2000 Florida Anthropological Society Inc.
The Florida Anthropological Society Inc. holds
source text of the Florida Anthropologist
considered the copyright holder for the text
all rights to the
and shall be
and images of
The Florida Anthropological Society has made this publication
available to the University of Florida, for purposes of
digitization and Internet distribution.
The Florida Anthropological Society reserves all rights to this
publication. All uses, excluding those made under "fair use"
provisions of U.S. Code, Title 17, Section 107 are restricted.
Contact the Florida Anthropological Society for additional
information and permissions.
THE FLORIDA ANTHROPOLOGIST is published quarterly in March, June,
September, and December by the Florida Anthropological Society, Inc., c/o
Room 102, Florida State Museum, University of Florida, Gainesville, FL 32601.
Subscription is by membership in the Society for individuals or institutions in-
terested in the aims of the Society. Annual dues are $4.00; student members
$2.00. Requests for memberships and general enquires should be addressed to
the secretary; subscriptions, dues, changes of address, and back issue orders
to the treasurer; newsletter items to the president; and manuscripts for publi-
cation to the editor. Second class postage paid at Gainesville, Florida.
XXV, No. 3
Clear Fork Gouge and Greenbrier Point by L. O. Warren . . .
Coleman Site, Dade County, Florida by D. L. vonBurger ......
Duda Ranch Shell Mound, Brevard County, Florida by C. F. Knoderer
Artifacts from the Trail Site, Dade County, Florida by W. F. Coleman
Panamanian Duhos by Deborah Brandt . .. . . . . ..
Early Historical Period Canoe by D. L. vonBurger . . . ..
Burials from Green Lake, Texas by R. J. Wingate and T. R. Hester .
Meadowbrook Farms No. 2 Site by B. L Waller . . . .
Concretions Associated with Glades Prehistoric Sites by B. Mowers .
An Engraving Tool by R. P. Bullen and M. T. Wallace. . . ..
OFFICERS OF THE SOCIETY
President George Magruder
Parkway Palms Apts., No. 235B
Indian Harbour Beach, FL. 32937
1st Vice President John W. Griffin
46 St. George St., St. Augustine, FL. 32084
2nd Vice Presidnet Adelaide K. Bullen
Florida State Museum, University of Florida
Gainesville, Florida 32601
Secretary Cliff E. Mattox
P.O. Box 521, Cocoa Beach, FL. 32931
Treasurer Leon Reyniers, P. O. Box 8451
Ft. Lauderdale, Florida 33310
Editor-Resident Agent Ripley P. Bullen
Florida State Museum, University of Florida
Gainesville, Florida 32601
Three years: Yulee Lazarus
Ft. Walton Beach, Florida
Two years: Wilma B. Williams
One year: Thomas Gouchnour
At large, for one year:
Charles A. Hoffmann, Jr.,
Benjamin I. Waller,
A CLEAR FORK GOUGE AND GREENBRIER POINT
FROM THE GULF OF MEXICO, PINELLAS COUNTY, FLORIDA
Lyman O. Warren
Cyrus Ray in 1938 described the "Clear Fork Gouge" from the town of
Clear Fork on the Brazos River of Texas. The context of the Clear Fork
culture complex was thought to be early, perhaps 4000 to 7000 B. C. The
gouge is defined as a core tool, bifacially worked, with a wide gouge end or
bit which is more often concave than straight or convex; the sides or edges
taper gradually to a rounded point or pole end. Gouges are percussion flaked.
They measure from 1 1/2 to 5 inches in length. A Clear Fork gouge is de-
picted on page 16 of Sun Circles and Human Hands (Fundaburk and Fore-
In Florida, at the Dixie Lime Caves near Ocala, Bullen and Benson
(1964) found Clear Fork gouges in association with side-notched Bolen and
Bolen-like (Greenbrier) points (as well as later stemmed forms). The
gouges have also occasionally been found on the surface in other parts of
the state (Simpson Collection, Florida State Museum).
The present paper reports a new flint workshop or workshop-habita-
tion site which was initially characterized by Hernando type projectile
points, but in one part of which, more recently, two side notched points
have been found in a dredged-up context with a Clear Fork gouge.
The site, known as Storm Harbor Marina, about 3 miles south of Tar-
pon Springs, is on the property of Mr. Robert Storm, now of Colorado, who
kindly gave permission to dig. It was first called to my attention by John
Parker of St. Petersburg, to whom I wish to express my thanks. The marina
is located at the most westerly end of the Klosterman Road, just north of the
road itself. The site, an artificial one, was created when Mr. Storm had
dug an eighteen foot deep channel in a north and south direction, opening to
the south on Klosterman Bayou, a part of the Gulf of Mexico. The channel
was dredged, not drag-lined, and the sand and green and black clay and peaty
soil were piped onto the low land to the north and east to give it a foot or so
of elevation. The surface was littered with spalls and a number of artifacts,
but sherds and mineralized bone fragments were not to be found. By the time
I had a chance to study the area it had been pretty well denuded of significant
surface finds, although a few beautifully executed examples of Hernando
points, and one or two non-descript stemmed specimens were picked up on
Florida Anthropologist, vol. 25, no. 3, September 1972
CLEAR FORK GOUGE
Fortunately, however, numerous tumuli or little piles or shallow
mounds of sand, clay, and concretions presented themselves here and there.
From our experience at the pumped up site at Apollo Beach in Ruskin we
knew that these hillocks represented where the outlets of the dredge pipes
had been, and where the heavier materials tended to become concentrated,
while lighter materials, mostly sand and water, had flowed off down hill.
Several of these tumuli produced a large number of spalls. Three of them
gave up the artifacts depicted in this paper.
Figure 1 (a) shows a typical Clear Fork Gouge measuring 2 1/2 by
1 3/4 inches in length and breadth, and 3/4 inches in thickness. One face
is a little more convex than the other, but both are bifacially worked by
heavy percussion flaking. The working edge or bit is slightly concave, and
its termini form sharply defined right angles with the two side edges. The
left edge is slightly convex and battered as for a finger hold, and the other
edges are sharp and somewhat sinuous.
From the same hillock came the two side notched points
depicted in Figure 1 (b). Neither point shows the classical beveling that
would be expected in a fully typical Bolen Point (or Big Sandy I, as they are
called in Alabama). They might be called non-beveled Bolens (Florida ter-
minology), or again, to use the Alabama nomenclature, "Greenbriers".
The Bolen base shown in Figure 1 (b), right, has the same color, patina, and
workmanship as the gouge, and the stone material is almost identical, a
silicified limestone or chert with small shell inclusions. The basal edge is
quite sharp, but the side notches are definitely ground. A basal longitudinal
flake has been removed producing a quasi-fluted effect.
The other point, from which the lower right tang or ear is unfortunately
missing, is a much more elegant object with skillful edge flaking, or re-touch-
ing, superimposed on larger horizontal flakes which meet at the midline axis
to form a shallow ridge (more pronounced on one face than the other). The
angle at the tip is unusually obtuse. A basal longitudinal flake has been re-
moved on the face depicted (Figure l(b), left) producing a pseudo-fluted
effect, clearly visible in the photograph.
The objects shown in Figure 2 came out of almost contiguous moundlets
and from an identical matrix of coarse, rust-colored sand and greenish clay
with admixtures of peat; they have a brown or peaty color like those of Figure 1.
Figure 2 (a-c) are small core tools, bifacially worked, probably better termed
scrapers than gouges. They may be completely unrelated, except spatially,
to the Clear Fork Gouge, or, conceivably, they may be Clearfork variants.
