of fur.
Osceola was of Creek origin.
A final specimen, no. 13, may be noted. This is worn by John Jumper in
a drawing already mentioned (Illustrated London News, May 21, 1853; repro-
duced in Porter, 1946). Unfortunately it is too obscured for detail but is a
sash, apparently for a pouch, with a complicated bead embroidery design.
The affiliation of John Jumper is unknown. His association with Billy Bow-
legs on the New York trip may indicate a tribal association but this is not
ANALYSIS AND COMPARISON
Material and Form. The three documented specimens (nos. 1-3) are all
made of woolen cloth, lined with cotton goods and bound with ribbon. All
are red, except the sash for specimen no. 3 which is black. It is impossible
to determine the material of the illustrated specimens.9 The twoundocumen-
ted specimens (nos. 4 and 5) are similar to the first three; one is red wool,
the other blue.
There is remarkable similarity in form among all specimens. The pouch
may be square or slightly wider than deep. The opening across the top is
covered by a long triangular flap which sometimes extends slightly below
the bottom of the pouch. A decorative fringe of yar or beads lines the
bottom edge of the pouch.
Sashes are generally wide, 4 inches or more in measured specimens, and
apparently of comparable width in the others (excepting specimen no. 12).
The ends are always lobed or fringed, specimens having 2 lobes (3), 4 lobes,
and 5 lobes respectively. Sashes are attached to the pouches by the edge at
the point where the sash lobes begin.
Design. All decoration is in bead embroidery using a two thread tech-
nique. No instances of cloth applique were seen. Bead colors are light
and dark blue, white, yellow, rose, green, and black. White and blue are the
favorite colors on all specimens with yellow being found on two out of five
Designs themselves exhibit a certain similarity in principle but vary
widely. The repetition of a simple geometric motif is common on sashes, for
example, the X motif (no. 2), the diamond motif (nos. 3, 6, and 10), and the
diamond with recurved ends (nos. 1 and 7). Finally there are motifs which
can simply be called complicated designs (nos. 3, 8, and 13), although no. 3
may perhaps be considered a variation of the double curve motif.
9Specimen no. 9 is excepted from this discussion. It will be considered later.
Decorations on the pouches themselves may (no. 1) or may not (nos. 2, 3?
and 4) be similar to those on the sash. Included are overall designs, (no. 2)
or the repetition of a single element, "stars," nos. 4 and 5; diamonds with
recurved ends, no. 1; and simple diamonds, no. 3.
Fingerwoven Pouch. Special recognition must be made of specimen no. 9.
This appears to be fingerwoven from wool yarn and glass beads. No Seminole
specimen of this type is known to exist at the present time. However, a
Creek specimen of this form seems to be close to the Seminole type (Fig. 6,
Tribal correlations. As we have noted a number of Muskogee and non-
Fig. 6. Comparative Southeastern Pouches. A, finger woven yarn sash and pouch;
Creek; B, beaded sash and pouch, Alabama; C, beaded sash and pouch,
Muskogee "Creek" Indians comprise the group known as the Seminole. At
present two major groups are recognized and we will endeavorto correlate
the earlier groups with one of those. The group called Alachuahere, the
first to enter the state, were Hitchiti speaking Oconee Indians. The St.
Johns Seminole were probably Mikasuki and related to the former. Both can
be correlated with the present Big Cypress group.
Specimens (excepting no. 9) which can be attributed to the Big Cypress
or Mikasuki division are nos. 1, 2, 6, 7, 10, and perhaps 11. As a group
they have certain similarities in sash designs they are all simple repeated
motifs. The use of the diamond with recurved ends is found twice in Billy
Bowlegs band (nos. 1 and 7).
Specimens from the Cow Creek group include no. 8, and probably no. 3.
Both are complicated designs. No. 13 is another complicated design, but its
tribal derivation is uncertain.
Temporal correlations. The great majority of the specimens fall within
a narrow time period 1830-60. Two were collected or in use about 1895, and
a third (no. 5) non-Florida specimen, was collected about the first decade of
the 20th century. By chance the time division happens to correlate with tri-
bal differences so differences between the two groups may be temporal or tri-
bal; probably the latter since the complicated design of the late specimens is
paralleled by another of uncertain tribe (no. 13) but dating from the 1850's.
OTHER INDIAN POUCHES
In order to determine the distinctiveness of the Seminole pouch a survey
has been made of other Southeastern pouches using the same sources of infor-
mation. Specimens illustrated in drawings and paintings are most commonly
Creek and Seminole, none are shown from other tribes in the McKenney and
Hall (1933-34) or Catlin (1844) illustrations. In a painting by Stanley, made
in Oklahoma in 1843, of an Inter-Tribal Council of the Five Civilized Tribes,
a number of decorated pouches are worn but their tribal identity is unknown
(Swanton, 1946, pl. 1).
Creek Pouches. Creek pouches with two exceptions all appear to be
bead embroidered cloth. One museum specimen (Brooklyn Museum, cat. no.
29.1449) and five illustrated specimens of bead embroidered examples have
Form is identical with the Seminole, a square or nearly square pouch
with a long triangular flap and a fringed bottom. The sash is broad and two
specimens whose ends can be seen are bi-lobed.
One example of a simple diamond design on a sash is known, that of
William McIntosh of Coweta worn in a painting, now in the Alabama State
Capitol (Swanton, 1946, pl. 35). The pouch has hourglass-shaped figures
and the flap bears heart-shaped motifs, presumably in bead embroidery.
