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.
U. OF F. LIBRARIES
-- --- ,'
RIPLEY P. BULLEN MEMORIAL ISSUE
PUBLISHED BY THE
FLORIDA ANTHROPOLOGICAL SOCIETY, INC.
THE FLORIDA ANTHROPOLOGIST is published quarterly in March,
June, September, and December by the Florida Anthropological
Society, Inc., c/o Room 130. The Florida State Museum, The
University of Florida, Gainesville, FL 32611. Subscription
is by membership in the Society for individuals and institu-
tions interested in the aims of the Society. Annual dues are
$6.00; student members $4.00. Requests for memberships and
general inquiries should be addressed to the Secretary; dues,
subscriptions, changes of address, and orders for back issues
to the Treasurer; manuscripts for publication to the Editor;
and newsletter items to the President. Second class postage
paid at Gainesville, Florida.
OFFICERS OF THE SOCIETY
President: Raymond Williams
Dept. of Anthropology, Univ.
of South Florida, Tampa, FL 33620
1st Vice President: George W. Percy
Div. of Archives, History and Records
Management, The Capitol,
Tallahassee, FL 32304
2nd Vice President: Paula J. Fields
1160 King Street
Merritt Island, FL 32952
Secretary- Nedra Lexow
1124 Harrison St.,
Hollywood, FL 33020
Treasurer: Norcott Henriquez
1510 Dewey St.
Hollywood, FL 33020
Directors at Large
Three years: Thomas C. Watson
203 Carolyn Avenue, Panama
City, FL 32401
Two years: Robt. C. Johnson
4250 Melrose Avenue,
Jacksonville, FL 32210
One Year: Ray. C. Robinson
1020 4th Street North,
St. Petersburg, FL 33701
Editor: Jerald T. Milanich
Department of Social Sciences
Florida State Museum
Gainesville, FL 32611
Kathleen A. Deagan
Dept. of Anthropology
Florida State University
Albert C. Goodyear
Institute of Archeology
University of South Carolina
Roger T. Grange, Jr.
Dept. of Anthropology
University of South Florida
Ann S. Cordell
John W. Griffin
Historic Key West Preservation
Board, Key West
George W. Percy
Division of Archives, History
and Records Management
Florida Dept. of State
VOLUME 30, NUMBER 2 JUNE 1977
Ripley Pierce Bullen, 1902-1976 ...................... 34
Editor's Page ........................................ 36
Excavations at the Old Oak Site, Sarasota,
Florida: A Late Weeden Island-Safety
Harbor Period Site,
by George M. Luer ............ .................. 37
Markham Park Mound No. 2. Broward County, Florida,
Broward County Archaeological Society
Project prepared by Wilma B. Williams and
Bert Mowers .................................... 56
Distribution of Paleo-Indian Projectiles in
by Ben I. Waller and James Dunbar .............. 79
Subsistence Systems in the Southeast
by Elizabeth S. Wing ........................... 81
RIPLEY PIERCE BULLEN, 1902-1976
Ripley P. Bullen, Curator Emeritus at the Florida State
Museum, University of Florida, died in Gainesville on December
25, 1976, at the age of 74. Although we of the Florida
Anthropological Society knew Ripley best for his archeological
work in Florida, his professional interests and research
additionally encompassed the Southwest and Southeast United
States, New England, the Lesser Antilles, the Greater Antilles,
and Central America. His bibliography, which includes more than
250 monographs and articles, reflects his tireless efforts to
advance our understanding of New World prehistory. Along with
his scholarly pursuits, Ripley devoted much time to aiding the
research of many non-professional archeologists. The accomplish-
ments of the Florida Anthropological Society and the state of our
knowledge concerning Florida's prehistoric peoples are in a large
part due to Ripley's efforts.
Ripley was born in Winthrop, Massachusetts, on September 21,
19,02. He received an M.E. degree in mechanical engineering in 1925
from Cornell University and in that same year began a fifteen-year
career with General Electric Company in New York and Massachusetts.
During that time he met his wife, Adelaide Kendall, in Massachusetts
and they were married in 1929.
By 1940 Ripley's expanding interest in archeology had grown
to such an extent that he left General Electric and joined the staff
of the Robert S. Peabody Foundation for Archaeology at Phillips
Academy, Andover, Massachusetts. At that time he also began graduate
work in anthropology at Harvard University, serving as a Teaching
Fellow in 1943-1945. During his eight years at Andover, Ripley
participated in a number of archeological investigations in the
American Southwest and Massachusetts.
Ripley came to Florida in 1948 as Assistant Archeologist with
the Florida Board of Parks and Historic Memorials in Gainesville.
In the year of his arrival he helped to found the Florida Anthropol-
ogical Society and its journal, The Florida Anthropologist. Not only
did he frequently publish in the journal, but he served the Society
by holding a number of offices: Treasurer, 1950 to 1956; Resident
Agent, 1966 to 1976; First Vice-President, 1967; President, 1968 and
1969; and Editor, 1970 until just prior to his death.
In 1952 the Florida Park Service discontinued its archeologi-
cal program and its collections and data were transferred to the
Florida State Museum. Ripley was appointed first Curator of Social
Sciences and also served as Chairman of the new Social Sciences
Department from that time until 1969. In 1973 he retired from the
Museum, but as Curator Emeritus he continued to work daily in his
office. His editing and writing were carried out at a regular pace
to within a few days of his death. He conscientiously made pro-
visions for transferring the responsibilities of his various edi-
torial chores and for having uncompleted manuscripts finished.
Many organizations saw fit to honor Ripley and his accom-
plishments. The University of Florida awarded him an honarary
Doctor of Science degree in March, 1976, in recognition of his
distinguished career and his service to the University and the
people of Floria. In May of that year the Florida Anthropological
Society awarded Ripley and his wife, Adelaide, honorary life member-
ships as a way of thanking them for their many contributions to the
Society. A posthumous award has been bestowed by the Florida State
Museum which has begun the Ripley P. Bullen Monographs in Anthropology
and History series, the first number to be published early next year.
In addition, the Florida State Museum Associates annually award the
Ripley P. Bullen Medal to the University of Florida student who has
written the outstanding paper on the anthropology of Florida and the
Ripley Bullen was a wonderful man; one who contributed much
to us. Perhaps the best way we of the Florida Anthropological Society
can honor Ripley is to dedicate ourselves to carrying on the work to
which he was so devoted. He would have wanted it that way.
The Board of Directors of the Society, with overwhelming
support from the membership, have made this number of The Florida
Anthropologist the Ripley P. Bullen Memorial Issue to honor
Ripley's service to the Florida Anthropological Society. For many
years he provided the base on which the Society was able to grow
The only other such memorial number of the journal was Vol.
17, No. 2, 1964, dedicated to John Mann Goggin. That issue
contained the proceedings of the third and fourth Annual Conference
on Historic Site Archaeology and is still in print.
Several individuals have written the editor with additional
information regarding articles that have appeared in past issues of
The Florida Anthropologist. A whole host of people noted that the
"Spanish" pipes described in Wilfred T. Neill and George R.
Ferguson's article "Spanish Trade Pipes from Marion County, Florida"
(Vol. 29, No. 3, 1976) are, in fact, 19th-century pipes of American
manufacture. They were probably used both by Seminole and early
settlers in Florida in the 19th century. Ivor Noel Hume's book, A
Guide to Artifacts of Colonial America (1970), illustrates similar
19th century specimens (p. 303).
Donald G. Hunter of Alexandria, Louisiana, has provided data
on a cane Coushatta (Koasati) basket from the mid-1930's that is
very similar to the early 19th-century Seminole basket described by
K.A. Deagan ("An Early Seminole Cane Basket," Vol. 30, No. 1, 1977).