In Figure 2 (d) is depicted what may be a small drill, bit end to the
left. Figure 2 (e) is a small, thin, plano-convex knife or scraper. Not shown
6 '1 N
Fig. 1. Small Clear Fork gouge
and notched points
and side-notched points
Fig. 2. Small core tools, drill,
uniface knife or scraper.
CLEAR FORK GOUGE
from this dredged up matrix are a small, high crowned, plano-convex scra-
per, having the shape of a small "horse' s hoof scraper" and the volume of
a golf ball; also two small hammer-stones of approximately the same size,
which look as if they had been used in the fingers for precise percussion
rather than in the palm for heavy blows. A number of brownish spalls of
small size were also found.
Two side notched points were found in a dredged up site in association
with core scrapers, one of which resembles strongly the typical "Clear Fork
Gouge" of Texas.
Ray, Cyrus N.
1938 The Clear Fork Culture Complex. Texas Archeological and
Paleontological Society, vol. 12, pp. 2ZZ3-237. Abilene.
Fundaburk, E. L., and Foreman, Mary D.
1957 Sun Circles and Human Hands. Luverne, Alabama.
Bullen, R. P. and Benson, C.A.
1964 Dixie Lime Caves. Florida Anthropologist, vol. 17, no. 3,
pp. 153-64. Tallahassee.
St. Petersburg, Florida
THE COLEMAN SITE, DADE COUNTY, FLORIDA
D. L. vonBurger
The Coleman site is a circular hammock, 100 feet in diameter, located
in the S.W. 1/4 of the S.E. 1/4 of Section 26, Township 53 S. Range 39 E.
Dade County, Florida. It sits 250 yards east of the site known as "Cheetums
Hammock" (Laxson 1962). The Coleman site was, at the time of our investi-
gation in 1969, in danger of destruction and before our second visit had been
cleared off to 6 inches below the original surface level. Vegetation consisted
of Ficas Aurea, red bay, Australian pine, wild papaya, elderberry, and Bra-
The midden consisted of 12 inches of black loam and mixed grey sand
followed by an underlying zone, 12 to 14 inches thick, of hard calcareous ma-
terial containing animal bones, fish bones, and minute particles of charcoal.
A third layer, much softer and with very little bone refuse, extended down-
ward to a total depth of 54 inches. At the 54 inch level a deposit of Miami
Oolite was encountered which terminated excavation.
In October 1969 tests were begun at the site by means of a north to south
trench, 5-foot wide. Due to the start of construction in this area, excavations
were terminated in December of the same year at which time the trench was
10 feet in length.
Besides the usual pottery types typical to this area, a number of sherds
were recovered that suggested possible trade from cultural areas to the north.
They are listed in Table 1 by excavation levels. Recovered from the 36- to 40-
inch level was an almost complete St. Johns plain bowl. Below the 40-inch
level, and almost on the basil Oolite deposit, were found two Strombus gigus
shell celts in an almost perfect state of preservation. Also, in the same gen-
eral area were two hollow bone points. Other non-ceramic traits, including
glass beads, are listed in Table 2.
The Coleman site is a small habitation midden which was occupied from
very early times (Glades Ia) through the beginning of the historical period. The
early occupation isborneout by the finding of St. Johns Incised sherds (750-500
B. C.) within the upper two inches of the hardpan level and the St. Johns plain
bowl below the hardpan or calcareous layer. This find is paralleled by the
finding by the Broward County Chapter of the Florida Anthropological Society
Florida Anthropoligist, vol. 25, no. 3, September 1972
ARTIFACT COUNT BY LEVELS
Inches below surface
Sherds 0-6 6-12 12-24 24-48 Totals
Spanish olive jar 3 3
Glades Plain 98 15 113
Glades Tooled 9 9
St. Johns Check Stamped 1 1
Belle Glade Plain 2 2
Little Manatee Zoned St'd 1 1
Opa Locka Incised 5 5
Miami Incised 1 1
Matecumbe Incised 2 2
Sanibel-like Punctated 5 5
Ft. Drum Incised 1 1
Ft. Drum Punctated 1 1
Deptford Simple Stamped 1 1
Deptford Linear Check St'd 1 1
St. Johns Incised 2 2
St. Johns Plain la 1
Other Artifacts Table 2
Seminole beads (Brickell) 7 7
Yenitian glass beads 4 4
Hollow bone points 2 1 2 5
Socketed bone points 1 1
Bone awl 1 1
Perforated shark's teeth 2 1 3
Perforated alligator tooth 1 1
Worked bone 2 2
Worked shell 1 1
Columella drill 1 1
Strombus celts 1 2 3
Busycon picks 1 1 2
Chert projectile point 1 1
Sk 1 m n
Fig. 1. Decorated sherds and other artifacts, Coleman site.
a-b, St. Johns Incised; c, Little Manatee Zoned Stamped; d,
Ft. Drum Punctated; e, unique, Sanibel-like Punctated; f,
Miami Incised; g-h, Glades Tooled; i-k, hollow bone points;
1, alligator tooth; m, chert point; n, grooved shark's tooth.
at the Peace Camp site in south Broward County of semi-fiber-tempered
(Norwood) sherds and shell celts below a similar hard layer.
This calcareous layer has been noted on sites in south-east Florida lo-
cated west of the coastal ridge, but does not show up in those sites on or east
of it. Evidence from the Coleman site would suggest that some time before
750-500 B. C. much of the area was covered by water for a period of time
sufficiently long enough for the midden material to silt and calcify into the
grey secondary layer. Sometime after 500-750 B. C. the site was again oc-
cupied, as evidenced by the St. Johns Incised sherds.
Occupation continued through Glades II times (Opa-Locka Incised and
Matecumbe Incised), and well into the Glades III period (Glades Tooled
sherds). The olive jar fragments and Spanish trade beads indicate occupation
after the coming of De Soto, while the "Brickell type" beads document at least
occupational use of the site by Seminoles.
Appreciation is due Troup Bros. Inc. for their permission to conduct
tests on the site, and Miami Oolite, Inc. for the use of their access road.
Thanks is also expressed to Ripley P. Bullen, William H. Sears, and Dan. D.
Laxson, for their help in correlating information gathered. Also, thanks is
extended to the members of The Miami/West-India Archaeological Society
who conducted the field work.
Laxson, D. D.
1962 "Excavations In Dade and Broward Counties, 1959-1961.
The Florida Anthropologist, Vol. 15, No. 1, Gainesville.
December 14, 1969
THE DUDA RANCH SHELL MOUND, BREVARD COUNTY, FLORIDA
Charles F. Knoderer
The habitation site of Archaic man and his successors covered in this
report was designated by Rouse as BR-18. On a much earlier map made by
Clarence B. Moore, dated August 1894, it was listed as Turtle Mound. As
"Turtle Mound" has been used to designate a well known midden mound and
State Memorial southeast of New Smyrna Beach, as shown on many automobile
road maps, we will call BR-18 the Duda Ranch Mound.
A. Duda and Sons, Inc. ranch is located in Section 21, Township 26
south, Range 35 east, on the east side of the St. Johns River in Brevard County,
Florida. The Duda Ranch mound, the largest of several on ranch land, is lo-
cated about 300 yards northeast of the present river channel and about half way
between Lake Winder and Lake Washington. It measures 320 feet on a north-
south and 205 feet on an east-west axis. The top is 13 feet above mean water
level. Composed mainly of mussel and snail shells, it closely resembles other
Archaic shell middens on the St. Johns River.
As mentioned above, the Duda Ranch mound is near but not on the present
St. Johns River channel. The river is shifting to the west in this area. This
shifting of the river is very evident upon examination from the air, as it is ob-
vious that numerous old channels may be seen to the east of the present main
channel. Further substantiation of this shift was observable in the bottom of
Test A. At 6. 25 feet we penetrated the habitation strata and entered river sand.
