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
f R 7 /7 -`q
THE FLORIDA ANTHROPOLOGIST
Membership is open to all interested in the aims of the
Society. The Florida Anthropologist is published by the Florida
Anthropological Society during the months of March, June, Sept-
ember, and December. Subscription is by membership; annual dues
are $3.00 ($1.50 for students),of which $1.50 is for subscription
to The Florida Anthropologist and other publications as issued.
The Florida Anthropologist is published at The Department of An-
thropology, The Florida State University, Tallahassee, Florida.
Application for Second Class privileges entered at Tallahas-
see, Florida on April 22, 1958 under the Act of March 3, 1879.
Applications and orders for back issues should be sent to
the Treasurer (each single number to members, $.50; each double
number, $1.00; to non-members, $.75, and $1.50 respectively;
Newsletters Nos. 1-35, $.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. 1. "Two Archaeological Sites in Brevard County, Florida,"
by Hale G. Smith. 32 pages, 4 plates.................0.50
No. 2. "The Safety Harbor Site, Pinellas County, Florida,"
by John V. Griffin and Ripley P. Bullen.
43 pages, 4 plates................................. 0.50
No. 3. "The Terra Ceia Site, Manatee County, Florida,"
by Ripley P. Bullen, 48 pages, 7 plates..............0.50
No. 4. "The European and the Indian," by Hale G. Smith,
150 pages, frontispiece, 6 maps....................2.00
No. 5. "Florida Anthropology," C.H. Fairbanks, ed............1.25
President: William H. Sears Fla. State
Museum, Gainesville, Florida
1st Vice President: John M. Goggin, 312 Peabody Hall
Univ. of Fla., Gainesville Fla.
2nd Vice President: Marvin J. Brooke,805 N.W. 15th Ct.
Miami 33, Florida
Secretary: D.D. Laxson, 231 W. 41st Street,
Treasurer: Hale G. Smith Box 3051 Florida
State University, Tallahassee Fla.
Editor: Charles H. Fairbanks Box 3051
Fla. State Univ.,Tallahassee,Fla.
Executive Committeeman: Irving Rouse, Yale Univ.
W.J. Armistead, Tampa
The Florida Anthropoloeist, vol.ll,no.2,1958
A STRATIFIED EARLY SITE AT SILVER SPRINGS, FLORIDA
Wilfred T. Neill
Various eastern states have yielded projectile points
typologically similar to Paleo-Indian artifacts, Rarely, how-
ever, have such points been found in the East under circumstan-
ces permitting a determination of their relative or absolute
age. It is therefore of interest to report on a stratified
early site at Silver Springs, Marion County, Florida.
I am especially indebted to Ripley P. Bullen, Curator
of Social Sciences at the Florida State Museum, for his valua-
ble suggestions as to techniques, interpretations, and identi-
fications, as well as for assistance in excavation. Thanks are
also due to E. Ross Allen, director of the Reptile Institute
with which I am associated, for his active support of the study
even though it lay outside his own field of herpetology. I am
grateful to W. Carl Ray, owner of the site, for his whole-hear-
ted cooperation. I am also indebted to John W. Griffin, of the
St. Augustine Historical Society, for useful information and for
aid in reviewing pertinent literature; to H. H. Winters, for-
merly of the Florida State Geological Survey, for suggestions
concerning the geology of the site; and to H. James Gut for
A report on the site was presented before the Fifth
Annual Meeting of the Florida Anthropological Society, at Gain-
esville, Florida, on February 28, 1953.
The site lies on, or more properly in, a wooded hill
bordering Silver Springs run, about a half-mile below the head
springs, on the south side of the stream. Over a period of years
a segment of the hill was removed and used as sand full in nearby
areas. This removal was accomplished by men with hand shovels;
their activities did not disturb the rest of the hill, except for
a little slump along the cut face. In spread sand from the hill
were found somewhat Clovis-like projectile points, as well as a
few sherds and flint artifacts of more usual types. The circums-
tance awakened the hope that excavation might reveal definite
The hill is dune-like, and composed of homogeneous, seem-
ingly windblown sand. Such hills characterize the surrounding
region; their geologic origin is uncertain. The sand is about
89 to 97 inches deep in the excavated portion of the site. It
ests on a laminated formation of alternating sand and reddish
clay. The thickness of this underlying formation is not known.
The clay laminate become progressively thinner, and the sand
laminae progressively thicker, toward the upper portion of the
laminated formation; and the uppermost clay lamina is very thin
and broken. In other words, the laminated formation passes gradu-
ally into the overlying sand, without evidence of a disconformity.
The laminated formation yields only claystone, quartz pebbles,
limonitic concretions, and what appear to be water-worn marine
fossils; but the overlying sand contains a number of approximate-
ly horizontal occupation levels. These levels suggest that the
sand of the hill has gradually built up through the centuries;
and that the hill top, at all times nearly horizontal, was occup-
ied off and on by different groups of people as the sand accumu-
The hill drops off gradually to the river swamp. The river
never encroached upon the excavated portion of the site during
the period of human occupancy.
The river, called Silver Springs Run or Silver River,
originates from large springs about a half-mile above the site.
The water wells up through fissures in a limestone formation. The
average annual flow from these fissures is approximately 500
million gallons per day. The stream meanders for a distance of
about seven miles before entering the Oklawaha River, a tribu-
tary of the St. Johns.
The general area today abounds with fish and game, and no
doubt did so in past times as well. Deer and turkey are still
common in the vicinity, along with opossup, raccoon, gray fox,
otter, wildcat, gray squirrel, fox squirrel, cottontail rabbit,
marsh rabbit, cormorant, herons, resident and migrant ducks,
Florida gallinule, coot, alligator, gopher tortoise, three
species of large water turtles, bullfrogs, gars, bowfin, large
catfishes, eel, black bass, various sunfishes, striped mullet,
giant shrimp (Macrobrachium), clam, aquatic snails, and other
edible creatures. Bear and panther still occur in the county.