A probable bead embroidered sash in a zigzag design is that of Mistippee
of Eufaula (McKenney and Hall, 1934, vol. 2, opposite p. 40). Complicated
floral or curvilinear geometric designs occur on three specimens; those worn
by Ben Perryman (Catlin, vol. 2, pl. 219); Yoholo Micco (Eufaula) in McKenney
and Hall (1934, vol. 2, opposite p. 36) and Menawa (Oakfuskee) in McKenney
and Hall (1934, vol. 2, opposite p. 178).
The sash worn by Tustenuggee Emathla may support a pouch (McKenney
and Hall, 1934, vol. 2, opposite p. 173). It is either bead embroidery or beads
plus cloth applique. The asymetrical angular design is unusual.
A finger woven pouch in the U. S. National Museum (cat. no. 315090) has
already been mentioned. It is woven of wool yarn and glass beads. A simi-
lar one may be that worn by Opothleyoholo of Tukabatchee. It has a running-
V design sash (McKenney and Hall, 1934, vol 2, opposite p. 18).
In summary we can note that the cloth and fingerwoven forms of the
Seminole pouch and sash are both found among the Creek. The former type
among both the Creeks and Seminole is very close in form but differs con-
siderably in decorative designs. Creek designs themselves, in the avail-
able sample, vary greatly. This may reflect the number of Creek "towns"
or tribes represented in the sample.
Alabama. In the introduction it was noted that a Western group of
Southeastern tribes was characterized by certain art styles. These tribes
include the Choctaw, Chickasaw, Koasati, and Alabama. Pouches from this
last tribe can be examined for comparative purposes.
Two specimens (Museum of the American Indian, cat. nos. 2/6752, fig.
6, C; 2/5462, fig. 6, B) are typical of a larger group. Like the Seminole
they are made of red wool cloth decorated with bead embroidery, and may be
fringed. There the resemblance in form ends as the pouches lack a flap and
the sash has rounded ends.
Pouches may have simple geometric designs (Fig. 6, B) or elaborate
floral designs (Fig. 6, C). The sashes are nothing more than the usual
decorative shoulder sash typical of the Western division and usually worn
without a pouch (Goggin, n.d.).
We can summarize this group by saying the pouch is square or slightly
deeper than square, often fringed but lacking a flap. Designs are usually
floral. Sashes are broad with rounded ends and of the usual Western form
combining cloth applique and bead embroidery.
Eastern Woodlands. Shoulder pouches have a wide erratic distribution
in the Eastern Woodlands. A major center lies west and south of the Great
Lakes. There the distinctive form is loom woven of beads and often backed
with cloth. Chippewa examples are illustrated by Krieger (1931, pl. 3); a
Sac form may be found in Skinner (1925, pl. 17, fig. 2). Another variety made
of cloth is covered with an overall surface of bead embroidery (Douglas and
D'Harnoncourt, 1941: 160). The pouch in both types is large and firmly
attached to the ends of the sash. It is quite different in form from the typical
Southeastern type, although close to the finger woven type. Embroidery
designs are distinctive local art styles.
Algonkin pouches to the East resemble more closely the Southeastern.
A Delaware example in Harrington (1913, pl. 5) has a 3 lobed sash and a
square pouch. The latter lacks a flap and the aperature is a slit in the
side near the closed top.
SUMMARY AND CONLCUSIONS
A series of Seminole bead embroidered shoulder pouches have been ex-
amined. They show a consistency in form but some variation in decorative
design. On the basis of decorative style there is a suggestion that the
Mikasuki (in its broadest sense) material may be separated from the Cow
Creek. In turn this last, but not too clearly, has more in common in decora-
tive designs with Creek specimens. Creek pouches which have been ex-
amined are identical in form to the Seminole but as a group vary more in
Seminole and Creek material stand together in contrast with pouches
from further west in the Southeast, the Alabama for example. This distinc-
tion between the eastern and western Muskighean speaking peoples has been
noted in other aspects of art and material culture and undoubtedly will be
strengthened as more studies are made in this field.
This letter of Anthony Breath is complete except for closing sentences
of personal nature:
Apalachicola, Florida, Oct. 29, '49
I put in this place this morning in a Norther, being on my way to New
Orleans to copper my steamer. In my last I wrote you from Tampa. I
left there three days since you will see by the papers that we have had
another talk with the Indian chiefs on the 18th. I was at Charlotte harbour
with Genl. Twigg, when Billy Bowleggs and all the other chiefs of the
Seminole and Mickasukie were on board and we had a grand council. They
delivered up three Indians, those that had committed the murder they
brot the hand of one that had been killed in capturing him, the fith one
escaped Sam ones also came in for the first time in his life, he being
old could not come on board but remained on shore. Billy Bowleggs and
myself are quite intimate, he presented me with a fine belt and shot bag
worked by one of the Mrs. Bowleggs. He also gave me his white flag of
truce made of white feathers, but I am fearful we will have trouble with them
yet when it comes to their removal. I was present at the grand council and
could inform you of many things but cannot at present make them public by
order of Genl. Twigg. He, the Genl., left for Pilatke the day before I left.
He will be back in about four weeks, the same time I will. He and myself
are most of the time together. He says my service in the present case will
be needed for a year or two to come ....
Comdg. U. S. Steamer
1853. "Billy Bowlegs and Suite," Gleasons Pictorial, vol. 3, p.
1844. Letters and Notes on the Manners, Customs, and Conditions
of the North American Indians. 2 vols., 4th edition. London.