Water eroded sherds (two were Orange fiber-tempered) and bits of bone were
found in the next 9 inches below this, sterile gray sandy clay. The amount of
erosion of these sherds is indicative of running water, namely, the ancient river
channel. A drain ditch northeast of the mound exposes Melbourne-Anastasia de-
posits. This unconformity appears to be about 2 feet below the base of the mound.
A large shallow, and un-natural, basin shows clearly in the marsh just a
few feet west of the mound. Marsh loam matches the surface loam on the mound
proper. This basin may be (1) a remnant of the river channel present when occu-
pation started, (2) the borrow pit for dirt used to increase the elevation of the
mound or to make the midden more salubrious for habitation, or (3) the borrow
pit for the fill for the superficial mound removed by Moore in 1894.
Twenty-three small test holes indicated the main structure of the Duda
Ranch mound to be relatively untouched. However, a thickening of surface black
loam was noted as the test holes neared the central west side. Some check stam-
ped sherds and scattered human bone fragments were found in this surface loam.
Florida Anthropologist, vol. 25, no. 3, September 1972
T-^r ifi ^'SyJ/ 79 A
I j C-
%P41 T K
-6' 7___ ~1
71 PLiTi'~ K
-- 1 F
^ [ ^\J^
mound locating tests
---and other features.
/ t ..
Fig. 1. Pln oD.u Id
a oth. er. feat ur e
; -L-t j)
II 0 n--
I I I I I i i I I I
Early residents of the area were questioned and all recalled that the western
portion of the mound was much higher in the 1930' s than at present.
Figure 1 is a plan of the Duda Ranch mound locating Tests A and B. Also
indicated is what appears to be the base of a superimposed ceremonial or rec-
tangular temple mound. The shape (Fig. 1) does suggest a temple mound with a
ramp leading southeasterly. However, scraps of human bones found in the high-
est loam indicate it was undoubtedly a sand burial mound. Dug by Moore in 1894,
it accounts for the thicker surface loam noted for the western part of the present
Neither Test A nor B showed any suspicion of disturbance under the loam.
The vertical distribution of pottery is given in Table 1 for both tests except that
the upper part of Test B has been omitted. The ceramic picture there did not
differ significantly from that for Test A.
Pit A was located on the south gradual slope of the main mound. It was
staked out 5 by 5 feet and was screened in one-foot levels. All bone, charcoal,
artifacts, and sherds were segregated by excavation levels. Quantities of river
mussel and snail shells were measured and proportioned as shown in Table 1
for Test A.
As mentioned earlier Pit A bottomed at 6. 5 feet. Due to ground water the
6-to 7-foot level was cut to 2. 5 by 2. 5 feet. Thereupon we started Test B, a 4-
by 4-foot excavation high on the mound and some distance away from ancient river
channel found at base of Test A. Test B was practically a duplicate of A for the
first 6 feet in sherd sequence and artifacts. The bone and shell tools were in
far better condition in Test B, probably due to location above all but the highest
ground water. Test B bottomed in sterile, gray, sandy clay at 11 feet. It was 4
by 4 feet to a depth of 6. 5 feet, then 2 by 3 feet to 10 feet, and only 1 by 1 foot in-
to the sterile sandy clay at 11 feet. This gave us two steps to work from. The
last two feet were excavated by use of a bucket and rope line to the surface.
A Columbia projectile point (Fig. 2) was found in the exploratory trench.
shown in Figure 1 at a depth of 2. 5 feet. Specimens from Tests A and B in addi-
tion to those listed in Table 1 include: a Westo-like point (Fig. 2) at 5. 5 feet and
a Sarasota point (Fig. 2) at 6. 5 feet, both in Test B, a deer antler handle (Fig. 2)
associated with the Sarasota point in Test B, 1 shell and 2 sherd pendants (Fig. 2)
at 3. 5 feet, 6 Busycon hammer (1 at 3. 5, the others at 5 feet), 3 sherd bones, a
deer bone fish hook (Fig. 2), at 2. 5 feet in Pit A, 3 shark teeth scrapers (Fig. 2),
various bone pins, awls, simple bone points, chisel, and spatulate tools (Fig. 2),
several hollow bone points (Fig. 2), a bushel of food bones from bear to bird, and
15 perfectly preserved but semi-fossilized examples of human feces (4. 5 to 9 feet.
The vertical pottery distribution in Table 1 is interesting. It clearly shows
DUDA RANCH SHELL MOUND
afq18 Dud a A._ ond s. Ile.
PEeFORA TD for /AIT/MG
COLLUBIA ZtOOAD wrsEJaT tZOO
PAOJECT/cLE Po/ir (fCferf)
(PUNCEw s -AWLS- PO4JECTL E ,WIYV)
Fig. 1. Various specimens from Duda Ranch mound.
the St. Johns II marker, St. Johns Check Stamped, at shallow depths over the
undecorated St. Johns I zone. The hole-tempered sherds were originally prob-
ably limestone tempered and indicate trade from north of Tampa Bay. At lower
depths we note the Florida Transitional period with thick St. Johns Plain, Nor-
wood (fiber and sand temper) Plain, and the Orange Series. This documents
occupation to about 1200 B.C. Two eroded fiber-tempered sherds suggest the
earlier Orange period.
We wish to thank the owners of the A. Duda and Sons Ranch for permitting
this research. We also are in debt to students of Anthropology at Brevard Junior
College and their professor, A. S. Dooley, Jr., who furnished much of the labor
expended on this project.
The quantity and size of both river mussels and snail shells in the refuse
VERTICAL DISTRIBUTION OF POTTERY AT DUDA MOUND
Test A Test B
Levels and 0-1 1-2 2-3 3-4 4-555-6 6-7 I 6.5-9*9-10 10-11 11-12
areas of tests 5x5 5x5 5x5 5x5 5x5 5x5 2.5x2.5 2x3 2x3 lxl
Check St'd 60 2
decorated 7 3
Scored 9 6 7 3 3
Plain 492 173 457 272 259 284 24 33 62
Perico-Pasco u t
Plain 4 2 d c~
scratched after ok 0
firing 5 U
Plain 1 4 2
Plain 2 3 3 u
% mussel vs. 80 m 30 m 40m 50 m 60m 50m 50 m
snail shells 20 s 70 s 60 s 50 s 40 s 50 s 50 s
Total sherds 568 184 464 279 269 284 27 41 67
*Sherds from 0-6.5 levels Test B are similar in distribution and type to
those found in Test A. All dimensions are in feet.
was noted by levels in Test A. We feel that they show an indication of changes
in climate and river level. The quantity and size of the mussels dropped rapidly
in the 1-to 2-foot level (beginning of St. Johns II times) indicating a low stage of
the river. This condition would affect river mussels more than the freshwater
snails which-could survive in stagnant water holes. This level, correspondingly,
reflects a great drop in numbers of sherds. Thin tree rings observed in char-
coal found in this 1-to 2-foot level are further indications of drought.
DUDA RANCH SHELL MOUND
The succeeding deeper levels indicate an approach to conditions similar to
the highest level until the 7-foot level in Test B where both size and density of
mussel shells decreased gradually. Both mussel and snail shells found in the
pre-ceramic level were extremely small (1/5 size of those found in upper levels).
This is understandable as the time sequence here falls in the latter part of the
Melbourne-Van Valkenburg interval when the ocean level was lower than at pre-
sent thus increasing the gradient of drainage of the St. Johns river and thereby
dropping the river level. This is in agreement with the finding of old fire pits
by H. Bruce Green well below low water level at the north end of Lake Washing-
ton in 1964.