Vegetation is diversified. Within three miles of the site
are such plant associations as high pine, rosemary scrub, live-
oak hammock, mesic hardwood forest, cypress swamp, low hammock,
and pine flatwoods.
The fissures of the head springs have yielded abundant re-
mains of Columbian mammoth, Florida Mastodon, black bear, all-
igator, various turtles, and large birds. Probably the area has
supported a rich biota throughout the period of human occupancy.
The hill top lies between the river and a small flint-/out-
cropping which had been quarried by Indians. The nearby prese-
nce of this quarry might explain why the hill had been occupied
sporadically by Indian groups throughout the centuries. It would
be quite natural, after quarrying some flint, to retire to the
adjacent hill top, which overlooks the river, Most of the occu-
pation levels at the site are very thin, consisting mainly of
scattered chips and bits of charcoal, as though small groups of
people had each spent but a short while on the spot.
As stated previously, a segment of the hill had been re-
moved by workmen, leaving behind an irregular cut face. Eleven
test excavations were made into this cut face, from the surface
to a depth of eight feet or more. The tests were confined to
the relatively flat hill top, in an apparently successful effort
to avoid slope wash. Eventually some of the tests were merged
with each other, forming three larger and three smaller exca-
vations. The positions and extent of these excavations are
shown in an accompanying map.
The sand was removed by troweling. Each artifact or flint
chip recovered was marked with the number of its test exca-
vation and the distance from the surface of that excavation.
Well-defined occupation levels appeared at each test. These
levels proved to be nearly horizontal. The distance from the
present surface to the laminated formation varied from about
89 to 97 inches among the eleven test excavations. About 500
square feet of the eight-foot level were exposed.
The site was visited almost daily throughout October,
November, and the first half of December, 1952; and also dur-
ing January and the last half of February,1953. Some additi-
onal excavation was done at intervals during latter 1953,
1954, and 1955.
The following account can best be understood by refer-
ence to the accompanying map, whereon the test excavations
are designated by letter or by a combination of letter and
number. Most of the important artifacts are illustrated in
An occupation level was found about seven inches below
the present surface. It was represented at D2 by charcoal
and flint chips, and at E by charcoal, chips, and a sherd
tentatively identified as Little Manatee Complicated Stamped
ware (P1. 1, A). Some of the chips were of a pink flint,
probably imported. A single chip of local flint occurred at
12 inches at F2. Charcoal and chips of local flint were also
found at depths of 16 and 17 inches at F2. Chips of local
flint occurred at depths of 25 to 27 inches at F2, and one
chip of imported (pink) flint at 27 inches in D2.
An occupation level was found at depths of 36 to 39 inches
and was represented at tests A2, A3, D2, E, Fl, and F2. It
yeilded numerous chips of local and imported flint, abundant
charcoal, sherds, cores, and various flint artifacts. A sherd
0 10 20 30
SCALE IN FEET
Pl. 1. Artifacts from upper (ceramic) levels of the site. For
further identifications, see text.
of Dunn's Crepk Red ware (P1. 1, B) was found at 36 inches in
D2, and another at 36 inches in A2. The latter was touching a
broken projectile point (P1. 1, C). This point was thin and
flat, with a fine pressure retouch. It was made of imported
pink flint. Three other broken points (P1. 1, D-F), similar
to the first in material, workmanship, and typology, were
found at depths of 36 or 37 inches in A2. A somewhat larger
point (P1. 1, G), unbroken, was found at 38 inches in A3. It,
too, was of a pinkish flint which apparently does not occur
locally. Along with the last was a broken leaflike knife (P1.
1,H) of a whitish flint which is perhaps of local origin. In
D2 were found a fragmentary projectile point (P1. 1, I) of
imported material, a fragmentary projectile point of local
material, a utilized flake of imported stone, a triangular
scraper of local flint, and a large chunk of local flint at
depths of 37 to 39 inches. The last mentioned item may be a
broken chopping tool, but is more likely an exhausted core.
Two similar chunks, almost surely cores, were found at 39
inches at Fl, and another core at 38 inches in E. An edged
fragment was found at 36 inches in A3.
Chips were found at 41 and 42 inches in F2.
An occupation level occurred at depths of 44 and 45 inches
in B, C, Dl, D2, E, Fl, and F2. It consisted of charcoal,
scattered chips, one sherd, and two flint artifacts. One flint
chip was of imported and the remainder of local material. The
sherd was a rim fragment of Orange Incised ware (P1. 1, J)
from 44 inches at F2. The flint artifacts were a worked flake
found at 44 inches in Fl, and the stem of a projectile point
(P1. 2, A) at 44 inches in C. Although but a few inches sepa-
rate this zone (44-45 inches) from the next (48-54 inches), it
is believed that the two are distinct. The 44-45 inch occupa-
tion zone was characterized by scattered chipping and a
paucity of artifacts; it was lacking from Al, A2, A3, and
A4. In contrast, the 48-54 inch zone was rich with chips, char-
coal and artifacts; it occurred throughout the site.
A well-defined level, from 48 through 54 inches, was pres-
ent at all tests. It yielded abundant chips, charcoal, and art-
ifacts. All the chips were of local material. A stemmed projec-
tile point (PI. 2, B) was found at 50 inches in Fl, and two others
at 51 inches in C. They were of local flint, rather rudely made
by percussion flaking, but with a fine retouch along the edges.
One of these is shown in P1. 2, C. A roughly heart-shaped tool
(PI. 2, D) was found at 52 inches in D1, and another (P1. 2, E)
at 48 inches in A2, They were of local material. A small drill
(P1. 2, F) of local flint was found at 54 inches in Fl, and a
larger drill (P1. 2, G), made from a flake, at 48 inches in E.
A large scraper (P1. 2, H), probably reworked from a broken
pick or spade-like tool, occurred at 52 inches in C. It, too,
was of local flint. A broken center section of a large blade
was found at 49 inches and the broken end of a pick-like tool
at 51 inches, in D2. These last two artifacts were of pink flint
believed to be imported. Probably they are quarry tools. An ed-
ged fragment was found at 50 inches in D1. A large, heavy flint
tool was found at 48 inches in C. Four similar tools were found
in spread sand from the site.