1896. Hunting and Fishing in Florida. 2nd edition. Boston.
1886. "The George Catlin Indian Gallery," Annual Report of the
United States National Museum for 1885. Washington.
Douglas, Frederic H. and Rene D'Harnoncourt
1941. Indian Art of the United States. New York.
1925. Billy Bowlegs and the Seminole War. Coconut Grove, Florida.
Goggin, John M.
n.d. "Style Areas in Historic Southeastern Art," Proceedings of
the XXIX International Congress of Americanists. In press.
Krieger, Herbert W.
1931. "Aspects of Aboriginal Decorative Art in America Based on
Specimens in the United States National Museum," Annual
Report of the Smithsonian Institution for 1930, pp. 519-556.
McKenney, Thomas and James Hall
1933-34. The Indian Tribes of North America. 3 vols. Edinburgh.
New Orleans Correspondent
1858. "Billy Bowlegs in New Orleans," Harpers Weekly, June 12,
pp. 376-377. New York.
1946. "The Negro Abraham," Florida Historical Quarterly, vol.
25, pp. 1-43. St. Augustine.
1925. "Observations on the Ethnology of the Sauk Indians, Part
III, Notes on Material Culture," Public Museum at the City
of Milwaukee, Bulletin, vol. 5, no. 3. Milwaukee.
Swanton, John R.
1946. "The Indians of the Southeastern United States," Bureau of
American Ethnology, Bulletin, no. 137. Washington.
Wilson, Minnie Moore
1896. The Seminoles of Florida. Philadelphia.
"Indian Costumes in the United States," Science Guide,
No. 63, American Museum of Natural History. New York.
EXPERIMENTS WITH RAW MATERIALS UTILIZED BY THE FLORIDA
INDIANS IN CERAMIC CONSTRUCTION
Hale G. Smith and William Watson
The work on various clay deposits in Florida has been, for the most
part, with those deposits which have a high firing temperature and are uti-
lized by various commercial enterprises (Calver, 1949).
The less vitreous clays and Fullers Earth have received little attention,
and therefore it is believed by the authors that the results of this paper will
be of some value to the Geologist as well as the Ceramicist and the Archeo-
logist. Our thanks is due Robert MacDonald for his assistance in this work.
CLAY AND FULLERS EARTH DEPOSITS AND TEMPERING MEDIUMS
The samples utilized for this work were from the following points in
Area I. This is the locus of a red (iron oxide) non-plastic sand with a
slight amount of clay body. Erosion from highway construction had exposed
these beds which average about 5 feet in thickness. This material was
originally collected to be used as a tempering material, but it was later
found (sample 1) that it was workable into vessel forms.
Area I. Sink hole 3.4 miles south of the City of Tallahassee. A Fullers
Earth deposit occurred at a depth of about 25 feet and was a stratum averag-
ing about 3 feet in thickness. Overlaying the Fullers Earth stratum was a
sandy stratum of a red (iron oxide) non-plastic sand which had penetrated the
Fullers Earth stratum through root action. Atop the red non-plastic sand was
a white sand stratum. The width of these latter two strata is not known due
to the fact that vegetation covered the sides of the sink and only small areas
were exposed to view.
Area III. The material collected occurred along the east bank of the
Suwannee River to the south of U. S. Highway 27 where it crosses the river.
The samples were collected from a depth of about 8 feet where the stratum
is 8 inches thick, although it varied in thickness from area to area. This
relatively thin stratum was overlaid by white sand and some admixture of the
two had occurred due, again, to root action. The material taken from this
locus was more plastic than the Fullers Earth from Area I and made a better
grade of pottery.
Area IV. The locus of the clay deposits here was to the north of the
bridge crossing the Suwannee River on U. S. Highway 19. This clay was
just above the water line of February 19, 1949 and was river levigated. It
occurred in laminated strara of various thicknesses intermixed with sand
layers. Pockets of levigated clay also occurred, but these generally showed
an admixture of sand and a muck-like material. However, some of the poc-
kets contained enough relatively pure clay by volume to be profitably utilized
by the aboriginal peoples of this area.
The assumption may be made, with a certain amount of certainty, how-
ever, no proof is present at this time, that the clay strata encountered at
Mayo, probably extends for some distance, both north and south, along the
A modem commercial ceramicist would consider all of the samples
procured most unsatisfactory for any type of work.
Tempering mediums of white sand, muck, Spanish moss, and crushed
sherds were used. The white sand is that which occurs to the south of the
Pensacola Terrace and overlays, in certain localities, the iron oxide non-
plastic sand. The muck came from a bed of a small lake on the southwestern
city limits of Tallahassee, while the Spanish moss was from an oak tree.
The crushed sherds were a plain type of the Swift Creek Period.
The raw materials, after drying, were pulverized and water added to make
the clay sufficiently plastic for shaping into vessel forms by hand. In some
instances the material was so non-plastic or short that a Bisque Bowl was
used as a support for the base. After the base was shaped in the Bisque
Bowl, coils were added and shaped for the shoulder and rim. Tablets nos.
1, 9, and 10 were short while tablet no. 6 was short and sandy.
The more plastic materials were made into vessels by the coiling and
pinching technique. Samples 2, 3, 4, 5, 15, 16, and 17 were made by this
method. The medium plastic sample vessel was pinched from a solid ball
of clay. Two samples, nos. 7 and 8, tempered with crushed sherds and
quartz pebbles respectively are of medium plasticity.
STEPS IN PRODUCTION
After the raw materials were gathered, they were only refined to the ex-
tent of removing large roots and other foreign materials. For the various
samples either single specimens were utilized or various specimens were
mixed together in various percent by volume.