According to Mr. Hunter, the Coushatta basket, pictured in The Florida
Anthropologist, Vol. 28, No. 1, p. 37, Fig. 7, b, resembles the
Seminole basket in size and manufacture, except for an "X" design
on the sides produced by two dyed strands. Regarding the use of
palmetto for Southeast baskets, Mr. Hunter goes on to say: "During
fieldwork with the Bayou Blue Coushatta near Elton, Louisiana a key-
informant, Mr. Bel Abbey, noted that his mother would often utilize
palmetto inter-mixed with split cane in her baskets. The leaf portion
of the frond was not used. However, the stem portion of the frond
was peeled in a similar fashion as that used in preparing the 'river
cane'. The palmetto was used because its inner surface was more
porous than that of the cane. When dyed, the palmetto absorbed more
of the dyes and produced richer colors. Palmetto prepared in this
manner was also used as the sole material in some Coushatta baskets.
Mr. Abbey states that only small baskets could be made from the
palmetto stems because of the short length of the fronds. Palmetto
was and remains today relatively scarce in the vicinity of the Bayou
Blue Community. This is reportedly the reason why more palmetto was
not used in this manner."
The Florida Anthropological Society has dropped its society
membership to the Eastern States Archeological Federation for finan-
cial reasons. Individuals, however, may join for $10.00. Apply to
Faye L. Stocum, Business Manager, ESAF, Island Field Museum, RD#2,
Box 126, Milford, DE, 19963.
EXCAVATIONS AT THE OLD OAK SITE, SARASOTA, FLORIDA:
A LATE WEEDEN ISLAND-SAFETY HARBOR PERIOD SITE
George M. Luer
In 1955 the family of the author purchased a two acre lot
on the shore of Sarasota Bay. Part of a shell midden extended
through the property. In 1956 the family built, upon midden de-
posit, a house overlooking Sarasota Bay. The author, born soon
after the completion of the house, has known of and been fascinated
by the midden "in my yard" as long as he is able to remember.
Figures 1 and 2 indicate the location of the midden, the
Old Oak site along Sarasota Bay. Pinelands, now residential areas,
once stretched to the north, east, and south of the site. Just
west of the Old Oak site and extending southward into Roberts Bay
is a large tidal flat now interrupted by the Siesta Key Bridge,
the San Remo development, and the Intracoastal Waterway and ac-
companying spoil islands. Big Pass lies two kilometers to the
west of the Old Oak site and downtown Sarasota three kilometers
to the north.
A hammock of live oak (Quercus virginiana) and cabbage palm
(Sabal palmetto) with some red cedar (Juniperus silicicola) covers
the two shell heaps of the Old Oak site. Man has cleared the un-
dergrowth from the northern midden but the southern is still in a
natural climax condition. It possesses a dense understory of wild
coffee (Psychotria undata) and marlberry (Ardisia escallonioides).
Such growth is typical only of well-drained, lime-rich coastal
situations such as shell middens or the naturally afforded old
coastal ridges of barrier keys. Immediately east of the shell
heaps the sandy soil supports a forest that is in an advanced
state of transition, due in part to the lack of fire, from one of
slash pine (Pinus elliotii densa) and palmetto (Serenoa repens)
to one principally of live oak and cabbage palm.
A contour map, Figure 3, depicts the configuration of the
two middens at the Old Oak site prior to the placement of fill-
dirt upon the northern half of the northern shell heap in 1973,
the construction of two houses on the northern heap in 1956 and
1974, and the addition of filled land west of the middens in 1955.
The shell middens were, and in places still are, well-defined and
the contour of the land underlying the middens (the pre-occupation
surface) is, noting the gradual'slope of the surrounding natural
ground surface, readily imaginable. The shell heaps extend about
50 m inland although shell is scattered eastward another 135 m
toward a natural spring. Sometime during the early 1950's
developers dredged the spring and accompanying small swamp to
create a lake and filled the encircling lowland for homesites.
A creek, which drained the swamp and continues to drain the lake,
flows by the southern shell midden and into Sarasota Bay. What
appears to be a causeway, composed of sand and shell, at places
The Florida Anthropologist, vol. 30, no. 2, June 1977
Fig. 1. Southern Sarasota Bay, ca. A.D. 1900; stippled area
represents grass or sand shallows, remaining shaded
area represents mangrove; 1. the Old Oak site; 2, the
Post Office site; 3, site at Mound Street.
Fig. 2. Southern Sarasota Bay, A.D. 1975; stippled area repre-
sents grass or sand shallows, remaining shaded area
represents mangrove; 1, the Old Oak site; 2, the Post
Office site; 3, site at Mound Street.
OLD OAK SITE
easily discernable but at other places not, roughly parallels
the stream on the north from the southern margin of the southern
shell heap inland for 140 m toward the spring. No burial mound
is known at the site but an indistinct sand hillock 60 m east
of the northern midden appears to be of artificial construction
and could be a small burial mound.
CERAMIC SURFACE COLLECTION
Over the past ten years the author has gathered a surface
collection of potsherds from the Old Oak site. Many plain sand-
tempered sherds were found. These sherds contained different
amounts of fine sand, coarse sand, or mixed fine and coarse sand.
Many sherds conformed to the definitions of Glades Plain or
Englewood Plain ceramics but, since the potsherds found appeared
to form a single natural group, these seeming artificial divisions
were not recognized. Most typological definitions used herein,
however, are to be found in Archeology of the Florida Gulf Coast
(1949) by G. Willey.
Many Belle Glade Plain and St. Johns Check Stamped sherds
were also discovered. A few potsherds of St. Johns ware, other
than St. Johns Check Stamped sherds, were found. One was a
textile-stamped sherd and others were St. Johns Plain potsherds,
one of which was a highly polished rim sherd.
The surface collection also included a sand-tempered cord-
marked sherd and six incised potsherds. One was a quartz-
tempered rim sherd with an incised line 1.3 cm below the rim
that parallelled the rim. Another rim fragment was a St. Petersburg
Incised-like sherd with several groups of parallel lines at
varied angles to the rim. Also incised with parallel lines that
appeared, however, to be perpendicular to the rim was an unclas-
sified potsherd that had half a suspension or repair hole in an
edge. A second unclassified incised sherd displayed many short
lines incised at different angles so that many lines overlapped.
An Indian Pass Incised sherd and a small Safety Harbor Punctated
sherd were also found. Rims typical of these sherd types are il-
lustrated in Figure 4; sherds from the surface collection are
shown in Figure 5.
The potsherd collection indicates that the Old Oak site
was occupied sometime during the late Weeden Island and Safety
Harbor archeological periods. St. Johns Check Stamped pottery
is diagnostic of these two periods (Willey 1949:446) which to-
gether ranged from approximately 850 A.D. into the historic
period (Bullen 1965:306).
The author dug during July and August, 1973, on the western
S 100 ft
Fig. 3. Contour map of shell heaps at the Old Oak site in feet
above mean high tide level. Stippled area represents
area of shell. Broken line indicates area excavated.
OLD OAK SITE
shoreward slope of the northern midden (Fig. 3). An area was
outlined (Fig. 6) 11 m in length that consisted of 11 units.
Eight of the rectangular one meter by one half meter units were
excavated. The units excavated were, beginning at the base of
the midden andproceeding toward the crest of the midden, units
1 through 5, unit 8, and units 10 and 11. Units 5, 8, 10, and
11 were expanded to create areas of one square meter. Units 1,
5, and 11 were excavated to a sufficient depth that the entire
midden deposit was penetrated. Intervening units were dug so
that the continuity of the different strata encountered could be
ascertained. All the units were excavated with a trowel in
arbitrary ten centimeter levels. Specimens uncovered were bagged
or boxed according to unit and level.
Strata Encountered. Unit 1, at the base of the midden,
exposed five strata, four of which were midden strata. The sur-
face stratum was a shallow recent accumulation of gray sand.