The 48-54 inch zone yielded no sherds. Fiber-tempered wares
had a long time span in this area, yet at the present site Orange
Incised ware was sandwiched between Dunn's Creek Red ware at
36 inches and an apparently preceramic level beginning at 48
inches. Of course, further excavation may yield fiber-tempered
sherds at greater depths. However, the 48-54 inch zone was the
richest at the site, and it seems as though a few sherds woula
have been found if pottery were present.
Scattered chips were found at 55, 56, 57, 58, and 59
inches in tests A2, A3, A4, D2, E, Fl, and F2. Apparently the
55-59 inch zone represents a period of time when the site was
mostly unoccupied although visited sporadically by flint knappers.
At depths of 60 through 63 inches was found evidence of a brief
occupation, in the form of two projectile points (P1. 2, I-J),
an edged fragment, a large utilized flake, and what may be a
rude hated scraper (P1. 2, K). The flake was found in test E,
F / 7 ;,
\i" .--'.-. /
Pl. 2. "A": stem of projectile point from ceramic level of the
site. "B-M": artifacts from supposed preceramic Archaic
levels. For further identifications, see text.
the others in F2. The edged fragment, hafted scraper, and
points are thick and heavy, crudely made by direct percussion.
Charcoal and scattered chips were found at depths of 64,
65, and 66 inches, in A2, A3, A4, E, and F2. At 67 inches in
F2 were found the tip of a projectile point (P1. 2, L) and the
end of a spade-like tool (P1. 2, M). Both of these are well
made; the workmanship is reminiscent of Paleo-Indian technique.
Charcoal and a few scattered chips were found at depths of
68 and 69 inches, in tests A3, A4, and F2. At 70 inches in A3
was found a large utilized flake. Chips and charcoal occurred
at depths of 70 through 74 inches in tests A3, A4, B, and F2.
Below 74 inches was a foot or more of nearly sterile sand.
Elsewhere in the hill one could often see the faint discolora-
tions resulting from charcoal and perhaps organic wastes. In
the 74-87 inch zone there were no such discolorations. However,
a large utilized chunk of flint was found at 77 inches in Fl,
and a few chips at 81 inches in Fl and F2.
Finally, at depths of 88 to 93 inches below the present
surface, and separated by one to four inches from the uppermost
clay lamina, was another occupation level, relatively thick
with chips, charcoal, and artifacts. This level, the deepest
at the site, yielded fragmentary Clovis-like projectile points,
a chopper, utilized flakes, graver-like tools, a scraper, sand-
stone abraders, a bit of fossil shell, and a problematical ob-
ject (perhaps only a concretion). The Clovis-like projectile
points from spread sand agreed with those found in situ, in
typology, material, workmanship, and degree of patination.
Items found at this lowest stratigraphic occupation level
will now be described. The broken base of a point (P1. 3, J)
was found at 90 inches in Al, along with a point minus its base
(P1. 3, A). Another base (P1. 3, K) was found at 89 inches in
A2, and one (P1. 3, G) at a similar depth in E. A broken point
(P1. 3, D), made of fossiliferous flint, was found at 93 inches
The most nearly perfect of the Clovis-like points (PI. 3,
B-C) were found in spread sand rather than in situ. This is re-
P1. 3. Artifacts from the lowest occupation level of the site.
For further identifications, see text.
grettable but not surprising, for laborers probably had removed
nine-tenths of the site before my studies began. The following
description of the points is based on two entire and two broken
specimens found in spread sand from the site, and five broken
specimens discovered in situ.
The points vary in length from 72 to an estimated 100 am.;
in maximum width from 26 to at least 40 mm.; and in maximum
thickness from 6 to at least 13 mm. The average dimensions are
as follows; length about 88, width 33, and thickness 9 mm.
These are mostly large, heavy points. However, they are all
symmetric and excellently made. Bases may be fluted or merely
thinned. When a flute is present, it is wide and shallow, and
extends decidedly less than half the length of the blade. None
of the points could be called strongly fluted. Near the base,
the sides of the point curve gently in toward each other and
then out again. Some trace of this "basal constriction" seems
to be present in all specimens. The base of the point varies
from truncate to shallowly concave, with a tendency toward
"ears". In the more nearly complete points, the basal edges
have been dulled; in the more fragmentary ones they have not
been so treated. The fragments may represent points broken
during manufacture. The abundant chips at the deepest level,
and the nearby presence of a flint outcropping, suggest that
projectile points were made on the spot. Some chips, chunks,
and one point are of a rather characteristic fossiliferous
flint, which occurs at the quarry.
The deepest level yielded nine utilized flakes. Four of
these seem to have been deliberately edged by knapping; the re-
mainder show fine chipping along the edge, probably as a result
of use. Two of the flakes bear small, sharp points; perhaps
they would be classed as gravers. One flake is of pink flint;
all other chipped stone artifacts and spalls from this level
seem to be of the local white flint. The nine utilized flakes
were found at 90 and 91 inches in Al, 90 inches in A2, 92 and
93 inches in B, 89 and 90 inches in E, and 90 and 91 inches in
Fl. Seven of them are figured(P1. 3, E, H-,I L-0.
A sandstone abrader was found at 90 inches in Al, and
another at 89 inches in F2. The latter bears several channels
which look as though they had been made during the sharpening
of some pointed instrument such as a bone awl.
A chopping tool was found at 89 inches in Al.
A rude, plano-convex side scraper was found at 89 inches
A roughly pentagonal bit of fossil shell was found at 90
inches in Al. It may have been worked. Fossil shells are com-
mon in the local limestone.