The white to gray sandy clay found underlyingthe Duda mound,upon examina-
tion with a 10 power glass, contains quartzitic and shell sand as well as small
bits of sea shells and fossil spicules of sponges indicative of river erosion simi-
lar to the Melbourne-Van Valkenbury unconformity.
The above mentioned micro-fossil spicules were first noted in many sherds
overlying the Orange series and some in a few Orange sherds. All sherds were
then re-checked and micro-spicules were also found in the paste of the check
stamped sherds in the 1-to 2-foot level of Test A, in the later check stamped sherds
of the 0-to 1-foot level. These later people also used a source of clay containing
these fossils. However, the inhabitants represented by the interval from 1 1/2
to 5 feet in depth did not use this spicule-bearing sandy clay for pottery material.
Upon checking Test B sherds, the same situation vertically was found. These
spicules are about 1/3 of a millimeter long, about 1/5 the thickness of human
hair and pointed at both ends. Consisting of lime and silica they do not burn
away when the pottery is fired as do plant fibers in Orange fiber-tempered ware.
The greatest concentrations of bone and shell tools were found in the 3- to
6-foot section in both tests. The deepest bone awl was found at 9 feet in
in Test B and appears to be partially mineralized. The perfect condition of
these artifacts, as well as that of the human excrement, is due to their lo-
cation above the flood level of the river plus infiltration of calcium and lime
carbondates absorbed by rain water filtering down thru the huge deposits of
mussel and snail shells.
The sherd collection indicates a rather complete sequence from Orange
Plain thru Orange Incised, Transitional, and the St. Johns Plain and Check
Stamped series. A few Belle Galde Plain sherds were also present. Artifacts
and strata in tests indicate a steady flow of migrant people with thin sterile
layers pointing out periods of non-habitation while nature restored depleted
game and fish resources. To summarize people lived intermittently at the
large Duda Ranch mound for well over 2000 years.
SOME ARTIFACTS FROM THE TRAIL SITE, DADE COUNTY, FLORIDA
Wesley F. Coleman
The artifacts illustrated in Figures 1-3 were recovered by William Car-
son of the Dade County Chapter of the Florida Anthropological Society at the
well known Trail site west of Miami (our reference T-R-4). Appreciation is
hereby expressed to James Shaffer for the excellent pictures.
The first (Fig. 1) is a bone adzee" handle similar to several that have
been found in Florida (Bullen 1953:15). They are well known but are not ex-
actly common. It has a rounded hole in one end and a slotted one in the other
end. The handle of the adzee" was probably made of wood with a short bent
lower portion which was fitted into the larger hole. A stone or shell tool was
inserted into the slot to form the working end. Carson's handle was found at
a depth of 13 inches below the surface and probably refers to a Glades II time
Figure 2 illustrates an odd-shaped clay pendant that our research over
the past year has led us to believe is unique. Found at a depth of 6 inches in
association with Glades Tooled rim sherds, it must be Glades III in date. It
is made of a Glades gritty paste and has two holes', one at each end. One pre-
sumedly was for suspension by a cord and the other, perhaps, for the insertion
of a feather. Most pendants found in Florida have a groove near the top for
suspension but a few, usually of shell, have holes.
In Figure 3, a, we illustrate a stage in the manufacture of a bone awl or
hair pin from the foreleg of a deer. The pin is nearly complete. All that remains
to be done is to remove it from the blank and to grind or polish the end and sides.
For comparison a plain pin is depicted in Figure 3, b.
Bullen, Ripley P.
1953 The Famous Crystal River Site. Florida Anthropologist,
vol. 6, no. 1, pp. 9-37. Tallahassee.
Elder, Robert A.
1971 Personal Communication. Museum specialist,
Flprida Anthropologist, vol. 25, no. 3, September 1972
ARTIFACTS TRAIL SITE
Fig. 1. Bone adze handle.
Fig. 2. Clay pendant.
of manufacture and b, finished.
The purpose of this paper is to present another use for certain aborig-
inal objects usually referred to as metates and duhos or stools. The first part
of the paper will briefly describe metates and duhos. The second part will de-
scribe two stools from Panama including their utility and physical attributes.
The term metate, derived from Nahuatl metatl, is normally applied to a
flat or concave stone on which maize is ground by hand with a stone muller or
mano (Mason 1945:217). The maize is prepared by putting the desired amount
into boiling water, into which a small amount of lime has been added in order
to remove the husk. Once it has been cooled and rinsed, the maize is ground
on the metate and kneaded into a dough called massa. The dough is then cooked
in pots, wrapped in leaves and steamed with meat or other items, or made into
unleavened cakes called tortillas (Joyce 1916:212). Examples of some metates
can be seen in Figures 6 and 7.
Duhos (Fig. 5) were probably used in connection with religious or cere-
monial events and apparently reserved for the seating of the caciques and hon-
ored guests (Loven 1935:507; Granberry 1955:256; Sauer 1966:60). Hostos and
im Thurn suggest that these stools were also used in areas infected with "jig-
gers", in order to raise the person out of the reaches of the insects. These
"jiggers" or "niguas" are the Pulex penetrans- an insect which burrows into
soft tissue and lays its eggs causing swellings in the limbs and in many cases
complete disability of them (Hostos 1923:252) (im Thurn 1883:297). Hostos men-
tions a further division of the duho where the size of the duho considerably af-
fects its use: the larger duhos being given utilitarian uses for the seating of
men while the smaller miniature ones were idol stands, altars or ceremonial
platters (Hostos 1941:79). They may also have served as head rests while
Duhos are carved from one piece of wood thus eliminating the need for
nails or glue. These stools generally have an anthropomorphic or zoomorphic
head on one end and a tail on the other, but cases of the plain platter type have
been recorded (Hostos 1941:83; Portuondo 1950:48).
Florida Anthropologist, vol. 25, no. 3, September 1972
Figs. 1-2. Side and end
view, two legged stool.
Two Panamanian Stools
The information on these stools was given to me by Dr. Louis Paganini
of the Geography Department of the University of Florida. He was present
during their acquisition while collecting data for his doctorate in Panama. The
function given here is based on Paganini' s observation of their use while he
was in that country.
The first stool(Figs. 1 & 2), a two-legged one, was especially made for
Dr. Paganini as a gift by an Indian who became a good friend of his, one of the
Choco Indians on the Pirre River, near El Real in the Darien province of Panama.
The second stool (Figs. 3 & 4), four-legged with a turtle head protruding
from the seat, was a gift to Dr. Raymond Crist of the Geography Department of
the University of Florida. Crist was in Panama with Paganini and received the
gift from a Yape Indian when he made the mistake of admiring it. This group
of Indians is located on the Tuira River in the Darien Province near Colombia.
These Panamanian duhos or stools are used primarily as seats in dugout
canoes. These canoes, which are used for many other purposes besides trans-
portation of people, cannot have permanent benches or seats because the cargo
and the number of passengers varies from day to day. The use of these stools
also makes it possible for people to easily get on the precarious craft without
having to step over permanent seats and risk losing their balance and turning the
canoe over. By carrying their own stool, they can step into the boat, walk to
where they are going to sit, and place the seat under them.
When they are not being used in the canoes, the Indians always take them
into their homes to function as stools there, for fear of losing them or having
them stolen. Dr. Crist also mentions the fact that these stools were sometimes
used as head rests or pillows when the Indians slept.