Broken pieces of glass were in this layer. The sand also contained
several valves of the pointed venus (Anomalocardia anberiana) and
worn valves of the broad-ribbed cardita and cross-barred venus
as well as the shells of several other small mollusks, all also
common bay species, probably deposited by high bay waters. The
second stratum encountered, Layer VI, rose to the surface at
unit 2 (Fig. 6) and extended up the slope. This uppermost midden
layer was composed of dark gray sand, bits of shell, the shells
of fighting conchs and left-handed whelks, and small worn potsherds.
Layer IV was a stratum of densely packed unbroken surf clam
valves. In it also were fighting conch shells and the valves
of bay scallops and oysters. Midway through this stratum in
unit 1, at a depth of about 40 cm, the shell began to be of a
yellow color. This 40 cm depth was approximately equal with the
modern level of high tide. The shell above the high tide level
was white and as unit 1 progressed downward the shell became
bright orange. The fourth stratum encountered, Layer III, con-
sisted of the shells of fighting conchs, left-handed whelks,
brown tulips, and horse conchs, and the valves of bay scallops.
Layer II was composed primarily of packed bay scallop valves.
At a depth of 70 cm in Layer II black sticky bay mud appeared.
Some shell continued into this soft mud. Shell that was buried
in mud was not orange but had remained white.
Layers II, III, IV, and VI continued upward to unit 4. Surf
clam valves disappeared, however, in unit 4. Unit 5 contained
Layers II, III, and VI plus another stratum. Below the bay
scallop valves of Layer II was a very thin stratum, Layer I,
containing oyster and tulip mussel valves. This thin layer was
directly above gray sand in which were mixed bits of charcoal.
In the northwest corner of unit 5 was a circular deposit of
blackened sand and charcoal 50 cm in diameter. The gray sand
quickly faded into sterile white sand as unit 5 extended downward.
b c d e f g
L I 5cm
Fig. 4. Rim profiles
b, St. Johns
from the Old Oak site; a, Belle
Check Stamped; c-g, plain sand-
Fig. 5. Potsherds from surface collection; a,b, Belle Glade Plain;
c, St. Johns Plain with burnished surface; d,e, St. Johns
Check Stamped; f, Indian Pass Incised; g-i, plain sand-
tempered; j, textile-stamped; k, incised sand-tempered;
1, quartz-tempered with Weeden Island rim; m, cord marked.
2 I 3 4
'-i 5 8 10 11
0.5m I --
.--i.-- ". II ,--H
Pits excavated. Top: units from above; center: side view
of units; bottom: side view of units with vertical scale
Units 10 and 11 exposed four major strata of the midden -
Layers I, II, V, and VI. The uppermost stratum, Layer VI, was
much thicker in these two pits near the crest of the midden
than in pits at the base. As at the base, the layer was primarily
dark gray sand. Charcoal and burned and unburned bone and shell
was mixed with the sand and potsherds abounded. In units 10 and
11, however, Layer VI was actually composed of four layers, three
of which were of sand and divided by bits of shell. The fourth
stratum was limited to the northern halves of units 10 and 11
and was a buried pile, sandwiched between two of the dark gray
sand strata, of large whole shells of left-handed whelks. Just a
portion of the buried pile was penetrated and 96 shells were ex-
posed. The next layer encountered in units 10 and 11, Layer V,
was a clearly defined stratum of light gray sand and fine crushed
shell. Bones, sherds, and charcoal were in the layer. Three
columns of dark gray'sand, linearly positioned, descended through
the shell bits of Layer V and just into the shell of Layer II.
These columns of sand probably represented post holes. All three
were separated by approximately 30 cm and the column diameters,
from east to west, were seven, nineteen, and ten centimeters.
The dark gray sand that filled these "post holes" was from Layer
VI. Under Layer V were Layers II and I, both of which were com-
posed of shell Layer II of bay scallop valves and the shells
of fighting conchs (the relative abundance varied in different
areas of the stratum) and Layer I of fighting conch shells,
oyster shell, and the valves of bay scallops, tulip mussels, and
rose cockles. The shell in both strata, especially Layer II,
was loosely packed and much air space was between the shell.
Both strata contained splinter-like shiny crystals of calcium
carbonate (CaCO3). The crystals were exceedingly abundant in
Layer I. Brown clumps of well-preserved fish scales were also
in the layer. Gray sand was below Layer I and some shell, charcoal,
and potsherds were mixed in the sand.
Unit 8 was dug to a depth of about 30 cm. Layer VI was
penetrated as was Layer V which thinned and disappeared within
the area excavated for unit 8. When Layer II was reached, digging
Summary of Stratification. The sand beneath Layer I was
the pre-midden ground surface and appeared to have been lived
upon. Layers I, II, III, and IV were trash strata essentially
composed of shell and represented an accumulation from perhaps
one or more nearby living areas. The pre-midden bayshore was ex-
posed during the accumulation of Layer I but the accumulation
of Layer II buried the shore and apparently spilled out onto bay
mud. Layers V and VI were the remains of a living area. Three
living area surfaces were contained in Layer VI. Much of the
refuse of this living area was probably discarded at nearby areas
of refuse where trash strata were added to the midden similar to
OLD OAK SITE
the four below Layers V and VI. The shell and sand of Layer V
and sand of Layer VI were apparently deposited by inhabitants
of the Old Oak site to create surfaces that could be lived upon.
Tools of stone and bone and tools and ornaments of shell
were concentrated in Layer VI. This layer also had the greatest
abundance of potsherds. Artifacts are shown in Figure 7.
Pottery. The sherds uncovered suggest a time of occupation
by or within that indicated by the surface collection of potsherds
from the Old Oak site. Five St. Johns Check Stamped sherds were
found, all of which were in Layer VI (one in unit 8, two in
unit 10, and two in unit 11). One St. Johns Scored-like sherd
was found in Layer VI (unit 8). In the top of Layer II (unit 11)
was a buff-colored sand-tempered check stamped sherd. One sand-
tempered linear check stamped sherd was discovered in Layer VI
(unit 5) and another in Layer V (unit 10). These nine sherds
were the only decorated sherds found and represent about two and
one half percent of the potsherds excavated.
Two St. Johns Plain sherds were found in Layer VI (unit 11)
and another in Layer V (unit 11). Several Pinellas Plain sherds,
all of which fit together, were in Layer IV (unit 1). Eighteen
Belle Glade Plain sherds, of which 16 were uncovered in Layer VI
(one in unit 5, three in unit 8, four in unit 10, and eight in
unit 11) and two in Layer V (unit 10), represent about five per-
cent of the sherds excavated.
The vast majority of the potsherds that were excavated,
about 92 percent, were plain sand-tempered sherds. Almost three-
fourths of these sherds were found in Layer VI and, unlike
some large sherds with sharp edges found in the trash strata,
almost all were small and worn, presumably from having been ex-
posed and walked on when upon the surfaces of the living area.
Clay. A small piece of unfired clay was found in Layer VI
(unit 10). The buff-colored clay contained a small amount of
Stone. Stones were most common in Layer VI. In unit 5
was a spherical piece of rough brown sandstone, perhaps a hammer-
stone, in which were fossil shells. In the stone were several
holes made by boring clams, and such holes were also in two
stones found in unit 10 and in one from unit 11. These latter
stones were rounded and of rough sandstone, one of which con-
tained fossil shells. Other pieces of sandstone as well as
limestone were also in the stratum. In Layer VI (unit 5) was
the only projectile point discovered, a quartz Westo point. The
layer (unit 11) also yielded a chip of mineralized bone that
was perhaps a scraper or small cutting implement.
Two stones were uncovered in Layer V (unit 10). One was
a small piece of limestone encrusted with worm shell, oyster
shell, and a barnacle. The other was a water-smoothed piece of
mineralized sea cow rib.