A problematical object was found at 90 inches in Al. In
appearance it suggested the tooth of a horse, broken off, worn
down, etched and softened by soil acids. However, H. James Gut,
paleontologist of Sanford, Florida, believes it to be only a
concretion. The object was found along with the base of a pro-
jectile point and the blade of another point, (PL3,F)all with-
in an area of about 15 square inches and at the same depth. 2
Digging was done in the laminated formation. The only
objects found were quartz pebbles, limonitic concretions, clay-
stone, a fragment of what appears to be fossil coral, and what
may be a water-worn fossil sea-urchin.
Flint artifacts from the upper four feet of the site show
little or no trace of patination. Items from the 48-54 inch
zone bear only a slight patina. (A single specimen from this
zone is rather heavily patinated; some variation in patina
might be expected as the result of a difference in material.)
Artifacts from the 60-63 inch zone are heavily patinated, and
those from the 67-inch level are even more so. In items from
the lowest level, the patina is extremely heavy, as compared
with material from the higher levels and also as compared with
Paleo-Indian artifacts from various western sites. One blade
has become chalky and crumbly along the edges, and in most of
the specimens the outlines of the chipping scars have been dull-
ed. This increase in patination toward the deeper levels is
evidence that stratigraphy was undisturbed in the excavated
portion of the site. Throughout the site were found scattered,
irregular lumps of claystone, apparently formed in place.
These lumps become larger and much more numerous toward the
deeper levels, thus providing additional evidence of undisturb-
A stump that had burned in place was found in test F2.
Its original base appeared to have been somewhere between the
50- and 60-inch levels. A large chopping tool was found at 56
inches, about four feet away from the stump.
The site yielded no trace of bone or unfossilized shell.
The only organic material was a gdod bit of charcoal. The
latter, although abundant in almost every zone, was usually in
the form of scattered grains or lumps.
Discussion and Summary
Eleven test excavations were made into the cut face of a
hill bordering Silver Springs Run, in Marion County, Florida.
The dune-like hill was composed of apparently wind-blown sand,
about 89 to 97 inches deep, conformably overlying a laminated
formation of sand and reddish clay. The laminated formation
yielded no evidence of human presence, hut several nearly hor-
zontal occupation levels were distinguishable in the overlying
sand. It is suggested that the popularity of the hill top, in
past times, was due to the nearby existence of a flint quarry.
A sherd tentatively identified as Little Manatee Complica-
ted Stamped ware was found at a depth of 7 inches below the
present surface. An occupation level occurred at depths of 36
to 39 inches; it yielded Dunn's Creek Red ware and flint arti-
facts made by pressure flaking. A sherd of Orange Incised pottery
(a fiber-tempered ware) was found at 44 inches. A relatively rich
zone of occupation at depths of 48 to 54 inches produced fairly
well made flint artifacts but no pottery. Rude artifacts made by
direct percussion were discovered at 60 to 63 inches. Two well
made artifacts occurred at 67 inches. A zone from 74 to 87 inches
was nearly, but not completely, sterile. The lowest occupation
level occurred at depths of 88 to 93 inches below the present
surface, and about one to four inches above the uppermost clay
lamina. It yielded Clovis-like projectile points, a chopper,
utilized flakes, graver-like tools and scraper-gravers, a scraper,
sandstone abraders, an apparently worked bit of fossil shell, and
a problematical object perhaps not artifactual,
Clovis-like projectile points from the Silver Springs site
are relatively large and heavy, although symmetric and well made.
They are characterized by a truncate or shallowly concave base,
a basal constriction, and a tendency toward "ears." Bases may be
fluted or merely thinned. When a flute is present, it is wide
and shallow, extending less than half the length of the blade.
None of the points could be called strongly fluted. Patina is
very heavy. I have referred to these points as being "Clovis-like"
but the term is used loosely. They might equally well be called
"Plainview-like." Goggin (1949, p. 20) stated, "One possible ex-
pression (of the Paleo-Indian tradition) are the many Plainview
and fluted points, locally called Suwannee Points, which occur
in the Central Florida region." In several later publications
Goggin has referred to "Suwannee Points." He has not defined the
term, but made it clear (Goggin, 1950, p. 46, footnote 2) that he
had in mind the artifacts described by Simpson (1948). Goggin
(1952, p. 65, footnote'3) applied the name "fluted Suwannee
Points" to the artifacts from the lowest occupation level of the
Silver Springs site.
At this site, the Suwannee Points occupied the lowest
stratigraphic occupation level, occurring below what is thought
to be a preceramic Archaic horizon. However, the 74- to 87- inch
zone, separating the Suwannee artifacts from the overlying
Archaic material, was not completely sterile; it yielded a few
spalls. I do not believe these spalls were intrusive into this
zone, for they were lying flat, as though on a former land
surface; and there was no visible evidence of disturbance.
Apparently, then, the site never went unoccupied for any great
length of time. The first comers were the makers of Suwannee
Points; and they were succeeded by people who left spalls in
the 74- to 87- inch zone. These in turn were succeeded by pre-
ceramic Archaic Indians, who either lived in the immediate
vicinity or (more likely) visited the locality at frequent in-
tervals in connection with quarrying operations. Late Archaic
visitors left a sherd of Orange Incised pottery. Early Wood-
land people were probably responsible for the sherd of Dunn's
Creek Red ware and associated projectile points; while the
sherd thought to be Little Manatee Complicated Staped ware
may be attributed to a middle Woodland group. Thus the mat-
erial found, while sparse, would seem to cover a remarkably
long time span; and throws light on the chronological position
of Suwannee Points relative to Archaic and later manifesta-
1 Using the term flint in its archeological sense, to denote
any of the cryptocrystalline varieties of silicon dioxide.
2 For some reason Seaberg (1955, p. 68) stated "mammoth
and mastodon bones at 88 to 93 inches are associated in situ"
with the Clovis-like artifacts at the Silver Springs site.
This is not the case; remains of extinct vertebrates occur in
the head springs but have not been found at the site.
Dn Robert 0. Vernon, Director of the Florida Geological
Survey, tells me that, in his opinion, this laminated sand and
clay is the result of cyclic ground-water deposition. It is pro-
bably not a member of the Hawthorn strata and is probably Pleis-
tooene in time. At any rate it seems not to be involved in the
deposition of artifacts discussed by Dr. Neill.