Description of the two-legged stool
This first stool (Figs. 1 & 2) is made out of a wood commonly called "Ca-
tivo", or Prioria copaifera, that closely resembles balsa wood. It is chopped
out of one piece of wood by a machete. This particular stool (7 inches high, 17
inches long, 9 1/4 inches wide and approximately 1 inch deep in the seat) was
finished by smoothing the rough parts down with small knives and later rubbing
it down with sand drom the river. Usually such finishing procedures are not
used and the stools remain roughly hewn. As it was a present, the Indian made
a special effort to do a neat job in this case.
2 PANAMANIAN DUHOS
''^ -^^ ^^^^^^^^^^
Figs. 3-4. Side and end view, four legged stool.
Fig. 5. Bahamian Duhos
Fig. 6. Oval metate
from Panama (courtesy
Fig. 7. Duho from
Puerto Rico (courtesy
Fig. 8. Duho from Cuba
Description of the four legged stool
This second stool (Figs. 3 & 4), is made of a very heavy and quite dur-
able wood commonly called "Pino Amarillo. Shaped by a machete from one
piece of wood, this particular stool, 13 1/2 inches long, 7 inches wide and 3
1/2 inches high was not further smoothed down with knives or sand. It has a
turtle head at one end. Dr. Paganini mentioned that these stools can occur in
a much more elaborate state with heads more finely carved and tails which
serve as handles when they are being carried from one place to the next.
I would like to point out an interesting observation which was made while
studying the duhos and metates. This was the startling similarity between the
two forms. The two-legged stool (Figs. 1-2) closely resembles the oval metate
(Fig. 6), and the four-legged stool in turn looks like Bahamian and Puerto Rican
duhos (Figs. 5, 7) and particularly one (Fig. 8) from Cuba.
This also raised the question of the precision in language which occurred
when discussing the use of the terms metate (as corn grinders) and duhos (as
seats). Peralta and Alfaro (1893:31) expressed concern about this as far back
One cannot be totally sure that all the stone speci-
mens, found during the recent excavations, which
represent four legged animals such as turtles, are
truly metates, because the corresponding manos
have not always been found in conjunction with them,
and many of the plate surfaces show no sign of wear,
which would indubitably occur if two rocks were
rubbed together for a period of time. (Translated
by the author of this paper).
A more recent writer, Mason (1945: 220-22) also briefly noted this problem
when he was doing archaeological work in Costa Rica.
In conclusion I would like to raise the question as to whether all the arti-
cles listed presently as metates are in reality corn grinders, instead of duhos
or seats. Also, it seems very probable the larger wooden stools from Key Mar-
co, Florida (Cushing 1896: Pl. 34, I) functioned as portable canoe seats as sug-
gested by Cushing in 1896 (p. 363). Except for the lack of a head the Key Marco
example closely resembles the 4 legged one illustrated here in Figure 3.
Cushing, Frank, H.
1896 Exploration of ancient key dwellers' remains on the gulf coast
of Florida. Proceedings of the American Philosophical Society,
vol. 35, no. 153, pp. 329-432. Philadelphia.
1955 A Survey of Bahamian Archaeology. Thesis presented to the Univ.
of Fla. for the partial completion of the degree of Master of Arts.
Hostos, Adolfo de
1923 Anthropomorphic Carvings from the Greater Antilles.
American Anthropologist, Vol. 25, no. 4, Oct. -Dec.
1941 Anthropological Papers. Office of the Historian.
San Juan, Puerto Rico.
im Thurn, Everard
1883 Among the Indians of Guiana.
London: Kegan Paul, Trench and
Joyce, Thomas A.
1916 Central American and West Indian Archaeology, London:
Philip Lee Warner
Lines, Jorge A.
1964 Costa Rica: Land of Exciting Archaeology, San Jose: Instituto
Costarricense de Turismo.
1935 Origins of the Tainan Culture, West Indies. Gothenfurg:
Elanders, Bokfryckeri Akfiebolag.
Mason, J. Alden
1945 Costa Rican Stonework: The Minor C. Keith Collection.
Anthropological Papers of the Amer. Museum of Natural History,
Vol. 39, part 3, pps. 193-317.
Peralta & Alfaro
1893 Catalogo Razonado de los Objetos Arqueologicos de La Republica
de Costa Rica. Madrid.
1950 Historea de Cuba. La Habana: Editorial Obispo.
AN EARLY HISTORICAL PERIOD CANOE
The Thornhill Lake District of Volusia County is well known for a large
mound complex located on its northern shore (Fig. 1). A visit by Clarence
B. Moore in 1894 disclosed artifacts which suggested extensive trade with ma-
terial from possibly as far as the Great Lakes. Of note are polished stone
artifacts resembling double-headed axe blades over the chests of burials.
Archaeological data, and information studied, suggest that this site was aban-
doned by about 1200 A. D. However a site located on the south edge of the
slough and beside the St. Johns River (Fig. 1) began at about 900-1200 A. D.
and continued until historic times.
The canoe depicted in Figure Z was discovered in Thornhill Lake slough
in two feet of water and three feet of mud. Mode of construction would suggest
its manufacture between 1550 and 1763 A.D. Though the specimen is Timu-
cuan in design and in the same style as earlier dated specimens from Lake
Apopka (Bullen and Brooks 1967), it shows distinct marks of an iron axe (Fig. 3).
This would date its manufacture as after 1550, but its old style design puts it
before the depletion of the Tamucuan Culture by 1763 as it is not Seminole sty-
listically. No explanation of the holes drilled along, its upper edge is possible
at this time.
The specimen measures 19 feet 7 1/2 inches in its present length, 30
inches at its widest point, and 17 inches in depth. It was hollowed out by fir-
ing (charring) and chipping, and a portion of the exterior also exhibited firing.
This canoe, which appears to be made of cedar, is at present on display at the
Museum of the Apopkans.
Bullen, Ripley P. and Brooks, H.K.
1967 "Two ancient Florida dugout canoes", Quarterly Journal of the
Florida Academy of Sciences, Vol. 30, No. 2, pp. 97-107,
Florida Anthropologist, vol. 25, no.3, September 1972
HISTORICAL PERIOD CANOE
i, - inoa.
* SHELL r0D.
X 54,p rno.
4 -- -
Fig. 3. View of mid portion of side.
TEN BURIALS FROM GREEN LAKE, TEXAS
R. J. Wingate and Thomas Roy Hester
This brief paper presents data obtained from site 41 CL 13 on Green
lake, Calhoun County, Texas (Fig. 1). This central Texas coast site was in-
vestigated by the senior author in 1960. At that time a number of burials were
exposed and removed. The work was initiated after part of the site was re-
moved by the Texas Highway Department for road fill. A portion of the burial
area at the site was destroyed by these land removal activities.
Green Lake (Fig. 1) is located on the central Texas coast, near the
mouth of the Guadalupe River and just north of San Antonio Bay. The lake con-
tains potable, although somewhat brackish, water; it extends four miles north
to south, and is four and one-quarter miles wide. The lake lies within the
floodplain of the Guadalupe River. This area of low relief is cut by creeks,
sloughs and marshes. It seems likely that the lake was once an extension of
San Antonio Bay, but was cut off from the present bay by the building of the
Guadalupe River delta. The area lies within the Tamaulipan Biotic Province
(Blair 1950: 102-105). C.A. Calhoun (personal communication) has recorded
the following vegetation along the lake and nearby waterways: cypress, water
ash, willow, sea cane, and palmetto. On the surrounding upland prairies are
mesquite, huisache, yucca, prickly pear cactus and motts of live oak. Wild-
life in the vicinity of the lake include alligators, turtles, shore birds, migra-
tory waterfowl, deer, rabbits, squirrels, opossum, raccoons, foxes, coyotes,
rodents of various species, peccary (javelina), and snakes. Annual rainfall is
36. 83 inches, January minimum temperature is 470, while maximum tempera-
tures of 920 occur in July (Unknown 1969:251).