In the bay scallop valves of Layer II (unit 10) was a cache
of six apparently water-smoothed stones. One stone was a reddish-
brown piece of mineralized sea cow rib and another was a piece
of light gray smooth sandstone with many small holes, of diameters
of about a millimeter, produced by the boring sponge, Cliona sp.
(G. D. Shaak, pers. comm.). The remaining four stones were of
a similar brown rough sandstone. All the stones were of a similar
size. A fifth rounded stone of brown rough sandstone was in the
bay scallop valves (unit 10) but not with the cache of six.
A piece of bog iron was in the sand below the midden in
Coral. Several pieces of fossilized coral were unearthed
in Layer VI. The pieces were of the genera Oculina and Solenastrea
(G. D. Shaak, pers. comm.).
Artifacts of Bone. A double-pointed bone sliver, fashioned
by man, was found in Layer VI (unit 11). In Layer II (unit 5)
was the anterior half of the plastron of a large pond turtle,
Chrysemys sp. This plastron exhibited a deep cut that passed
laterally across the hyoplastrons just to the posterior of the
entoplastron. A shallower groove of less width, perpendicular
to the deep cut, ran across the posterior half of the entoplastron
and between the hyoplastrons to the suture with the hypoplastons
where the plastron was broken. A smaller cut ran between the
epiplastrons and onto the anterior half of the entoplastron.
Two notches were in the edge of the right hyoplastron above the
Artifacts of Shell. A columella of the shell of a horse
conch (unit 3) and four columella of left-handed whelk shells
(one from unit 3, two from unit 5, and one from unit ll--the
latter slightly burned) were found in Layer VI. One pick (unit
10) and one hammer (unit 8), both made from the shells of left-
handed whelks, were in the layer. Layer VI also yielded what
are commonly referred to as shell ornaments. An olive shell
"bead" was uncovered in unit 11 as well as three neatly perforated
valves (the umbos removed) of ponderous arks.
The shell bits of Layer V (unit 10) yielded one left-handed
whelk shell hammer and a large fragment of the worked shell of
a left-handed whelk. This fragment was apparently of a shell
vessel or dipper.
In Layer II was discovered the large columella of a left-
handed whelk shell (unit 1) and a pick made from a shell of the.
same species (unit 5). A perforated ponderous ark valve was
also found (unit 11).
c, quartz Westo point; d, double-pointed bone sliver;
5cm 5 cm
Fig. 7. Artifacts; a-f, from units excavated; g,h, from surface
collection; a, Busycon pick; b, mineralized bone chip;
c, quartz Westo point; d, double-pointed bone sliver;
e, Busycon hammer; f, incised Chrysemys plastron; 9,h,
mineralized shark tooth points (?).
Charcoal. Unfortunately not all pieces of charcoal un-
earthed were saved but, of those that were, pieces were identi-
fied that were of the burned wood of slash pine.
Food remains were in every layer. Molluscan, crustaceous,
echinodermatous, cartilaginous and osseous piscatorial, amphibian,
reptilian, avian, and mammalian remains were found.
Mollusk. The shells of 35 species were discovered and the
shells of perhaps 25 species represent mollusks consumed by in-
habitants of the Old Oak site. The letters on the left of the
list that follows roughly indicate the relative abundance of
shells in the eight excavated units: (a), abundant; (c), common;
(u), uncommon; (r), rare.
1. (a) Argopectenirradians concentricus (bay scallop)
2.(a) Spisula solidissima similis (surf clam)
3.(a) Busycon contratium (left-handed whelk)
4.(a) Strombus alatus (fighting conch)
5.(c) Trachycardium egmontianum (rose cockle)
6.(c) Busycon spiratum pyruliodes (pear whelk)
7.(c) Fasciolaria tulipa (brown tulip)
8.(c) Pleuroploca gigantea (horse conch)
9.(c) Crepidula fornicata (slipper shell)
10.(c) Crepidula maculosa (spotted slipper shell)
11.(c) Crepidula aculeata (spiny slipper shell)
12.(u) Anomia simplex (jingle shell)
13.(c) Fasciolaria lilium hunteria (banded tulip)
14.(c) Mercenaria campechiensis quahogg)
15.(u) Macrocallista nimbrosa (sunray)
16. (u) Polinices duplicatus (cat's eye)
17.(c) Crassostrea virginica (common oyster)
18.(c) Ostea equestris (crested oyster)
19. (c) Melogena corona (king's crown)
20.(c) Genkensia demissa granosissima (ribbed mussel)
21.(u) Euglandina rosea
22. (c) Neotia ponderosa (ponderous ark)
23.(c) Anadara floridana (cut-ribbed ark)
24.(c) Arca zebra (turkey wing)
25.(u) Dinocardium robustum vanhyningi (giant cockle)
26.(c) Atrina sp. (pen shell)
27.(c) Miodulus americanus (tulip mussel)
28.(u) Murex pomom (apple murex)
29.(r) Conus spurius atlanticus (alphabet cone)
30.(u) Carditamera floridana (broad-ribbed cardita)
31.(u) Chione cancellata (cross-barred venus)
32.(r) Plicatula gibbosa (cat's paw)
33.(r) Spiroglyphus sp. (worm shell)
34.(r) Vermicularia sp. (worm shell)
35.(r) Oliva sayana (olive)
OLD OAK SITE
As is evident from the list, the shells of almost all the
large mollusks native to the shallows of the Sarasota area were
found. The fact that the shells of so many species were "abundant"
or "common" in the midden reflects not only edibility of the
species but also the location of the Old Oak site on an open bay
opposite a pass (Fig. 1) where there was much, and a variety of,
exploitable shellfish habitat.
Bay scallops live in bay waters below low tide level in
and around beds of turtle grass (Thalassia testudium). Turtle
grass beds still exist in Sarasota Bay but were much more exten-
sive before the dredging and filling of the past 65 years (Figs.
1 and 2). The mature free-swimming bay scallop is a gregarious
animal. The great number of bay scallop valves encountered in
Layer II seem to indicate that inhabitants of the Old Oak site
found schools of bay scallops, collected the mollusks in great
numbers, and brought their catch to the Old Oak site. Areas of
Layer II consisted entirely of many hundreds of "stacked"
bay scallop valves. Valves, often four of five, were cupped in-
side each other and lying upon others that were similarly cupped
and upon still others. Bay scallop valves were perhaps cupped
and "stacked" when inhabitants of the Old Oak site shucked large
quantities of bay scallops. When excavated, many of the "flat"
upper valves (flatter than the more convex white-colored lower
valves) still exhibited the original coloration of reddish brown
with white spots. The valves of several bay scallops were found
still together the ligament of each pair unbroken.
Surf clams live, and sometimes abound, in fine white sand
below the level of low tide on beaches of passes and the Gulf.
Perhaps Layer IV represents a collection of a single day of
thousands of surf clams gathered on nearby Gulf beaches, trans-
ported by canoe to the Old Oak site, and shucked on the bayshore
where perhaps canoes were beached the shell tossed onto a
trash area and the clam meat consumed at a living area.
The left-handed whelk is a carnivorous univalve that moves
about bay flats, turtle grass beds, and deeper bottoms of sand
and shell in search of prey primarily hard-shelled clams (Abbott
1968:136). Several piles of large left-handed whelk shells
occur on the surface at the Old Oak site and one such accumula-
tion, as has been written, was found sandwiched between two living
floors in units 10 and 11. The shells of this accumulation were
unbroken and thus perhaps indicate that the whelks were extracted
in a manner that did not necessitate the breaking of shells.