The tempering mediums when utilized, were added in a percentage by
volume to the clay specimens of white sand, muck, and quartz pebbles. The
amount by volume of muck and crushed sherds was undetermined.
The tablets and vessels were then dried for 36 to 38 hours in a warm
room, and separated into two groups: one to be fired in the electric kiln,
and the other in an open fire on the ground.
The electric kiln at Florida State University is a Glo-Bar type and the
materials were fired in a completely oxidizing atmosphere at 18000 F.
The firing on the ground by methods and techniques utilized by various
aboriginal groups was done near the top of a hill. The wind was blowing at
about 10 miles per hour (subjective judgment) from 1790 south.
The wood for the firing was composed of various oaks, long leaf pine,
and gum. A shallow semi-circular hole was cut out of the red clay (4 inches
maximum depth and 2 feet long) and a fire of hard wood was started and
allowed to become reduced to coals.
The tablets and vessels during the preparation of the fireplace and
fire, were placed in a baking oven (Philco, 1949, Apartment Model) at 1000
and slowly heated to 5000 in order that they would not crack at the higher
temperatures they were to be exposed to later.2
When the tablets and vessels were sufficiently warmed through they
were placed in the hot coals and long leaf pine wood was added in quan-
tities to cover the vessels. When this was burning intensely, oak and gum
were added periodically. In order to get "flash-points" (momentarily higher
temperatures) dried palmetto fronds and pine needles were added. The fir-
ing was continued for two hours and fifteen minutes. The vessels and tab-
lets were then removed and allowed to cool.
It is to be noted that none of the samples were vitrified or approached
vitrification at 18000 F, the highest temperature of the electric kiln. The
temperatures of the ground fire probably did not exceed 12000 F. For this
experiment the amount of vitrification was measured on a scale of 1 to 4,
4 being the highest degree of vitrification in the samples with 1 the lowest.
1This kiln was built according to specifications furnished by Edgar Littlefield,
Head, Ceramic Art Department, Ohio State University, Columbus, Ohio.
2Aboriginal groups after building the initial fire, placed their vessels several feet
away from it. Periodically the vessels would be moved closer to the fire.
A vitrification of 4 (on this scale) is equal to that found in aboriginal cera-
mic types from Florida.
For each mixture of the various earth samples and tempering media,
control tablets were constructed as well as vessels. A 5 an. line was in-
cised upon each control tablet so as to determine the amount of shrinkage
due to drying.
Two sets of tablets were made, one for kiln firing and the other for
open firing. In the tablets themselves there was a differential shrinkage
before firing. This differential shrinkage was due to the amount and quality
of the tempering media present. Because of the nature of the outside fire
some of the tablets were warped. The uneven distribution of heat in the
open fire caused stresses that appeared in the control tablets, but had
little or no effect upon the vessels as a whole. However, shrinkage was
so great in vessels 5, 14, and 17, that cracks appeared. These vessels
were all open fired and therefore the unequal distribution of heat and the
presence of draughts caused these fractures.
Sample 1. The vitrification was low and the plasticity of the material
was short. No tempering media was added due to the already high content
of sand in the sample. A shouldered vessel 6.5 cm. high and 10.9 cm. in
diameter (shoulder diameter) was constructed. Although this vessel could
have been utilized by aboriginal peoples, it is too friable to have been very
practical. Sherds have been collected from aboriginal sites that appear to
contain this material, although they have a higher clay content. These
sherds are early, dating from the Deptford Period on the northwest Florida
coast. Therefore, it is concluded that this material was not utilized to any
great extent by the prehistoric peoples of Florida, and that as their pottery
techniques advanced the use of this material was abandoned.
Sample 2. The vitrification was high and the material was plastic
enough to make a long necked waterbottle. The waterbottle constructed from
this sample is 11 cm. high with a 6.3 cm. neck. The diameter of the body
is 5.3 cm. Applique strips and circular disc-like buttons were also applied.
These decorative features adhered well to the surface without pulling away
or falling off during firing. The clay was utilized in its natural state with-
out any tempering material added. A small amount of sand was already pre-
sent, due to root action carrying it down from the superimposed sand strata.
This material also was probably utilized by aboriginal peoples. Sherds
have been found from various aboriginal sites throughout central Florida and
the northwest Florida coast that have a similar appearance to sample 2.
These sherds fall into the archaeological periods that are post Deptford.
Sample 3. This sample was made up of equal proportions of the iron
oxide stained non-plastic sand (Area I) and Fullers Earth. This is a more
workable and satisfactory paste that sample 1. The vitrification was high. A
waterbottle was constructed from this sample 12.5 cm. high with a 6 cm. neck.
The diameter of the body is 7.2 cm. A strap handle was attached from the
bottom of the neck to the base of the vessel. In comparing it to archaeolo-
gical materials it is most similar to Deptford and Santa Rosa-Swift Creek
types of the northwest Florida coast.
Sample 4. The vitrification was relatively high when compared with the
other samples fired. The material was 50% Fullers Earth and 50% white
sand. A small shallow dish was constructed 8.3 cm. in diameter and 3.1 cm.
high. In all of its properties it was similar to vessel 3 in
and hardness. The tablet upon firing in the open fire was
is similar to Swift Creek ceramics.
warped. This paste
Table 1. Summary of Experiments by Sample.