The site (designated as 41 CL 13 by the Texas Archeological Research
Laboratory, Austin) lies along a 20 to 25 foot bluff on the southeastern edge of
Green Lake (Fig. 1). This bluff extends for about three-fourths of a mile
(roughly north-south) and parallels the lake shore. A burial area (Fig. 2) is
present near the central part of the bluff, and it was this portion of the site
which was investigated by Wingate in 1960. However, archaeological debris is
widely distributed along the bluff surface and it is impossible to define the ex-
act limits of the site. C.A. Calhoun has recorded site 41 CL 1 (described in a
later section of this paper) at the eastern end of the bluff. At present, we can-
not determine if 41 CL 13 and 41 CL 1 are separate sites, or whether they are
parts of a single site.
Florida Anthropologist, vol. 25, no. 3, September 1972
BURIALS FROM GREEN LAKE
LOCATION OF 41CL13
Fig. 1. Site location map.
Edge of Guadalupe River floodplain indicated by dashed lines.
WINGATE AND HESTER
The surface of the site is covered with grasses, prickly pear cactus and
scrubby thorn brush (mesquite, huisache and others). Larger trees, includ-
ing live oak, are present behind the bluff on the edge of the prairie.
A burial area, or cemetery, was partially excavated by Wingate. Bur-
ials were found at depths of 26 to 28 inches below the present surface, and had
been intruded into midden deposits. The midden soil is brownish in color, and
contains numerous shells of Rangia sp. (a brackish water clam) and common
oyster (Crassostrea virginica). Also present in the midden are occasional
flint flakes and Rockport ware sherds. A plan of the burial area is shown in
The burials are briefly described below, based on notes and observations
provided by Wingate:
Burial 1 (Fig. 2). Adult. On left side, with skull directed north and fac-
ing east. Arms flexed, with hands to face. Lower portions of skeleton missing,
probably due to post-burial disturbances.
Burials 2 and 3 (Fig. 2). Adults. Burial 2, tightly-flexed on right side
with the skull directed north and facing west, faces Burial 3, also a tightly-flex-
ed individual on left side with skull directed north and facing east. Hands of
both individuals were placed at the face. The two burials are only about one
foot apart. The proximity of the two skeletons, as well as the similar burial
positions, suggest that they were interred at the same time.
Burial 4 (Fig. 2). Adult. Appears to have been disturbed. Skull of this
individual directed east and facing up. One of the legs placed along side of and
parallel to the torso, with the foot resting against the right side of the face; the
other leg flexed. Disposition of arms uncertain. A large piece of asphaltum.
appeared to be associated with the burial, though it is possible that it was a part
of the surrounding midden fill.
Burial 5 (Fig. 2). Adult. Tightly flexed on left side, with skull directed
south and facing west. Hands to face.
Burial 6 (Figs. 2, 3). Adult. Immediately east of Burial 5, and in a
similar burial position (tightly flexed on left side; skull directed south and facing
west; hands placed near face). A cluster of Rangia shells near the forehead may
have been a burial offering as many were unopened. Burial was evidently placed
on a layer of Rangia shells. Resting within the fold of the right arm was a small
broken biface (Figs. 3, 4). About three inches from the forehead was a very
large oyster shell, within which was a smaller oyster with modified edges. Near
BURIALS FROM GREEN LAKE
the base of the skull were three other pieces of burial furniture: (1) a piece of
baked clay; (2) a ball of soft white clay; (3) a small incised bone bead (Figs.
3, 4). Decoration on the bead consists of two parallel incised lines; the bone
is that of either a small mammal or bird.
The grave fill for Burial 6 was distinct from the surrounding midden de-
posits, and Wingate believes that part of it was sand obtained from the nearby
Burial 7 (Figs. 2, 3). Adult. Flexed, in an upright sitting position
(possibly a bundle burial). In the area of the mouth were two pieces of modified
deer bone decorated with incised lines (Fig. 4, A). Among the ribs of this in-
dividual, a stemmed dart point (Fig. 4, B) was found. This artifact, reworked
along the lateral edges long after its original manufacture, exhibits heavy pa-
tina on its unreworked portions.
Burial 8 (Figs. 2, 3). Juvenile. Very poorly preserved; the head was
directed south and facing east. The individual was flexed on its back.
Burial 9 (Figs. 2, 3). Adult. Also poorly preserved. Probably was on
right side, with head directed south and facing west. A carved round stone disc
was associated (Fig. 4, E).
Burial 10 (Fig. 2). Adult. Located during excavations, but not fully ex-
posed. Flexed on right side, with skull directed west and facing south.
Summary and Discussion
Ten burials have been excavated from a restricted burial area or cemetery
at site 41 CL 13, central Texas coast. Other burials were destroyed by land
removing activities during the construction of a nearby highway.
The intermittent nature of interments at the site is suggested by the variety
in burial position and orientation. Burials 2 and 3 represent individuals in-
terred at the same time. Burials 5-9 are oriented with the skull to the south,
and with the exception of Burial 7, all are flexed. Grave goods are definitely
associated with three burials, and possibly with a fourth. Some of the burials
have the hands at or near the face, a recurrent trait along the central and south-
ern coast (Hester 1968).
The burials had been placed in a midden deposit which contained sherds of
Rockport pottery. Sandy-paste Rockport ware is the predominant pottery type
on the central and lower Texas coast, originating in late prehistoric times and
surviving into the historic period (Suhm and Jelks 1962: 66-68; Campbell 1962:
331-336). The burials were made either during a late stage of midden accumu-
WINGATE AND HESTER
........... ......... .... ... .. .. .
. . . . .. .. .. . ... .
I4 1 I1: I :: :1~~
Fig. 2. Burial plan (not to scale).
lation, or perhaps during subsequent brief occupations. In either event, the
burials can be assigned to the late prehistoric or protohistoric eras (there is
no evidence of historic contact at the site). We should point out that Karankawa
Indian groups were known to inhabit this region during the historic period (Gats-
chet 1891). Though these burials might be attributable to prehistoric or proto-
historic representatives of this tribal group, we currently have no data to sup-
port such a link.
None of the artifacts associated with the burials are temporally diagnostic,
based on our present evidence. Dart points such as the one shown in Figure 4,
BURIALS FROM GREEN LAKE
Fig. 3. Plan view Burials 6 9.
Sketches of Burials 7-9 made from photographs taken before
burials had been completely exposed.
O R shell
00 Ranaia shells
WINGATE AND HESTER
Fig. 4. Artifacts from burial area.
A, incised bone object; Burial 7; B, stemmed dart point, Burial
7; C, chipped biface, Burial 6; D, incised bone bead, Burial 6;
E, carved stone disc, Burial 9.
BURIALS FROM GREEN LAKE
are supposedly linked to preceramic occupations on the coast. But as we noted
above, this specimen had been reworked suggesting that it had been "salvaged"
from an earlier context and re-used. Incised deer bone artifacts are known
from other burial sites on the Texas coast (Hester 1968). The specimen from 41
CL 13 (Fig. 4) has closest parallels with incised bone grave goods from the Al-
bert George site on the upper coast (Walley 1955: Pl. 36).
Almost nothing is known of the archaeology of this part of the Texas coast.