Perhaps large left-handed whelks were brought to living areas
and boiled in large earthenware pots. Then, after boiling,
whelks could have been easily removed from unbroken shells, con-
sumed, and the heavy large shells (clean with the animals re-
moved) simply cast into piles at certain locations about living
areas hence the pile in Layer VI. One left-handed whelk
shell in the Layer VI accumulation and another in Layer II
(unit 11) had been extensively broken while the whelks within
were alive and each shell, due to the survival of the mollusk,
exhibited considerable growth past damaged areas.
The fighting conch lives below the level of low tide in bays.
Fighting conchs consume algae (Abbott 1974:143) and are easily
overlooked on dark bottoms because of the brown periostracum
covering the shells. Shells of fighting conchs in the midden
were whole or broken. Perhaps whole shells represent fighting
conchs that were boiled to extract the animal for consumption.
Broken shells were without spires and perhaps represent a second
manner of extraction. Breaks in shells indicated that spires
had been removed by blows delivered to the shells near the suture
between the body whorl and penultimate whorl. Perhaps left-
handed whelk shell hammers were used to remove the spires. The
shells of two young fighting conchs were uncovered but all other
shells were of adults and possessed a thick flared lip.
The rose cockle occurs in bays buried in sand below the
level of low tide in turtle grass beds. Pear whelks, brown
tulips, and horse conchs frequent grass beds. Often horse conchs
are also found on the periphery of oyster beds below low tide
level. Slipper, spotted slipper, and spiny slipper mollusks
as well as jingle shells often live on the shells of other
mollusks. Some shells of these species were still attached to
shells of horse conchs in the midden but others, however, were
found unattached and perhaps the animals within had been con-
sumed. The banded tulip lives on oyster beds and bay flats
that are exposed at low tide. Such flats often harbor quahogs,
sunrays, and cat's eyes. Many quahog valves in the midden had
a lip that was chipped perhaps from blows dealt while opening
the clams with left-handed whelk shell hammers. Other valves
were broken probably also a result of shucking. Oysters were
gathered from oyster beds where king's crowns abound. In the
midden the body whorls of king's crowns were almost always
broken, probably as a result of extracting the animal. Ribbed
mussels live in the intertidal zone buried in sandy mud about
the roots of red mangroves (Rhizophora mangle) or on shores
covered with smooth cordgrass (Spartina alterniflora). The
carnivorous land snail Euglandina rosea, a consumer of other
land snails (Solem 1974:'209), often lives in damp rotting
material in high grass about the margins of hammocks and swamps.
Ponderous arks, cut-ribbed arks, and turkey wings (two matching
valves were in Layer IV), and giant cockles are mollusks most
commonly found in sandy Gulf shallows. The pen shell is found
in both the Gulf and bay.
Probably brought with fill from the bayshore were worn
valves of the broad-ribbed cardita and cross-barred venus -
several worn valves of the latter were in the shell of Layer
V. A rock encrusted with worm shell was also in the layer.
A small piece of a worm shell was in Layer VI where also was an
52 OLD OAK SITE
olive shell "bead". Olives live on sand of pass and Gulf
Crab. Pieces of the claws of stone crabs (Menippe mercenaria)
were common throughout the midden. Several pieces of the claws
of blue crab,. Callinectes sp., were also found. Blue crabs live
in a variety of habitats and are abundant in bays. Stone crabs
are often found in burrows in turtle grass beds.
Sea Urchin. Spines and plates of the tests of the sea
urchin Lytechinus variegatus were present throughout midden
deposit. This species is very common on rock and shell bay bot-
tom below the level of beds of turtle grass in which the species
occurs less frequently. Perhaps the gonads of the sea urchin
Shark and Ray. Remains of six species of shark and three
species of ray, all of which are shallow Gulf and estuarine
species (P. W. Gilbert, pers. comm.), were discovered.
1. Ginglymostoma cirratum (nurse shark)
2. Carcharhinus leucas (bull shark)
3. Carcharhinus maculipinnus (spinner shark)
4. Galeocerdo cuvieri (tiger shark)
5. Negaprion brevirostris (lemon shark)
6. Sphyrna sp. bonnetheadd shark)
7. Pristis sp. (sawfish)
8. Aetobatus narinari (spotted eagle ray)
9. Rhinoptera bonasus (cownose ray)
Boney Fish. Fish of all the species of which remains were
unearthed occur in bays. In addition to those species listed,
remains of ladyfish (Elops saurus) and flounder (Paralichthys
albigutta) were found on the surface at the Old Oak site. The
diversity of the species represented indicates that fishing
methods had to have been employed that could secure bottom fish
such as flounder and black drum, grassbed fish such as seatrout,
fast-swimming species of open bay such as crevalle jack and
ladyfish, and surface fish such as mullet and snook. Bows and
arrows and fish spears were probably used in fishing and fish-
hooks, nets, and weirs also could have been utilized.
1. Arius felis (bay catfish)
2. Bagre marinus (sail cat)
3. Opsanus sp. (toadfish)
4. Centropomus sp. (snook)
5. Epinephelus sp. (grouper)
6. Pomatomus saltatrix (bluefish)
7. Caranx hippos crevallee jack)
8. Haemulon sp. (grunt)
9. Archosargus probatocephalus sheepsheadd)
10. Lagodon rhomboides pinfishh)
11. Cynoscion sp. (seatrout)
12. Menticirrhus sp. (whiting)
13. Pogonias cromis (black drum)
14. Sciaenops ocellata (redfish)
15. Mugil sp. (mullet)
16. Chilomycterus schoepfi (striped burrfish)
17. Diodon hystrix (porcupinefish)
Amphibian and Reptile. In addition to the remains uncovered,
remains of the green sea turtle (Chelonia mydas), Florida snapping
turtle (Chelydra serpentina osceola), gopher tortoise (Gopherus
polyphemus), and Florida soft-shelled turtle (Trionyx ferox) were
found elsewhere at the Old Oak site.
1. Siren lacertina (mud eel)
2. Alligator mississipiensis (alligator)
3. Chrysemys sp. (pond turtle)
Bird. Many bird remains were found but no remains could be
attributed to any species.
Mammal. Remains of the Atlantic bottlenose dolphin (Tursiops
truncatus) and eastern woodrat (Neotoma floridana) were found at
the site in addition to the remains of the four mammal species
1. Didelphis marsupialis (opossum)
2. Procyon lotor (raccoon)
3. Odocoileus virginianus (deer)
4. Sylvilagus sp. (rabbit)
Potsherds indicate that aborigines lived at the two oc-
cupation and refuse middens of the Old Oak site sometime during
the period embracing the late Weeden Island and Safety Harbor
archeological periods. They procured much of their sustenance
by hunting, fishing, and gathering shellfish. Considering the
number of food species represented in midden refuse (6 mammal,
6 reptile, 1 amphibian, 28 fish, 1 urchin, 2 crab, and perhaps
25 mollusk), the Indians extensively utilized the food resources
of both terrestrial and marine ecosystems. It is not known
whether Indians occupied the Old Oak site seasonally, at ir-
regular intervals, or continuously. Whether they did or did not
practice horticulture is also unknown but Indians of the Weeden
Island and Safety Harbor cultures are believed to have been
horticulturists (Willey 1949:403,477). The spring and rich land
surrounding it, located east of the Old Oak site, probably pro-
vided conditions required for horticulture, and, if the Indians
of the Old Oak site practiced horticulture, probably led them
OLD OAK SITE
to inhabit the site. Perhaps the Indians of the Old Oak site
were associated with the Indians of the large village, ceremonial,
and burial site 5.5 km to the north at Whitaker Bayou dating
from the Weeden Island and Safety Harbor periods (Willey 1949:
344; Bullen 1950:29; Milanich 1972:37).