Goggin, John M.
1949. "Cultural Traditions in Florida Prehistory." In
The Florida Indian and His Neighbors (John W.
Griffin,editor), pp. 13-44. Winter Park.
1950. "An Early Lithic Complex from Central Florida."
American Antiquity, Vol. 16. No. 1, pp. 46-49.
1952. "Space and Time Perspective in Northern St. Johns
Archeology, Florida." Yale University Publications
in Anthropology, No. 47. pp. 1-147, ple. 1-12.
1955. The Zetrouer Site: Indian and Spaniard in Central
FlIrida. Master's thesis, University of Florida.
Simpson, J. Clarence
1948. "Folsom-like Points from Florida." The Florida
Anthropologist, Vol. 1, nos. 1-2, pp. 11-15.
ROSS ALLEN REPTILE INSTITUTE
SILVER SPRINGS, FLORIDA
- 49 -
A i A
P1. 4. Test B. The two yardsticks point to spalls in their
original position, in the lowest stratigraphic occupa-
tion level of the site. The close-up (right) shows the
two uppermost clay laminae (indicated by arrows).
A portion of the site in 1957, after work had ceased.
Some slump has been dug away to show the laminated formation and
its relation to the overlying sand.
PI. 5. Laminae 11 to 12 feet below the surface and about 3 to 4
feet below the lowest occupation level.
Al A2 A3 A4 B C D1 D2 E Fl F2
N. ___ -
N... A P ,A P. ... .*.
STRATIGRAPHY OF THE SILVER SPRINGS SITE
Lt Sherd of Little Manatee Complicated Stamped ware
Q: Quarrying tool (?), spade- or pick-like
Dt Sherd of Dunn's Creek Red ware Ut Utilized flake
Ps One or more projectile points Ks Knife
OQ Sherd of Orange Incised ware Si Scraper
Hs Heart-shaped tool of flint Cs Core
Tt Flint tool, large and heavy Dri Drill
F: Fragment of worked flint At Abrader
Stippled areas represent levels at which flint chips and/or bits
of charcoal were found. For lack of space a few finds are omitted
from the chart, in cases where several artifacts were found at
the same level in one test.
The Florida Anthropologist, vol.ll,no2,1958
Some Problems of the Origin of Creek Pottery*
Charles H. Fairbanks
The ceramics of the historic Creek settlements of the Geo-
rgia- Alabama area have been known for some time, but the origins
of this pottery are not yet clear. It is evident that a number of
distinct traditions are involved in the formation of the Creek
ceramic tradition. This paper will attempt to define some of the
complexes involved. They are graphically shown ih the accompany-
Creek ceramics are best known from the excavations in the
vicinity of the historic Carolinian Trading Post at Ocmulgee Nat-
ional Monument, Georgia (Kelly, 1939, and 1938). Additional hist-
oric Creek sites investigated by Gordon R. Willey in central and
western Georgia show a highly similar ceramic complex. Of these,
only the Kasita Site near Columbus, Georgia has been reported
(Willey and Sears,1956). Unreported sites are the Big Sandy Site,
the Ennis Site, and the Tarver Place, all in central Georgia.
Collections from these sites are at Ocmulgee National Monument
and exhibit essentially identical pottery to that from the Ocmul-
gee Fields and Kasita sites. All of these date from the closing
years of the 17th and the first half of the 18th centuries. Dur-
ing this period the Creek were in full contact with the French
and English traders as well as Spanish friars and military pers-
onnel. Of these influences, the British deer-skin trade seems to
have exercised the greatest impact on the Creek.
The Creek ceramic complex is composed of two general forms
as a general rule, open bowls with incised or painted decoration
and jars with "roughened" surfaces. Specifically the tradition
contains Ocmulgee Fields Incised (Jennings and Fairbanks, 1939),
Ocmulgee Fields Plain, Ocmulgee Check Stamped (Fairbanks, Ms B),
Walnut Roughened (Jennings & Fairbanks,1940), Chattahoochee Bru-
shed (Bullen, 1950), Kasita Red Filmed (Jennings & Fairbanks,
1940), and a few miscellaneous forms. Each of these will be dis-
cussed in turn with some speculations on its origins and rela-
tionships. Together they make up a consistently recurring cer-
amic complex with minor regional variations in form, decoration,
and proportion among the various types.
Ocmulgee Fields Incised is perhaps the most characteristic
of all the historic Creek pottery types and may well be the most
abundant type in the central Georgia area. It is found on two
vessel shapes: the cazuela and the plate. In shape the type seems
clearly derived from the earlier Lamar Bold Incised cazuela. In
both types we find a rounded or occasionally flattened base with
definite shoulder angle rather high up. The Ocmulgee Fields Inc-
ised bowls have plain basal parts below the shoulder, whereas the
Lamar type has, generally, a complicated stamped basal part.
There are plain cazuelas in Lamar however. It is not known at
present just how common this plain type is at the earlier level.
The rim area is incised with a band-like design in both cases,
this feature being largely the connection between the two types,
Lamar Bold Incised and Ocmulgee Fields Incised. As has been
pointed out frequently, Lamar Bold Incised is a vigorous, bold,
rythmic incising, while Ocmulgee Fields Incised has a sloppy,
careless, often smothed-over incising (Fairbanks, 1952,pp.298-9).
Thus we have a similar form with closely similar pattern or loc-
ation of the decoration, suggesting that the relationship is
closer than simply a derivation from a common ancestor. I see the
loss of tamping and the smoothing of the incising as part of the
general trend away from elaborately textured surfaces which seems
to be characteristic of the southeast as a whole during this time.