C.A. Calhoun (personal communication) has investigated site 41 CL 1 (Traylor
Ranch site) about 1000 yards from 41 CL 13. This site has an 18-inch thick
midden accumulation of Rangia sp., oyster, and other marine shells, and land
snails. Calhoun also noted the presence of baked clay lumps, asphaltum, sherds
of Rockport ware, bone awls, milling stone fragments, flake scrapers, seven
Scallorn arrow points (Suhm, Krieger and Jelks 1954: 506), and a stemmed dart
point. Calhoun excavated a burial at the site and noted fragmentary remains of
two others. Calhoun has observed that archaeological materials are scattered
all along the northern, western and eastern margins of Green Lake. The data
from 41 CL 1 and 41 CL 13 suggest that there are probably a number of scatter-
ed burial areas along the lakeshore. Indian burials have also been noted on
nearby Hynes Bay, but are not described (Bracht 1879:442).
There are a number of cemetery sites present along the Texas coast
(Collins, Hester and Weir, 1969; Hester 1969). On the central and lower coast,
these take the form of specialized burial areas separated from occupation sites.
On the upper coast, evidence suggests that burial areas are usually within shell
middens (Campbell 1957; Hester1968). These sites are currently being investi-
gated in detail by Lawrence Aten of the Texas Archeological Salvage Project.
It is intriguing to note that the burial area at Green Lake is within midden de-
posits, as are the cemeteries on the upper coast. However, since burial prac-
tices along most of the Texas coast are so poorly known (or in many cases,
poorly published), we have little basis for further speculation.
Blair, W. Frank
1950 The biotic provinces of Texas. Texas Journal of Science,
Vol. 2: 93-166. San Marcos.
1879 Note. Annual Report, Smithsonian Institution: 442. Washington.
1957 Archeological investigations at the Caplen site, Galveston County,
Texas. Texas Journal of Science, Vol. 9, No. 4:448-471. Austin.
1962 Origins of pottery types from the coastal bend region of Texas.
Bulletin, Texas Archeological Society, Vol. 32 (for 1961):331-
WINGATE AND HESTER
Collins, M.B., T.R. Hester and F.A. Weir
1969 The Floyd Morris site (41 CF 2), a prehistoric cemetery site in
Cameron County, Texas. Bulletin, Texas Archeological Society,
Vol. 40: 119-146. Dallas.
Gatschet, Albert S.
1891 The Karankawa Indians, the coast people of Texas. Archaeo-
logical and Ethnological Papers of the Peabody Museum, Harvard
University. Vol. 1, No. 2. Cambridge.
Hester, Thomas Roy
1968 A study of the cemetery sites on Oso Creek (Nueces County) and
other cemetery sites on the Texas Gulf Coast. Paper presented
at the annual meeting of the Texas Archeological Society. Waco.
1969 The Floyd Morris and Ayala sites: a discussion of burial prac-
tices in the Rio Grande Valley and the Lower Texas Coast. Bulle-
tin, Texas Archeological Society, Vol. 40:157-166. Dallas.
Suhm, Dee Ann and Edward B. Jelks
1962 Handbook of Texas archeology: type descriptions. Texas Archeo-
logical Society and Texas Memorial Museum. Austin.
Suhm, Dee Ann, Alex D. Krieger, and E. B. Jelks
1954 An introductory handbook of Texas archeology. Bulletin,
Texas Archeological Society, Vol. 25. Austin.
1969 Texas Almanac. A. H. Belo Corporation. Dallas.
1955 A preliminary report on the Albert George site in Fort Bend Coun-
ty. Bulletin, Texas Archeological Society, Vol. 26:218-234. Abilene.
Department of Anthropology
University of California
NOTES ON THE MEADOWBROOK FARMS NO. 2 SITE
Ben I. Waller
The purpose of this brief paper is to record a significant site in northern
Marion County. Meadowbrook Farms No. 2 (now called Stavola Farms) is lo-
cated approximately 7 miles north of Ocala and east of U.S. 441 and 301.
The Meadowbrook Farms No. 2 Site was first called to the attention of
the author in October of 1967 by Miss Carol Smith. Braxton Jones and the au-
thor were given permission to "conduct any survey we like" at the site by the
farm manager Mr. Henry "Hank" Smith. Unfortunately, due to the nature of
the excavation, stratigraphic evidence from the site was not available. Con-
sequently, all artifacts recovered, were picked up on the surface. Although
much of the farm showed evidence of human occupation, this paper deals pri-
marily with the artifacts located around the pond which was being enlarged and
The pond itself was situated in the western half of the farm and was nearly
500 feet long by 175 feet wide. It was surrounded by white sand which made up
the bank. The bank had a good elevation with all sides being higher than the pre-
sent water level. On the northern and eastern sections, however, a 9- to 17-foot
elevation made that area especially desirable for habitation, and it is on that
prominence that most of the artifacts were recovered.
Many of the projectile types listed by Bullen (1968) in his guide to the
projectile points of Florida were present and most showed a preponderance of
pre-ceramic forms although at least 4 St. John Check Stamped sherds were
collected at the site.
Of special interest was a Suwannee point collected on the northwestern
bank and a broken paleo-Indian point base found on the north bank. A very fine
Greenbrier was collected 350 feet north of the pond some time later and
is in the collection of the author.
In summary, I feel this site was a long-time habitation site as the projec-
tile points recovered range from Paleo-Indian to the Pinellas points of proto-
historic Indians. The scarcity of pottery indicates largely pre-ceramic com-
plexes; however, a few Pinellas points and St. Johns Check Stamped sherds
demonstrate a late pre-Colombian reoccupation. Some 1830 glass fragments may
indicate a historic settlement although none is listed on available maps for this
Flroida Anthropologist, vol. 25, no. 3, September 1972
CONCRETIONS ASSOCIATED WITH GLADES PREHISTORIC SITES
Excavation of a prehistoric Indian site in the eastern part of the Ever-
glades often discloses a layer of hardened material below the usual soft mid-
den surface. This occurrence has been mentioned in various reports in the
literature but is not described in detail. For instance, Dan Laxson (1970:157)
tells of "hard gray marl conglomerate found at the bottom level of some of the
sites.... (consisting) of minute pieces of bone and shell, with some sherds. "
Perhaps the word concretionn" is not exactly appropriate, in that the cement-
ing agent does not deposit radially around minute centers, but the association
with the word "concrete" certainly is apt, as we who dig through this material
The best example of concretionn" to my knowledge is at the Peace Camp
mound, Br-52. Here a veritable dome of armor uniformly 16 to 18 inches
thick, covered the entire mound. Below it were two undisturbed habitation
levels. While other Glades sites have not shown the thickness, uniformity, and
hardness of the Peace Camp concretion, a surprising number have produced
similar hard layers, indicating that the factors producing concreting are often
present in the eastern Glades area. The Markham Park sites, now being dug
by the Broward County Chapter, are both furnished with hard layers, particu-
larly Mound I where burials are found within the hardened zone. Bishops Head,
Cagles Hammock, and Plantation Acres sites all have some areas of concretion.
At Peace Camp and elsewhere the surface layers of the middens, usually
containing most of the pottery, and dating back around 2000 years, sharply ter-
minates in a stoney layer. This layer may have a pebbly, light gray upper level
resembling old lime mortar. This is where easy digging stops and pick-work
Some sites have only an inch or so of hard crust, with a region of incipient
hardening below it. The hardening is obviously a process proceeding from the
top down. Usually a thick layer of more or less barren material is found below
the crust. However, when present, previous occupation levels below the barren
layer show up plainly with color and texture changes.
Since about 2000 years of occupational deposits cover most of these con-
cretions while Transitional period artifacts are found below them, as at Peace
Camp, the age of the concretions must lie between these times; that is, from
as early as 1200 B.C., to perhaps as late as 50 B. C. At Peace Camp, both
St. Johns Plain and Incised and Norwood Plain (with mixed fiber sand temper)
sherds were found in the concretion itself. This would indicate a Florida Tran-
sitional period age for the concretion. Our C-14 date of 3050 years (plus or
Florida Anthropologist, vol. 25, no. 3, September 1972
minus 140 yrs.) came from several inches below the concretion (Williams and
Mowers N. d.). Surely it is no coincidence that many Glades mounds show con-
creted layers of approximately the same age; i.e., 2000-3000 years B. P. We
may speculate why these layers occur, and why at this particular point in time.