The author offers his sincere thanks to his friend Stanley
T. Stokes for aiding in the excavation of several pits and to
the late Ripley P. Bullen, former Curator Emeritus of Anthropology
at the Florida State Museum, for identifying many potsherds. The
author would also like to thank his brother, Bert Luer, for
catching many of the fish from which bones were obtained for
later comparison with bones unearthed from the midden. Through
the generous permission of Dr. Elizabeth S. Wing of the University
of Florida the collections of faunal remains at the Florida
State Museum were made available to me. These collections
proved invaluable. The aid of Dr. Graig D. Shaak of the University
of Florida and of Dr. Perry W. Gilbert of Mote Marine Laboratory
was also appreciated. Last but by no means least, the author
would like to thank friends and members of his family for their
aid and encouragement given in innumerable ways.
Abbott, R. Tucker
1968 Seashells of North America. Western Publishing
Company, Inc., Golden Press, New York.
1974 American Seashells, The Marine Mollusca of the
Atlantic and Pacific Coasts of North America.
Van Nostrand Reinhold Company, New York.
Bullen, Ripley P.
1950 Tests at the Whittaker site, Sarasota, Florida.
Florida Anthropologist, 3:21-30.
1965 Florida's Prehistory. In Florida From Indian
Trail to Space Age, by Carlton Tebeau, Chapter
23, Southern Publishing Co., Delray Beach.
Milanich, Jerald T.
1972 Excavations at the Yellow Bluffs-Whitaker
Mound, Sarasota, Florida. Florida Anthropologist,
1974 The Shell Makers, Introducing Mollusks. John
Wiley and Sons, New York.
Willey, Gordon R.
1949 Archeology of the Florida Gulf Coast.
Smithsonian Miscellaneous Collections, Vol. 113.
56 MARKHAM PARK MOUND NO. 2
BROWARD COUNTY, FLORIDA
Broward County Archaeological Society Project
prepared by Wilma B. Williams and Bert Mowers
The Markham Park mounds are two small tree islands not unlike
others in the Everglades, When Levee 35A was built, it formed the
eastern boundary of Conservation Area No. 2 and changed the
physical characteristics of the land to the east. As a result, the
newly made land now drained into the North New River Canal. Broward
County acquired some of this non-arable land for use as a park. It
is approximately ten miles west of the intersection of U. S. 441 and
State Road 84, in Section 33, Township 49 South, Range 40 East.
The Broward County Commission requested that the Broward County
Archaeological Society excavate the two mounds in the park and
write the reports. Excavation began in November 1971 and concluded
December 16, 1973. This report will concern Mound No. 2, Bd. 183.
The elevation of the mound is about four feet above the
surrounding land and has the remnants of a borrow pit on the west
and north sides while the mound slopes gradually to the east and
south. A post hole survey showed occupation debris over an area
at least 100 feet in diameter, making the site a very considerable
one as compared to the usual Glades hammock site. A single huge
ficus tree crowns the mound, with younger ficus trees on the
western slope, with holly, grapefruit and guava trees.
The contour readings were taken and a grid plan of numbered
five-foot squares east and west and lettered squares north and south
was laid out (Fig. 1). A total of 57 squares were excavated by
six inch levels, to sterile sand or bedrock. Samples were taken
of concretions encountered and of clay layers above bedrock.
Because the mound has been known as an Indian site for many
years, evidence of vandalism and of disorganized digging were seen.
This resulted in reversed stratification in some areas, and probably
the loss of some artifacts. However, we think the great area of
our excavations minimized the effects.
Eighty-eight people participated in the excavations during the
The antiquity of this site is confirmed by several potsherds
that are not usually found in south Florida hammock sites. The
most important would be two Orange fiber tempered sherds (Rouse
1951) found in Level 4. This would appear to give the site a Late
Archaic horizon. The very early St. Johns Incised and mat impressed
certainly fit into Bullen's Transitional (Bullen 1965) (Table 1).
The Florida Anthropologist, vol. 30, no. 2, June 1977
15 16 17 18 19
* .. 9" .
20 21 1 22 23 24 25 26 27 28 29 30
Fig. 1. Contour map and grid plan.
Table 1. Pottery distribution by levels, all
Pottery Types L-l L-2 L-3 L-4 L-5 L-6 L-7 Total
St. Johns Check Stamped
Peace Camp Plain
Key Largo Incised
Opa Locka Incised
Ft. Drum Rim-ticked
Cane Patch Incised
Gordons Pass Incised
North Miami Beach Incised
Natural Bridge Incised
Markham Incised (tentative)
Markham Plain (tentative)
Belle Glade Plain
4 14 50 19
Dunns Creek Red
St. Johns Plain
St. Johns Incised
12 15 5
1 1 34
59 9 3 22079
4 5 862
Totals 12273 10684 2900 724 100 17 8 26706
WILLIAMS AND MOWERS
Level 1 (0-6") is the surface humic layer, consisting of grass
and plant roots, partially rotted woody refuse, wind-blown dust
and sand, and other materials dropped by man. It is blackish,
granular in texture and contains some pottery, beads, bones, and
modern artifacts, such as shot gun shells, cartridge cases, and
bottle caps. In this level are found most of the Glades III
sherds, giving a rough age of the level of about 500 years. This
is a rule of thumb estimate, but it is surprisingly consistent
across the mound. Level 1 produced 45.956% of the sherds found,
which indicates a great deal of late use of the site. Naturally,
population growth would provide more sherds in later centuries and
especially in the last millennium.
Level 2 (6-12") is, in part, physically different from Level
1. About five inches down from the surface a hardening occurs,
producing a denser, blackish dirt with food bones, pottery sherds
and lenses of sand with ashes. Compaction of humus over the cen-
turies has compressed the soil and consequently the temporal spread
is greater for this level than for Level 1. Much of the Glades II
pottery, along with the remainder of the Glades III sherds, and a
variable amount of Glades I sherds, is found in this level. Our
rule of thumb estimate of this level is that it goes back in time
about 1,500 years. Level 2 contained 40.006% of the sherds.
Again, this shows relatively great use of the site during the last
Level 3 (12-18"). In appearance, much of this level is quite
similar to Level 2. However, pottery frequency drops off sharply.
Disturbances lessen, but location of pit assumes more importance.
Pits near the mound center yield less pottery than perimeter pits;
some in the central area have substantial amounts of hard, black-
ish organic concretion, often seen on Glades hammock sites. This
was a minor, local affair, not continuous or even very general.
It may be produced in concentrations of ash, or fire pits with
Temporal judgments become more uncertain in this level. Most
of the Glades II pottery disappears and only the older types of
Glades decorated sherds, such as Fort Drum Punctated and Rim-ticked
or tooled designs are found; Cane Patch, Sanibel and Gordons Pass
Incised or Punctated designs are sparingly found (Goggin 1964).
Level 3 contained 10.859% of total sherds dug, showing less use of
the site than at higher levels.
Level 4 (18-24") is usually sandier than Level 3, and contains
much less pottery. Glades Plain and St. Johns Plain predominate
and unclassified wares, mostly of west coast or upper east coast
sherds, are still found. Semi-fibered sherds, Orange fiber tem-
pered sherds, and early St. Johns Incised sherds, appear sparingly
(Fig. 2, d, g). This level seems to be of Florida Transitional
age (Bullen 1965), with possible extension into Late Archaic
(Orange) time. A number of Strombus celts and other shell material
found with the sherds are available for C-14 dating. Level 4 con-
tained 2.711% of the total sherds produced.
0 1 2 3 4 5 6
4 ()N[ i)[ (IME TF'[
Ji 11 1 I I1 1 111111 1 11i111Ull 1ll l i ll 1 (
0 1 2 3 4 5
Fig. 2. Ceramics; a-b, unclassified incised; c, unclassified
shell impressed; d, St. Johns Incised; e, triangular
punctated; f, Perico Incised-Punctated; g, Orange
WILLIAMS AND MOWERS
Level 5 (24-30") contained even less pottery, with types
similar to Level 4. Many bones and shell artifacts occur, and
animal species appear to be larger in size. At least Florida
Transitional age is indicated. Level 5 contained only .374% of
Level 6 (30-30"). Only 17 sherds were found, all of which
could be Transitional in age. This represented .064% of total
dug. The dirt grades into clean Pamlico sand.