At the Lamar site there are a few vessels that represent a
transitional phase between Lamar and Ocmulgee Fields in this re-
spect. These cazuelas have plain bases, incised rims, and a part-
ial smoothing of the rim which blurs and reduces the strength of
the incising. They would be counted as Lamar Bold Incised without
any hesitation but seem to forecast the Ocmulgee Fields type. This
intermediate stage is also found in northwest Florida in the type
In the lip and shoulder angle we find a minor, but con-
sistant, difference between Lamar and the Historic type. Ocmulgee
Fields Incised usually has a complex, extruded, and varied lip
form. This may be basically flat or rounded. It is almost always,
however, characterized by an extrusion or protrusion of one sort
or another. This is in marked contrast to the universally simple,
flat,or rounded lips of Lamar Bold Incised. This seems to be the
only case where the Ocmulgee Fields form is more complex than the
earlier type. Lamar cazuelas are often decorated with a row of
circular or annular punctations along the shoulder angle at the
base of the incised band. As far as I am aware this feature is
never found on Ocmulgee Fields Incised. It is replaced by a rare
series of notches. The use of notches in place of circular or
annular punctations reflects the general trend of simplification
from Lamar to the historic level. Finally, Ocmulgee Fields cazue-
las seem, on the average, to be smaller than the equivalent form
In summary, the common form of cazuela bowl in Lamar and
the historic Creek levels is the same, is decorated in the same
way, and in the same place. Some design motifs can even be traced
from Lamar to a somewhat slovenly form in Ocmulgee Fields. The
two pottery types differ, aside from ware characteristics, mostly
in the minor differences of lip form, shoulder-angle punctation,
and in the smoothing after incising of the later type.
The second form of Ocmulgee Fields Incised, the plate or
bowl with flaring rim, is less easy to trace from a Lamar proto-
type. Flaring rimmed bowls are known from Lamar but are far from
common. They seem much more common in Ocmulgee Fields and become
a regular feature of the type Ocmulgee Fields Incised. In both
cases the incised decoration is placed on the upper or interior
surface of the rim. Multiple-line guilloches and twin-lined loops
are found in both forms. It seems clear that the plate in the
historic Creek levels is derived from the earlier bowl with flar-
ing rim of Lamar. Goggin has suggested, in conversations, that
this form in the San Marcos complex of coastal Georgia and north-
eastern Florida may be derived from Spanish majolica plate forms.
It is certainly a common shape in Spanish ceramics. Its presence
in both Lamar and Ocmulgee Fields may indicate this source. The
incising on these plates is the same, weak, slovenly, smoothed
type as on the cazuelas.
Ocmulgee Fields Plain is a common type representing mostly
a "sorting" type for the plain basal parts of Ocmulgee Fields
Incised. There is, however, a small number of completely plain
jars and cazuela bowls that seem to be identical in ware charact-
eristics with Ocmulgee Fields Incised. The type seems to repre-
sent a continuation of Lamar Plain into the historic horizon.
This material offers little in the way of information on the ori-
gens of Creek pottery because of the difficulty of dealing with
Ocmulgee Check Stamped is a smothed-over check stamped
type found on the deep jar characteristic of Walnut Roughened and
on the cazuela bowl characteristic of Ocmulgee Fields Incised. It
represents the survival of the late check stamped types like Mer-
cier Check Stamped (Sears, 1951, p. 32) or Boyd Check Stamped
(Caldwell, 1955). It also illustrates the common historic Creek
trait of smoothing surfaces after decoration. I have pointed out
(Fairbanks, Ms A) that the stamping technique in the southeast is
probably at least partly functional in the construction of the
jar. Thus the smoothing of a stamped surface is not the contra-
diction that it seems to be if it is considered as wholly a dec-
orative feature. Rim and lip forms and decorations in Ocmulgee
Check Stamped are very similar to those for Chattahoochee Brushed
and Walnut Roughened. Their derivation will be discussed with
those types. Here it is sufficient to point out that Ocmulgee
Check Stamped forms one of many links in historic Creek pottery
with the Lamar period and seems to follow the same general pro-
gressive trends as other members of the ceramic complex exhibit.
Walnut Roughened and Chattahoochee Brushed must be consid-
ered together as they form one decorative constellation. Walnut
Roughened is shell tempered; Chattahoochee Brushed is grit or
sand tempered. In other respects they seem to be very similar.
Some shell temper is present in Lamar but it is by no means typi-
cal. The bodies of the types are roughened by a variety of tech-
niques such as brushing, stippling,and cob-marking. In many cases
it is difficult to decide just what to term the roughening
process. In all cases it is not the deep scoring of Mississippi
Valley types and I have previously expressed the opinion that it
is a sort of substitute for the earlier stamping. This would seem
especially relevant in the case of cob-marked forms. The neck
area in all cases is either plain or incised withsharply cut
lines, generally widely spaced chevrons or diagonals. This type
of incised or plain neck is not common on the Lamar period deep
jars but is occasionally found, at Irene Mound. The lip treat-
ment is more similar to Lamar in that it has an added fillet of
clay at or just below the lip. This fillet is usually pinched or
more rarely notched. In this respect is closely resembles the
pinched rim fillet of Lamar Complicated Stamped. It is probably
derived from Lamar rim fillets rather than from the notched fill-
et on open bowls as in the Dallas Focus and other Late Mississ-
ippian foci. At least in Walnut Roughened and Lamar Complicated
Stamped we have the shared feature of a deep jar with slightly
flaring rim and a pinched or notched rim fillet. Of course, all
the notched or pinched rim fillets of the Late Mississippian foci
are surely related. I am simply arguing for a closer relationship
between Lamar and Walnut Roughened than between Walnut Roughened
and forms like Dallas. Walnut Roughened and Chattahoochee Brushed
sometimes have vestigial strap handles just under the lip. This
is evidently to be traced to the prevalent use of strap handles
to the west as illustrated in the presumably Upper Creek ceramics
in the vicinity of Montogomery, Alabama.
Kasita Red Filmed is perhaps the second rarest (after
Ocmulgee Check Stamped) of the members of the historic Creek ce-
ramic complex. It is usually decorated only with red paint of a
rather fugitive type. Now, red, and even red and white direct
painting, is known from Lamar times. It is tempting to
derive Kasita Red Filmed from the nameless red filmed of Lamar.