1-4 after Surink- Aver- Hard-
Sample Plasticity firing Area Temper age Open age ness
1 Short 1 I None 3 mm. 4 mm. 3.5 2.0
2 Plastic 4 II None 6 mm. 5.mm. 5.5 3.5
3 Plastic 4 1/11 50% Mixture 6 mm. 5 mm. 5.5 3.5
4 Plastic 3 II 50% White Sand 5 mm. 4 mm. 2.5 3.5
\ 5 Plastic 4 II 20% Spanish Exploded Explode 3.5
6 Short 2 II 50% Muck 0 mm. 0 mm. 0.0 2.0
7 Medium 2 II 10% Crushed 0 mm. O.mm. 0.0 3.5
8 Medium 2 II 50% Quartz 3 mm. 2 mm. 2.5 3.5
9 Short 3 IV 50% White Sand 2 mm. 3 mm. 2.5 3.5
10 Short 3 II 2% Ground 3 mm. 2 mm. 2.5 2.5
11 Medium 3 1 50% Aluminum 3 mm. 2 mm. 2.5 2.5
Plastic Silicate Clay
12 Plastic 4 III None 2 mm. 2 mm. 2.0 3.5
13 No Material
\14 Medium 3 IV 50% Muck 0 mm. 0 mm. 0.0 2.5
T5- PTIastic 4 III/1 wU% Mixture 3 mm. 2 mm. 2.:5- 3..
16 Plastic 3 IV 50% Red Sand 3 mm. 3 mm. 3.0 3.5
\ 17 Plastic 3 IV 50% Quartz 2 mm. 3 mm. 2.5 3.5
Sample 5. The tempering medium in this sample was 20% by volume of
Spanish Moss mixed with Fullers Earth. The vitrification was high and it
has good plasticity. However, because the Spanish Moss had not completely
dried, the formation of gas bubbles occurred and the shallow bowl constructed
split and warped at the base. No aboriginal sherds are known to the authors
that resemble this specimen. Therefore, it is believed that the makers of
the fiber-tempered'ware used a more plastic clay, and if Spanish Moss was
the tempering medium, the vessels were allowed to dry more thoroughly than
we did this sample.
Sample 6. The paste of this sample was composed of 50% muck and 50%
Fullers Earth. In plasticity it was short. It had a very low amount of vitri-
fication and has a hardness of 2.0. A shallow bowl was constructed 4.5 cm.
high and 7.9 cm. in diameter that maintained its structure during firing, al-
though, it is very easy to crumble the vessel by pinching. This sample has
no correlations to any known aboriginal Florida sherds or vessels. Due to
the poorness of this mixture it is highly doubtful that it was utilized by
any prehistoric group.
Sample 7. The mixture of Fullers Earth with 10%, by volume, of crushed
sherds resulted in a medium plastic material. The cause of this sample
being medium plastic apparently is due to the fact that a higher percentage
of sand was present in this Fullers Earth than in the other Fullers Earth
samples. A cup was made 7.7 cm. high and 6.1 cm. in diameter.
This sample made a very poor paste and is a type that might be com-
parable to some of the Deptford Period pastes.
Sample 8. This paste was medium plastic, again due to the high sand
content in the Fullers Earth. The addition of 50% of quartz pebbles (average
diameter 1.6 mm.) made an even more unsuitable medium for vessel construc-
tion. There are no aboriginal sherds or vessels known to the authors that
approximate this sample.
Sample 9. Clay from Area IV, on the Suwannee River was modified by the
addition of 50% white sand. This resulted in a paste of short plasticity. In
texture it appears similar to gritty aboriginal wares from the Glades Area.
However, it crumbles more easily than does Glades Plain (Goggin, 1939).
Sample 10. Fullers Earth was mixed with 2% by volume of ground moss.
This made up a paste with a short plasticity, although a shouldered vessel
was constructed from this material that made possiblee" ware.
The clay sample had included white sand and therefore it resembles the
Glade Area Glades Plain type of pottery.
Sample 11. Aluminum Silicate clay was mixed 50%, by volume, with the
non-plastic iron oxide stained sand. This yielded a medium plastic paste
that had a relatively high amount of vitrification. The fired pieces resemble
in type and structure, Weeden Island horizon ceramics of the northwest
Sample 12. The clay from Area III was utilized in this sample in its raw
1500 Ft. Walton 4'
A. D. 1 Sample
Fig. 7. Correlation of ceramic tests with archeological horizons of the Northwest
Gulf Coast, Florida (after Goggin, 1950).
form without the addition of any tempering material as a certain amount of
sand was already present in the clay. The paste of this sample is similar
to Weeden Island and Fort Walton types and produced a fairly vitrious form
Sample 13. No material.
Sample 14. To the clay sample from Area IV was mixed 50% muck that
yielded a medium plastic clay with a vitrification of 3 upon the scale set up
for this paper. The clay itself had a certain amount of sand present, as did
the muck. The fired paste is more similar to Fort'Walton paste than any
other Florida aboriginal ware.
Sample 15. An equal proportion of clay from Area Il was mixed with
the Fullers Earth from Area II. This made a very plastic paste. The vitri-
fication of this paste upon firing was relatively high and made a ware com-
parable to the archaeological ware known as Swift Creek (see Fig. 7).
Sample 16. With the clay from Area IV was mixed 50% red sand by
weight. The paste was plastic and had the same vitrification as sample 7.
This paste has the texture and consistency of Weeden Island Paste.