Level 7 (36-42"). Only 8 sherds were found, representing
.03% of the pottery. At this level, most pits contained no
In general, the distribution of familiar Glades III, I and
I sherds in this mound corresponds to that of other Everglades
hammock sites excavated in Broward County. Glades III types,
such as St. Johns Check Stamped, Glades Tooled, Peace Camp Plain
and Surfside Incised, are mostly found in Level 1 with some in
Incised types of the Glades series ascribed to Glades II are
found in both Level 1 and Level 2, with some lower down. A number
of sherds of the recently named Arch Creek series, such as North
Miami Beach Incised and Natural Bridge Incised, as reported for
the Arch Creek site, were found. Most of these are variants of
other known designs (Miami-West India Archaeological Society and
Broward County Archaeological Society 1975).
An often seen Glades gritty ware type decorated with round
bosses of clay pressed out from the inside and arranged usually
in a single row around the pot below the rim is called Glades
Noded. No evidence of the point of origin of this design was
found. Laxson has reported similar sherds (Laxson 1970).
Many of the unclassified incised"sherds found are tiny rims
with incomplete patterns; others have rim ticks and rim grooves
not definitely identifiable. Many of the unclassified pot sherds
found are of sand tempered pastes not produced in the eastern
Glades area. Some are from the west coast of Florida and probably
represent vessels from the Naples area to Tampa and farther up
the Gulf Coast. These would include Perico, Pinellas, and other
wares from the many towns producing pottery in that area (Fig. 2,
a, c, e, f). Similarly, unclassified wares may be from the
St. Johns River Valley and upper east coast villages.
Belle Glade Plain comprises about 2% of the sherds found and
ranges from late Glades III to early Glades I. This small amount
is surprising in view of the short distance from this site to the
Belle Glade area. Travel up and down the eastern Glades was
common to southwest Florida Indians. Possibly travel by water
was more convenient on the east, with its plentiful camp sites,
than on the western side of the Glades. A few sherds each of the
Glades Red and Dunn's Creek Red were found.
Glades Plain, the local eastern Glades pottery, was by no
means exclusively the Arch Creek variety, as the hard, well-made
pottery found on the Arch Creek site is called (Miami-West India
Archaeological Society and Broward County Archaeological Society
1975). Probably less than one-half of the sherds were of the
Arch Creek variety. The origin of the rest is unknown. Glades
Plain comprised 82.7% of all sherds.
St. Johns Plain, the chalky were made of clay containing
sponge spicules and most common in the St. Johns River basin, ranges
widely in time and is the commonest component in the lowest pottery
levels. Much of the material found on this site contains con-
siderable fine sand. Only a few sherds were composed of the thick,
soft, pink-beige paste usually associated with Transitional
ware (to 1,000 B.C.). Many sherds of "muckware", a dark oily
appearing paste, were found.
A sand and fiber tempered ware (semi-fiber tempered), of which
five sherds were found, is presumed to be a Transitional ware be-
tween fiber tempering and sand tempering. Two sherds of Orange
fiber tempered ware, from different pots and different pits, but
both from Level 4, were found. One was a typical Orange sherd,
like the ones from South Indian Field (Rouse 1951), but the other
was different. It was very thick, soft, eroded and with fibers
in layers, still unburned. This may represent experimental pottery
or some transitional variety quite unlike semi-fiber tempered.
Luckily it was found in close association with seven Strombus celts,
from which it may be dated by C-14.
This orange fiber tempered ware is the most unusual sherd
found and is perhaps the most important from the site because the
possibility exists that it might be older than any other sherd yet
found in South Florida.
Two types of potsherds found in two different squares at
this site are tentatively named Markham Park Plain and Markham
Park Incised. Other sherds of these types either have not been
found in the Glades area or have not been recorded. As the distri-
bution of this "type" becomes better known, formal type definitions
may be warranted.
In Level 4 (18-24") of pit M-19 were found two rims and nine
body sherds of Markham Park ware. They are light, bluish gray in
color, sand tempered, and with an incised design. Groups of four
parallel lines were applied diagonally in alternate directions
below the rim. This is a pattern reminiscent of the much later
Miami Incised, but the latter design is of a larger number of lines
and not always with the alternating feature.
The pattern also is rather like certain Perico Incised rims
(Willey 1949), except that no horizontal lines appear on the
Markham sherd and the Perico Incised parallel lines are two and
five in number. Also, Perico Incised ware contains lumps of lime-
stone temper which Markham Park ware does not have.
WILLIAMS AND MOWERS
Miami Incised apparently is an Arch Creek variant of Glades
II age, while Markham Park ware, as described here, is Florida
Transitional or very early Glades I. The pastes are different,
although both are sand tempered. However, the bluish gray color
of Markham Park sherds may be due to marl in the clay or to
finely ground limestone. Possibly, the sherds represent the
short term use of a new clay source.
In Level 4 of pit P-25 were found one rim and 16 body sherds
of another pot, undecorated, with the same Markham Park bluish
gray color and similar paste, except that the sand tempering was
of larger grains of quartz sand, and the sherds are quite soft
and easily ground away. The interior is quite dark with a hard
skin and a softer interior with a cindery look, uniform in texture
and appearance. Possibly this is due to sand temper overloading,
or to a clay with poor firing characteristics; even a poor job of
firing this pot may account for the difference between the two
pots. Other body sherds found in this level seem to be from
another pot of the same ware. Rims are plain, rounded and thinned,
andpots are 12 to 14 inches in mouth diameter.
Another important pot found was of St. Johns Plain, five
pieces of which were found in pit L-24 in Level 7, 37 inches below
the surface. The pieces were glued together and formed the most
unusual vessel dug in Broward County. It is flat bottomed, with
sides curving slightly upward from the flat bottom at a 30 degree
angle. Another angle in one of the side pieces indicates that it
is a multi-sided pot, not necessarily a four-sided one. The flat
bottom of the pot was impressed with a woven design, obviously
from having been formed on a flat surface of matting. To our
knowledge, this is a first for Broward County (Fig. 3, a). The
shape is similar to Orange and early St. Johns vessels from east
This pot, while unquestionably a St. Johns ware, does not
exhibit the thick, soft pinky beige paste we have come to asso-
ciate with Florida Transitional St. Johns ware. Instead, it is
a thin, dirty, taupe-gray paste and contains considerable very
fine sand, perhaps naturally contained in the original clay body
from which the pot was formed.
The biconial clay balls of the Poverty Point type were not
used in the Glades, although one was found in G-23, Level 3 (Fig.
4, b). This ball follows exactly the description of the Poverty
Point cooking balls in size and form (Hunter 1975).
Ceramic objects other than pottery numbered six and consisted
of a few drilled sherds that may have been used as gorgets, pen-
dants or beads.
Fig. 3. Ceramics; a, St.
Johns mat impressed; b, early St.
r _, ft_
WILLIAMS AND MOWERS
SI II Ij~
IC N TIM ETERS111 1111
) 1 2
Fig. 4. Artifacts; a, microlith drill; b, bi-conical clay ball;
c, shell pendant; d-e, bone beads; f, silver bead.
I I II
Artifacts other than pottery were quite common. They con-
sisted of bone, shell and stone objects, with a few exotic
materials. Bone and shell artifacts, particularly shark teeth
and vertebrae, are usual on Everglades hammock sites, but the
number and variety of stone objects is unusual (Table 2).