The vessel form,however, seems to be a plate, or extremely shal-
low bowl with wide flaring rim. Only the rim area may be painted,
or designs outlined by incised lines may cover much of the inter-
ior. This plate form seems to be definitely related to the plate
form so common in majolica and other Spanish ceramics of the per-
iod. The zoning and the use of outlining incised lines is, how-
ever, more directly related to tendencies in late Weeden Island
times. Mission Red Filmed (Smith, 1948, pp. 316-7) in Florida is
a very similar type and seems to share the same origins. The ring
base is fairly common in Kasita Red Filmed and Mission Red Filmed.
This seems clearly related to the prevalent use of ring bases on
majolica or other wheel-made forms. It is essentially foreign to
southeastern ceramics. Wherever it occurs it is, I believe, to
be derived from European forms or from native pottery that was
heavily influenced by European forms.Rarely in Kasita Red Filmed
I have seen black and red paint on buff backgrounds. The black
is a thin, watery pigment that resembles in appearance, at least,
the black of various negative painted pottery types in the east-
ern United States. I believe this is a vegetable paint but have
no accurate information on it. This is on a paste similar to
Kasita Red Filmed and is not to be confused with the rare sherds
of Mexican polychrome wares found in the historic levels of the
southeast. Red filming is also rather common in the historic
Guale pottery found at Ft. King George on the Georgia Coast and
at present classified as San Marcos Red. Thus we see a general
appearance of red filming, all over interiors or in zoned areas,
in the southeastern historic period in the Georgia interior and
on the coast as well as on the northwest Florida red hills. It
may be only remotely related to Lamar red painted forms. It is
closely related to Weeden Island Zoned Red (Willey, 1949, p.422.)
and is certainly heavily influenced by Spanish wheel-made ceramics.
A number of minor types or forms are present that present a
miscellaneous category of historic Creek ceramics. One vessel,
from the trading post at Ocmulgee National Monument, was shaped
like a typical southeastern shallow basket with four corners and
a rounded base. It is typical of Ocmulgee Fields Incised in hav-
ing an incised band on the rim area. This is so similar to the
southeastern basket shape that it must surely have been derived
from that form. It is interesting in that it seems to take the
basket form back to the early historic period, rather that for
any light it throws on the origins of Creek pottery. From the
trading post area and an area north of the railroad cut just out-
side the monument boundary come two vessels that clearly indicate
the European influences effecting Creek pottery. The first is a
small Kasita Red Filmed mug with single loop handle from rim to
base. The red filming covers the entire surface of the mug. The
shape is quite obviously a direct copy of English or German salt-
glazed, stoneware mugs of the end of the 17th century. Its shape,
ring base, and single, large loop handle are quite foreign to the
southeastern ceramic tradition. The second jar might almost be
called a wide-necked bottle. It has a ring foot, rounded body,
long, curved and slightly flaring neck, and pinched rim-fillet.
It might be argued that this shape is derived from the Mississip-
pian carafe bottle shape. The total impression, however, is that
of an European "vase" shape and I feel sure the form is heavily
influenced by European, probably English, pottery.
In summary we find that Walnut Roughened and Chattahoochee
Brushed show a number of characteristics that relate them to
Lamar Complicated Stamped. Ocmulgee Fields Incised derives in de-
coration and shape from Lamar Bold Incised. Ocmulgee Check Stamp-
ed derives directly from Mercier Check Stamped. Kasita Red Film-
ed may derive from an unnamed Lamar red painted type but is clear-
ly heavily influenced by Spanish majolica types and from Weeden
Island Zoned Red. Creek ceramics, then, are shown to have a num-
ber of resemblances to the ceramics of the Lamar period which im-
mediately preceded the historic Creek occupation of most of the
Georgia area. It is tempting to derive the Ocmulgee Fields pot-
tery complex from Lamar and I have done so in previous papers. We
should, however, consider the ceramics identified as Cherokee be-
fore making any comprehensive statement.
- 59 -
Sears (1952) and Caldwell (1955) have recently described some
of the pottery types found in the historic Cherokee settlements
of northern Georgia. Complicated stamped pottery of the Lamar
style, check stamped, and pinched rim fillets are common. The
ceramic complex is clearly and closely related to the Lamar ceram-
ic complex. Sears (1956) points out that this argues that if
Cherokee is Lamar, then Lamar must be Cherokee and can never be
Creek. I think this syllogism is false. Certainly historic
Creek pottery contains a heavy increment of Lamar traits. What
is sure is that Creek pottery is not unmodified Lamar, while
Cherokee pottery is relatively "pure" Lamar. Now, Lamar is the
end product of a long evolution in Georgia of the Southeastern
Stamped Tradition (Sears, 1952; Fairbanks, Ms a). The Cherokee
are Iroquoian in linguistic affiliation;the Creek are Muskhogean.
It seems that one or both of these groups must have migrated into
the Southeast. Both have migration legends, that of the Creek
being somewhat the more elaborate. Both share a general culture
pattern that is definitely southeastern with the Cherokee being,
perhaps, somewhat more specialized towards a mountain environment.
I do not believe we can determine the linguistic affiliation of
the prehistoric Lamar people from the ceramic affiliations of
either historic Creek or Cherokee.
This situation is not unique in the eastern United States,
as we find a somewhat similar situation in the New York State
area. There the rather well known Iroquoian pottery has been
shown by Ritchie and MacNeish (1949) to have a number of relation-
ships with the Owasco and Point Peninsula series which preceded
it. They say that "In both the Point Peninsula and the Owasco
series definite trends are observable and, indeed, certain of
these trends appear to link the two discrete series, while in
turn some of the observable Owasco tendencies seem to continue on
into Iroquois," (Ritchie and MacNeish, 1949, p. 119). They go on
to suggest that the discontinuities between Point Peninsula, Owas-
co, and Iroquois ceramics may be due to their lack of excavation
of key sites or to movements of new peoples. They also state that
"Both the Bell-Philhower and Bainbridge components appear to be
derivatives of the Owasco developmental stage revealed at the
Castle Creek station, but to represent respectively, different
later historic tribal groups," (Ritchie and MacNeish, 1949, p.