Sample 17. Clay from Area IV was mixed with 50%, by volume, of
quartz pebbles having an average diameter of 1 mm. The addition of this
amount of tempering media did not effect the plasticity. This paste, how-
ever, does not have any resemblance to any aboriginal pottery. It is a mix-
ture that would be unsuitable to make large vessels since after firing there
is a tendency to crumble upon slight pressure.
These experiments have given both the archaeologist and the ceramicist
new insight into problems of a very different nature. For the Archeologist,
who, for some time, has been in doubt as to what clays were utilized by the
aborigines of Florida, this controlled experiment helps in determining not
only what materials were used, but also the capabilities of various raw ma-
terials. In Figure 7 we see that certain combinations of clays and temper-
ing media produced samples similar to aboriginal wares of most of the cul-
tural horizons of the northwest Florida coast. It is seen that both short
and plastic materials might have been utilized by the Deptford peoples.
On the other hand, the more plastic materials were evidently utilized ex-
clusively by the Swift Creek, Weeden Island, and Fort Walton peoples. The
differences that occur are caused mainly by the variety of tempering media
It is seen now that more work will have to be done in the location and
analysis of various clay deposits, and the mixing of various proportions of
tempering material with clays before our series will be complete.
To the archeologist and the ceramicist this experiment gave a clearer
understanding of the problems the Indians had to meet in reference to the
gathering and processing of the materials, and the techniques of manufacture
of ceramic vessels.
The teaching ceramicist, through a study of the aboriginal techniques
of material preparation, construction and design, finds these things a val-
uable teaching aid for the instruction of beginning students. The utiliza-
tion of the Bisque Bowl and the coiling technique requires little equipment
and gives the student necessary basic training in techniques of ceramic
Calver, James L.
1949. "Florida Kaolins and Clays," Information Circular No. 2,
Florida Geological Survey. Tallahassee.
Goggin, John M.
1939. "A Ceramic Sequence in South Florida," New Mexico
Anthropologist, vol. 3, pp. 36-40. Albuquerque, New
1950. "Florida Archaeology -- 1950," The Florida Anthropolo-
gist, vol. 3, nos. 1-2, pp. 9-20. Gainesville.
Florida State University
THE GARD SITE, HOMOSASSA SPRINGS, FLORIDA
Ripley P. Bullen
During the spring of 1951, Mr. James Harman of Homosassa Springs,
Florida, preparatory to building a house and driveway for Mr. Neal Gard of
Cleveland Heights, Ohio, spread the upper part of a small burial mound on
Rendevous Island on the north shore of Homosassa River. A large palm
tree, standing near the original center of the mound, was then removed.
Many human bones including five skulls were found in the roots of this
tree. It was also noted that spreading of the mound had scattered frag-
ments of human bones and a few sherds. The latter were collected by Mr.
A. C. (Dazzy) Vance and given to Mr. Gard.
Upon reading a newspaper account of this discovery, the Archaeological
Survey of the Florida Board of Parks and Historic Memorials visited the
site the next day and was given various bones including parts of four skulls
and the pottery mentioned above. The fifth skull, said to have been the best
preserved, had been left in the roots of the tree overnight but had disappeared
before morning. It broke during removal.
As the bones showed but little evidence of deterioration in some cases
the marrow voids had been filled with a lime deposit excavation was indi-
cated to secure measurable skeletal material. Permission to do this work
was kindly granted by Mr. Gard, the owner.
Originally the mound was two or three feet high and about twenty-five
feet across. It is said to have been partially surrounded by low areas or
barrow pits. In removing the upper part of the mound many limestone boul-
ders were found. These, undoubtedly, came from the barrow pits and were
included by Indians in the mound fill.
Excavation revealed the remaining site profile to consist of four zones.
Uppermost was a superficial disturbed zone, two to three inches thick near
the center of the mound, which contained bones, scattered and broken by
the work of spreading the mound. The next zone, about one foot thick and
composed of black-brown sand, contained bundle burials, a few sherds and
other artifacts, a fair number of chert chips, some charcoal, rarely a food
bone, and occasionally, fragments of limestone. The third zone, made
up of dark gray sand with which were mixed a great many pieces of lime-
stone, up to eighteen inches in greatest dimension, was culturally essential-
ly sterile. The larger pieces of limestone tended to concentrate in the lower
part of this zone. Below all was light gray marl.
Excavation centered around the hole left by the uprooted palm. Tests in
other parts of the mound did not produce burials but excavation was in no
sense complete. Occasional isolated bones in the black-brown zone and a
broken humerus, laterally displaced an inch and a half, as well as lack of
skulls with burials, suggested the mound may have been dug into many years
ago. However, all of this may have been the result of root action as well as
aboriginal disturbance. Also, as will be suggested later, lack of skulls with
bundle burials may reflect burial habits of the users of the mound.
Practically all bones excavated by us were broken. To a large extent
this may be explained by pressure exerted by the passing of trucks over the
present surface during and after spreading of the mound. Skeletal material,
while slightly eroded, did not exhibit the extreme chemical decomposition or
eating away commonly found in many of the mounds of Florida. However,
none of the excavated material was impregnated with lime deposits as were
some of the bones from under the palm tree.
We found no bones lower than the base of the black-brown zone. Examina-
tion of bones in the roots of the palm had suggested that bones would be
found down to the top of the marl. Apparently, the palm tree not only gave
mechanical protection to the bones below it but also forced many of them
down into the gray sand zone where, due to the large number of limestone
rocks, there was an excess of calcium carbonate. Unfortunately, such
conditions of preservation did not occur in the rest of the mound.