Experience leads me to believe that our Glades midden mounds were
raised up above their previous heights at about the same time from necessity
by action of the inhabitants. Something compelled the Indians to raise the level
of their home-sites, and that something could only have been a general rise in
water-levels in the Glades. Whether the annual rain-fall increased or the ac-
cumulation of peat displaced the water upward is immaterial; the necessity of
facing the rising waters remained. There is a possibility that this water-level
change may be correlated with world-wide differences in temperature and rain-
fall or rise in sea level in the post-glacial period. This I cannot prove at present.
The source of fill for raising mound heights unquestionably was the sur-
rounding area, which became borrow-pits. Most Glades middens show rings of
lower elevation, still visable, around mound centers. Baskets of muck and sand
were carried from the borrow-pits and dumped where needed. This process
may well account for cases of reversed stratigraphy on sites of long occupancy;
for example, Archaic projectile points lying near the surface may be dumped
above later Florida Transitional period artifacts.
An interesting possibility arises here. In digging in the borrow-pits the
surface of bedrock may often have been reached. Above the rock surface fre-
quently is found a layer of clay and sand, the result of limerock weathering.
Such clay could be a source of ceramic raw material, requiring only removal of
excess sand. At every site we excavate, we routinely dig a hole in the center to
record depth to bedrock and frequently find this clayey deposit. Clay with sand
makes a hard durable material, even on air-drying. Mixed with muck and sand,
it would surely contribute to the stoney consistency of the future concretion.
However, we feel that rain-water percolating through the soft midden, and
picking up lime from shells, bones, and sometimes the limey clay just mentioned,
furnishes most of the cementing agent in the concretion layer. Chemical analyses
should be run on these concretions. Only lack of laboratory facilities prevents us
from securing an accurate knowledge of calcium percentages which would illu-
minate this problem. The whole story of concretions in midden mounds would
make a fine graduate thesis.
In addition to the marly, clayey concretions described above, there is
another variety of hardened substance found on a few mounds. This is what we
call an "organic concretion", because it is visibly stuffed with bony bits, shell
fragments, charcoal, and occasional artifacts. It is a true midden consolida-
tion, very dark in color, brittle and rough in fracture. Often it has a leached,
eroded appearance. This type of concretion is usually found some distance be-
low the gray, marly layer, and obviously indicates an earlier occupation zone;
the living debris of the raisers of the mound level. At Peace Camp, a still
earlier midden was cemented with reddish, iron bearing deposits. It is thought
that organisms living at water level concentrated iron in their structures which,
when dried, oxidized into red, iron basic oxides. They added their distinctive
colors and cementing properties to the midden deposit.
In conclusion, we find that concreted layers are much more prevalent
than was formerly thought and that several kinds of concretions are known by
their differing physical characteristics. We feel that further study should be
made by chemical means of this rather surprising phenomenon, heretofore
neglected in the literature.
Laxson, D. D.
1970 Seven sawgrass middens in Dade and Broward Counties, Florida.
Florida Anthropologist, vol. 23, no. 4, pp. 151-58. Gainesville.
Williams, Wilma B. and Bert Mowers
N. d. Excavation records, Peace Camp site, Broward Chapter,
Florida Anthropological Society.
AN ENGRAVING TOOL FROM NORTH FLORIDA
Ripley P. Bullen and Mary Teller Wallace
In the spring of 1972, Mrs. Morton H. Teller of Gainesville, Florida,
brought to the Florida State Museum a small collection of Archaic period chip-
ped-stone tools found by the junior author in 1954 at her grandmother' s home
on the eastern side of Santa Fe Lake in Bradford County, Florida. These
tools were found in shallow water immediately in front of the house. At that
time, 1954, the water in Santa Fe Lake was the lowest known to residents near
the lake. This probably explains the location of the finds because other speci-
mens have not been found at higher elevations in the same location.
While most of the specimens, including several large stemmed points
and flake knives, clearly belong in the Archaic period, four specimens stood
Florida Anthropologist, vol. 25, no. 3, September 1972
132 ENGRAVING TOOL
out as different. One was a Pinellas point which seemed out of context. The
other three are illustrated, full size, below.
The one on the left is a Dal-
ton point with ground base and on-
ly a suggestion of side notching.
Si The one in the middle exhibits
an engraving or incising point very
neatly chipped on the end of a blade
or rectangular-shaped flake. The
third is a crude, percussion chip-
ped, lunate scraper. It is chipped
? "all over and could be used as a scra-
per on almost any edge.
Both Dalton points and engraving tools are fairly rare in Florida. How-
ever, very similar engraving tools are illustrated for the Dalton zone at the
Stanfield-Worley bluff shelter in Alabama (Dejarnette, Kurjack, and Cambron
1962: Fig. 42). These authors (1962:85-87) present two radiocarbon dates for
this zone averaging 7320 B. C. We believe the same date would apply in Florida
to the Dalton point and engraving tool in Figure 1. The cresant-shaped scraping
tool is not included in the Stanfield-Worley Dalton complex and hence is probably
not of the same date.
DeJarnette, David L. Edward B. Kurjack, and James W. Cambron
1962 Stanfield-Worley Bluff Shelter. Journal of Alabama Archaeology,
vol. 8, nos. 1-2. University, Alabama.
Copies of Volumes I through XX may be purchased from the Treasurer
at $6.00 per volume or $1.50 for single and $3.00 for double numbers if avail-
able. Copies of Volumes XXI through XXV may be purchased from the Treas-
urer at $4.00 per volume or $1.00 for single and $2.00 for double numbers.
Some numbers are out of print. The Society has no resale discount for dealers.
The FLORIDA ANTHROPOLOGICAL SOCIETY PUBLICATIONS, Nos. 1-5,
may be purchased from the Treasurer at $2.00 per available numbers. No. 6, a
symposium on fiber-tempered pottery of the Americas, published in 1972, may
also be purchased from the Treasurer for $2.00.
INFORMATION FOR AUTHORS
THE FLORIDA ANTHROPOLOGIST publishes original papers in all
subfields of anthropology with an emphasis on archaeology. Contributions
from allied disciplines are acceptable when concerned with anthropological
problems. The journal's geographical scope, in order of importance, in-
cludes Florida, Southeastern United States, Caribbean, and related regions.
Manuscripts should be double spaced and typed on one side only of an
8 1/2 by 11 inch sheet leaving at least a 1 inch margin on all sides. Contri-
butors are referred to American Antiquity and the Chicago Manual of Style
for matters of style and reference. Textual references should be similar to
"Smith 1964: 78-80; Jones 1969: P1. II, b). No footnotes. In the "Refer-
ence Cited" section, at the end of an article, do not abbreviate names of ser-
ies or journals. Books need full title, publisher's name, and place.
All illustrative material-line drawings (in India ink) arid photographs-
should be included in one numbered series of "Figures" suitable for a maxi-
mum reduction to 6 1/2 by 8 1/2 inches. Parts of figures are designated by
lower case letters. Figures must be accompanied by brief descriptive cap-
tions. All tabular material must be included in separately numbered "Tables. "
Twenty-five reprints without covers of each article are furnished free
to each author or group of authors. Additional copies, with or without covers,
may be purchased but the editor must be informed of the number desired when
the article is offered for publication.