122). They see Bainbridge as proto-Mohawk while Bell-Philhower
is quite definitely Munsee Delaware (Ritchie, 1949). The point
of all this is that in the New York State area and, on the histor-
ic level, a single ceramic complex was shared by peoples of two
separate linguistic stocks, Iroquoian and Algonquian.
Whether we assume that the Iroquoian speakers were at home
in the southern or northern Appalachians, it is obvious that one
or both of these groups must have derived their pottery from peo-
ples already existing in the area they occupied in historic times.
When we consider that the Algonquian Munsee also share many of
the ceramic traits of the Iroquois, we see that the situation is
much like that in the southern Piedmont. In each area two lin-
guistically diverse groups shared ceramic complexes that have con-
tinuities with change from earlier ceramic complexes. The con-
clusion, it would seem to me is that one or more of these groups
migrated into their respective areas and became heavily accultur-
ated by the ceramic styles of the pre-existing population. It is
apparent that we will need to use much more detailed comparisons
before we can assume that any prehistoric ceramic complex is the
work of a specific linguistic group. Anthropologists have for
long maintained that there need be no connection between specific
material traits and either race or language. They have also for
long been aware that cultural content often diffuses faster than
cultural style. It behoves us to be careful of making one-to-one
assumptions of identity between language and pottery.
- 61 -
Bullen, Ripley P.
1950. An archeological Survey of the Chattahoochee River
Valley in Florida. Jour. of the Wash. Acad.of
Caldwell, Joseph R.
1955. Cherokee Pottery from Northern Georgia. American
Antiquity, 20:3. 277-80.
Fairbanks, Charles H.
1952. Creek and Pre-Creek In "Archeology of Eastern U.S.",
J. B. Griffin, ed., 285-300.
Ms a. The Southeastern stamped Tradition.
Ms b. Ocmulgee Check Stamped, type description.
Jennings, Jesse D. and Charles H. Fairbanks
1939. Type description Ocmulgee Fields Incised. South-
eastern Archeological Conference, Newsletter, vol. I
1940. Type descriptions Walnut Roughened and Kasita Red
Filmed, Southeastern Archeological Conference, News-
letter, vol. II.
Kelly, Arthur R.
1938. A Preliminary Report on Archaelogical Explorations at
Macon, Georgia. Bureau of American Ethnology, Bul-
letin 119, 1-68.
1939. Macon Trading Post, an Historical Foundling. American
Antiquity, 4:4, 328-33.
Ritchie, William A.
1949. The Bell-Philhower Site, Sussex County, New Jersey-
Indiana Historical Society Prehistory Research Series,
Ritchie, William A., and Richard S. MacNeish
1949. The Pre-Iroquoian Pottery of New York State. American
Antiquity, 15:2, 97-123.
Smith, Hale G.
1948. Two Historical Archaeological Periods in Florida.
American Antiquity, 13:4, 313-9.
Willey, Gordon R.
1949. Archeology of the Florida Gulf Coast. Smithsonian
Institution, Misc. Collections, vol. 113.
Willey, Gordon R., and William H. Sears
1956. The Kasita Site. Southern Indian Studies, 4, 3-18.
Department of Anthropology
The Florida State University
- 63 -
Continuities Be+ween Lamar And Historic Creek Ceramics
OCMULGE FIELDS ROUGHENED, OCMULGEE
INCISED CHATAHOOCHEE PLAIN
I I I
/ I I
MERCIER BOLD INCISED
LAMAR COMPLICATED LAMAR
LAMAR ISLAND SPANISH
RED FILMED ZONED
KEY: Continuity with change in vessel shape.
-: -- Continuity with change in decoration or surface finish.
COmings, William P.
1958. The Southeast in Early Maps. With an annotated Check
list of printed and manuscript regional and local maps
of Southeastern North America during the colonial
period. Princeton, Princeton Univ. Press, $12.50
A valuable and welcome discussion of the "families"
of maps of the southeast. Portrays the development of
geographical knowledge of the area. Scholarily, com-
plete and of real use to all historical researchers.
Harris, Walter A.
1958. Here the Creeks Sat Down. Macon, Georgia, J. W.
Based on South Carolinian English documents and
does not cover the material from Spanish sources. This
book reflects the late General Harris' lifelong inter-
est in Ocmulgee Old Fields, Emperior Brim, and Henry
Woodward. Mainly concerns central Georgia, but there
is some Florida material.
Tebean. Charlton U.
1957. Florida's last frontier. The history of Collier County.
Miami, University of Miami Press.
Specifically concerns Collier County but is a veil
researched and written account that is informative a-
bout a much broader field.
Copeland, Leeila S. and J. E. Dovell
1957. La Florida. Its land and its people. Austin Texas,
A handy, concise history of Florida from Indian
THE FLORIDA ANTHROPOLOGIST
A Stratified Early Site at Silver Springs, Florida
Wilfred T. Neill 33
Some Problems of the Origin of Creek Pottery
Charles H. Fairbanks 53
Booknotices Editor Inside back cover
The Florida Anthropologist published manuscripts on any
subject pertaining to Florida or southeastern anthropology.
Manuscripts should be typed on one side of the sheet only,
with 65 spaces to the line, 35 lines to the page. Tables
must be typed in final form with inked lines. Citations
should follow the style of The American Anthropologist.
Footnotes and bibliography should be typed on separate pages.
Florida Anthropology, issued as a Joint publication of
the Florida Anthropological Society and the Department of
Anthropology. Florida State University, was mis-numbered an
the title page. This should read:
Published jointly by
The Florida Anthropological Society Publications. No. 5
The Florida State University Department of Anthropology,
Notes in Anthropology, vol. 2.