Spaces among the tree roots, where burials had been found, suggested
bundle burials. As they had been removed prior to our arrival, this point
could not be definitely determined. All burials uncovered during excavation
were of the bundle type. The following comments include what may be said
on the bones and their associations. The fifth skull from the base of the
tree has been omitted.
Burial A Reconstructed calvarium from base of tree. About 163 mm.
long, 142 mm. wide, cephalic index 87%. Skull has lateral
asymmetry as left parietal appears depressed. Deep and
wide grooves behind mastoid processes. Middle to old aged
adult, probably female.
Burial B Reconstructed calvarium from base of tree. About 170 mm.
long, 137 mm. wide, cephalic index 80.5%. Skull has later-
al asymmetry as right parietal appears depressed. Deep
grooves behind mastoid processes. Old adult, possibly
Burial C Fragmentary skull from base of tree. Bones are those of a
middle aged adult.
Burial D Fragmentary skull from base of tree. Bones are those of a
young child, approximately 12 years of age.
Fragmentary long bones and mandibles, presumedly be-
longing to the above skulls (and to the purloined fifth skull),
seem to merit but little comment. Except in the case of the
child, tooth wear is considerable but not excessive. One
caries was noted. One lowe. jaw had lost all six molar
teeth in life. Some shovel-shaped incisors are present.
Burial E Very fragmentary skull and jaw of a child. Shovel-shaped
incisor teeth. Age about 10 years. Piece of turtle carapace
Burial F Bundle burial consisting of femura, tibiae, fibulae, and
one humerus laid parallel with ulna and radius at an angle
below, all occupying an area 9 by 12 inches. Part of a
left inominate bone exhibits a wide, shallow notch. Pro-
Burial G Femura and fibulae lying parallel, fragmentary bones ex-
tending 12 inches to one side. Pelvis at one end with toe
bones in pelvic cavity. All lying on area of charcoal, %
inch thick. Bones not burnt. Probably middle aged male.
One chert chip, one St. Johns Check Stamped, and two other
sherds in fill.
Burial H Area of broken bones, apparently disturbed. Three chert
chips in fill. St. Johns Check Stamped sherd and fragment
worked Busycon shell below bones.
Burial I Skull, foot bones, and few ribs. Skull on north side of palm
tree hole with jaw touching edge of hole. Skull on left side,
face to east, other bones below. Partial reconstruction
suggests brachycephalic female. In left part of coronal
suture is what appears to be the edge of an old break or
Burial J Fragments of child's skull with small femur extending
southwest of skull. No teeth or jaw.
Burial K Bundle of parallel long bones occupying an area 12 by 20
inches with, just below, a humerus extending towards east
and a femur extending towards west. Three inches above
extended head of this femur was a greenstone celt (Fig. 8,
G). In center of bundle was an undecorated sherd. In-
cluded in fill were two St. Johns Plain, one St. Johns
Check Stamped, a cord-marked sherd, a chert knife (Fig. 8,
B), and eight chips of chert. Bones include various long
bones, an articulated tibia, fibula, and talus, parts of two
inominate bones, portions of a maxilla, and a right temporal
bone. The later exhibits a large mastoid process with a
deep and wide groove behind it. Undoubtedly, a fairly large,
Discussion of the Gard burial mound is limited by the incomplete nature
of the data. Damage to the mound and to the skeletal material occurred
during spreading of the mound and the uprooting of the palm tree. There is
also a possibility the mound was dug into many years ago.
There seems to be no question, however, but that all interments were
bundle burials. As all known skulls are from the tree hole or adjacent
thereto, it seems likely these remains may represent a group burial with the
skulls more or less centrally located.
Apparently, the bones were laid on the surface or in a shallow hole
(or holes) and earth from the periphery piled up to form a mound. Chips,
sherds, and occasional bits of shells and of turtle carapaces suggest the
site to been a habitation area before becoming a burial mound.
We have two collections of pottery from the Bard mound, sherds from
the excavations and those from the surface. The latter, scattered by the
spreading operation and collected by A. V. (Dazzy) Vance, must have come
from higher zones than the excavated sherds. As indicated in the follow-
ing lists, these collections differ substantially from each other.
Lake Jackson Plain 1
Gritty Plain 8 1
St. Johns Plain 1 5
St. Johns Check Stamped 2 6
Limestone and sherd (?) tempered 6
Wakulla Check Stamped 1
Totals 13 21
The Lake Jackson Plain sherd has a loop handle. Six of the eight
gritty plain sherds in the surface collection are similar in paste and sur-
face to the Lake Jackson Plain sherd. Chalky pottery, on the other hand,
came chiefly from the excavations. Some of the St. Johns Check Stamped
sherds were found below burials. The cord-impressed sherd from the surface
collection exhibits crossed imprints of cords adjacent to an area with paral-
lel imprints of cords (lands between individual cord imprints in both areas)
while cord-marked sherds from the excavations bear only parallel closely
The pottery, therefore, appears to pertain to two archaeological periods,
Safety Harbor and (probably very late) Weeden Island II. The bundle burials
may belong to either period. The stone celt (Fig. 8, G) is of a type usually
considered Weeden Island in date. As it was found over some bones, the
latter should, in part at least, date from that period. If a transitional Weeden
Fig. 8. Stone and shell material, Gard Site.
Island-Safety Harbor date is assumed for burials at the Gard mound, they
should have been interred at about 1400 A. D.
Artifacts from the Gard mound include a greenstone celt, two chert
knives, two large shell beads, a piece of cut Busycon shell, and a
Busycon pick. Of these the celt and one of the knives (Fig. 8, B) may
have been associated with Burial K. The other items did not appear to
have any special burial associations. All are illustrated in Figure 8.
Florida Park Service
Excavations at Kolomoki, Season I 1948 by William H. Sears, University
of Georgia Series in Anthropology, No. 2., 48 pp., 7 pli., 8 figs., map.
Athens, Ga. 1951.
The extensive Kolomoki site in southwestern Georgia, now a state park,
includes one large and two small temple mounds and four burial mounds
surrounded by village debris. Work described in this monograph tests
and small excavations made in midden deposits, was done in 1948 to aid
in park planning and park development. Information was secured to permit
location of roads and to appraise the site in terms of archaeological com-
Sears found evidence of four occupations as follows, from early to late:
Weeden Island and/or Swift Creek, Kolomoki, and Lamar. Kolomoki is a
new cultural unit based on middens which produced nearly exclusively two
new pottery types, Kolomoki Complicated Stamped and Kolomoki Plain. Many
of these vessels had flat bases, either square or disc-shaped. In the case of
the complicated stamped vessels, decoration was applied in a zone separated
from the rim by an undecorated area. Previously, Kolomoki Complicated
Stamped containers would have been included under the broader term, Swift
Creek Complicated Stamped (Late Variety).
As there is little evidence of an intensive occupation by makers of Lamar
pottery, Sears discounts them as builders of the temple mounds. He notes
from surface collections that Kolomoki remains are distributed in a semi-
circle surrounding two of the temple mounds. Pending further excavation,
it would seem that bearers of that culture constructed these mounds.
Kolomoki is placed later than Weeden Island at the site because Weeden
Island sherds were present in small quantities in Kolomoki middens but the
reverse was not true. Similarly, the late Swift Creek manifestation is placed
earlier than Kolomoki. Sears sees no evidence so far at the site to indicate
the relative temporal position of Weeden Island and Swift Creek.
The horizontal distribution of Weeden Island pottery differs from that of
Kolomoki although there is some overlap. No doubt one or more of the burial
mounds will prove to be of the Weeden Island period. As Sears points out,
this is the most northern Weeden Island village so far known.
The 1948 excavations at Kolomoki give a good view of cultural complexes
present and suggest what may be expected as further work is done. As might
be expected, various interpretive problems are raised. These Sears discusses
in a stimulating manner which indicates his grasp of the situation and under-
standing of culture dynamics. Prudently, he omits any definite conclusions
until more data is available.
Such problems include the origin of Kolomoki Complicated Stamped pot-
tery and the lack of any Aakulla Check Stamped sherds at the site. Addi-
tional work may fill in the latter gap or indicate one or more periods during
which the site was not occupied. The data available are too meager for
complete evaluation of the large Kolomoki site at this time.
Excavations at Kolomoki would have been improved by more careful
editing. The few typographical errors are of minor importance but lack of
a directional arrow on the frontispiece and incorrect and incomplete textual
references to plates are serious obstacles to one trying to use the report.
Sears has given us new information about southwestern Georgia. We
await with interest publications describing more recent work so that these
preliminary findings may be properly evaluated.
Ripley P. Bullen
Florida Park Service
CONTRIBUTORS TO THIS ISSUE
Hale G. Smith. Dr. Smith, past President of our society is Assistant
Professor of Anthropology at Florida State University.
William Watson. Mr. Watson teaches ceramics in the Department of Art,
Florida State University.
Ripley P. Bullen. Well known to our readers, Mr. Bullen is Assistant
Archeologist, Florida Park Service.
John M. Goggin. Dr. Goggin teaches Anthropology at the University of
FLORIDA ANTHROPOLOGICAL SOCIETY
Albert C. Holt, 3011 Riverside Ave., Jacksonville,
Ist Vice President:
2nd Vice President:
E. Y. Guernsay, Cocoa, Florida.
Frederick W. Sleight, P. O. Box 94, Mt. Dora.
John W. Griffin, Florida Park Service, Seagle Build-
ing, Gainesville, Florida.
Ripley P. Bullen, Florida Park Service, Seagle
Building, Gainesville, Florida.
John M. Goggin, Department of Sociology and
Anthropology, University of Florida, Gainesville,
Hale G. Smith, Florida State University, Tallahas-
H. James Gut, Sanford, Florida.
T. Lynn Smith, University of Florida, Gainesville,
Address general inquiries to the Secretary. Membership is open to all
interested in the aims of the Society, dues $3.00 per year. Application may
be made to the Treasurer. Address him also concerning receipt of publica-
Send all manuscripts to the Editor.
Address items for the Newsletter to the President.
LIST OF PUBLICATIONS
FLORIDA ANTHROPOLOGICAL SOCIETY
The Florida Anthropologist
Vol. I, Nos. 1-2, 3-4
Vol. II, Nos. 1-2, 3-4
Vol. III, Nos. 1-2, 3-4
Vol. IV, Nos. 1-2 Each Double Number
Non-Members - - - - -1.50
News Letter, Florida Anthropological Society
-----___ ____ _0.15
Publications, Florida Anthropologica Society
No. 1 Hale G. Smith, "Two Archeological Sites in Brevard
County, Florida" ------------ -0.50
No. 2 John W. Griffin and Ripley P. Bullen, "The Safety
Harbor Site, Pinellas County, Florida" -0.50
No 3 Ripley P. Bullen, "The .Terra Ceia Site, Manatee
County, Florida" - - - -0.50
These may be obtained from the Treasurer:
Mr. Ripley P. Bullen
Florida Park Service