Shark teeth were drilled for hafting and used as cutting-
graving tools. Some shark vertebrae became decorative items,
such as ear plugs.
The most important shell tools found were 165 Strombus celts.
These tools, made from the protruding lip of the Strombus gigas
conch, were made of the hardest material available to the Indians
(Laxson 1964). Celts were chipped to size and shape, then ground
to either axe or adze blade, hafted and used principally, we
think, for building dugout canoes.
The distribution of celts in the levels demonstrates that
the earliest inhabitants on this mound brought the knowledge and
use of shell celts with them. Most of the celts were found in
the lowest levels, from 18 to 42 inches below the surface. Not
one celt was found in Level 1, which produced the most pottery
sherds (Fig. 5, a-b).
Busycon tools and vessels parallel the distribution of Strombus
celts found. These objects, made from big welk shells, were
widely used as picks and hammers after piercing and hafting. An
interesting additional use, in the periods before the invention of
pottery, was as cooking utensils. Several shells and fragments of
shell vessels were found with holes burned through the bottom from
exposure to fire. This appears to be a preceramic trait only
(Webster 1970). The 26 worked shells, most of them with drilled
holes, are all worked differently (Fig. 4, c-d). Several Strombus
and other shell scrapers and cups were found.
Macrocallista knives were whole or fragmentary shells of the
large edible clam, Macrocallista nimbosa, a marine species not
common on the east coast of Florida, but very common on the west
coast of Florida, in channels, on islands and on the mainland.
This shell can be used as a knife, and is common on west coast
sites. Five sting ray spines and 14 pieces of sailfish bill were
From Level 4 (18-24") came two Hernando points of tan flint.
These are a little larger than usual for this type. Their time
range is estimated at 400 B.C. to 400 A.D. (Bullen 1975). Their
occurrence in a Transitional level indicates that they may date
from prior to 500 B.C.
WILLIAMS AND MOWERS
0 1 2 3 4 5 6 7 3 9 10 11 12 13
o00 oa6 oea oiL 09l 0i O CI, OEi ia on 00L 06 0o oz
Il ll l l lllll ll 11 1 1111 l11111 11 1 ti ll] 111 11 f|| | i |||ll||| [III |||||||||||||||| (|| | | |II || | | || I || || || ) | |
Fig. 5. Strombus celts from Level 7.
Table 2. Non-ceramic artifacts by levels, all
Artifacts L-l L-2 L-3 L-4 L-5 L-6 L-7 Total
Shark teeth, drilled
Bone socketed points
Bone gouges, awls, etc.
"Foreign" stone tools
Amber (?) bead and mat.
Clay pipe bowl
.9 20 17
5 2 13
WILLIAMS AND MOWERS
Table 2. (Continued)
Artifacts L-l L-2 L-3 L-4 L-5 L-6 L-7 Total
Dogbone with hole 1 1
Glass pendant, facet 1 1
Wood, root 1 1
Wood, worked object 1 1 2
Human cranial bones 4 4 58 9 75
Human teeth 3 11 26 64 13 117
Human mandible 1 1 1 3
Human feces 2 6 11 2 21
Cart. cases 41 3 2 46
Totals 241 460 729 368 177 100 23
Table 3. Contents of Pit Q-21, Level
Human long bones (femur, humerus)
Small limb bones (radius, ulna,
Carpals and phalanges
Miscellaneous unidentified bits
Raccoon jaw bone
Rodent jaw bone, plus loose teeth
Alligator vertebrae, scute
Fish, bird, snake vertebrae
Shark tooth, drilled 1
Bone pin fragment 1
From Level 3 (12-18") came a number of points and fragments.
From pit J-19 came a basal fragment of an eared point similar to
the Simpson type. From pit L-28 came a large fragment of a tan
flint point, base missing, but very much like a Hernando type.
From pit L-29 we took a whole point made of light tan flint
similar to a Levy projectile point, an Archaic point dating from
5,000 to 1,000 B.C. (Fig. 6, a) (Bullen 1975).
Pit N-17 produced a large broken point of whitish chert of
poor material and workmanship, looking like a Hernando type.
Pin N-23, Level 3, yielded a fine Broward point of chert,
similar to one found at Peace Camp site (Mowers and Williams
1972). This point may be dated by C-14 by use of associated
Level 2, pit H-19 produced a clear translucent quartz or
quartzite projectile point or knife. It appears to have been
broken and reworked, giving it a waisted appearance (Fig. 6, d).
Pit M-17 produced a white chert point, a short, side-notched
Archaic type close to Bolen Plain, circa 7,000 B.C.
Although several points found here are similar to Paleo-Indian
or Archaic point types, there is no reason to suggest similar
early dates for occupation of the site. Often crude and reworked
points similar to earlier types appear at Florida sites.
Five exotic stone tools found were: the corner of a ground
blade of light gray igneous rock (pit L-23, Level 4); a small
fragment of a ground stone tool of gray metamorphic rock and of
indefinite shape (pit 1-19, Level 2); a small fragment of a
greenish-gray, ground slate tool (pit L-30, Level 2); and, two
foreign stone fragments of unknown origin and use. Also found
were 14 flint chips and spalls. A fine polished steatite plummet,
a steatite polished celt and a rim sherd from a steatite bowl were
traded into South Florida from the north (Fig. 6, f,g,e).
Projectile points of chipped stone are rarely found on Ever-
glades sites. Markham Park Mound No. 2 is unusual in this respect
because nine stone points were found here. All were found in
Levels 2 through 4, indicating a possible time range from Late
Archaic through Glades II.
Another unusual artifact for the Glades area is the small
microlith drill found in P-17, Level 2 (Fig. 4, a). On examin-
ation of this tool, Dr. Bullen (personal communication) said
that it had a small spur at one time, which was broken off.
The nearest site for this type of flint work is the Nalcrest site
(Bullen and Beilman 1973).
WILLIAMS AND MOWERS
I111111111111111111 i II I1111111 I 11111111 lii I I I I I I I I II l l j
0 1 2 3 4 5 6 7
Fig. 6. Lithic artifacts; a, Levy-like projectile point; b-c,
Hernando projectile points; d, quartz point or knife;
e, steatite sherd; f, pendant: 7, celt.
- -- -- -- I
Deer were plentiful in the Everglades. The potential for
usage of bone to make tools and weapons was unlimited. There
were 207 bone bipoints, 33 socketed bone points, 10 bone gouges
and awls and 82 pieces of worked bone of various kinds found in
this site. The most interesting bone artifacts (Fig. 7, a-c)
were a small incised pin of flat bone, a broken awl or point 7.2
cm. long, a long pin broken in three parts possibly made of turtle
bone (personal communication, William H. Sears, Florida Atlantic
University). Another unusual bone piece is a zoomorphic bird
effigy carved, incised and drilled for suspension. As in other
south Florida sites, bone played a prominent part in the life
of the people.
Several tiny scraps of mica and five small, irregular pieces
of pumice were also found. Pumice, highly prized for grinding
and honing, often floats to Florida shores from volcanic islands
in the West Indies.
Two pieces of clear transparent, crystalline material are
being studied. Since they scratch glass easily, they may be quartz
or even sapphire. The single bead, possibly amber, was found with
many bits of the same material, not known to be a local substance.
A clay pipe bowl of unusual ribbed design was found in Level
1. It probably dates from the 19th century.
Modern artifacts, in addition to the usual cartridge cases,
bottle caps and glass, include six Seminole blue glass beads, a
small rolled up strip of metal, which may be silver, and a silver
bead. All are from Level 1, indicating a historic period occu-
pation. The silver bead is large and crude and may be of Spanish
origin, although reworked by the Florida aborigines (Fig.4). A hexagonal,
faceted, drilled pendant, of water-white material, found in the
first level, is probably a modern glass ornament.
Seeds were quite common in most levels; one kind was identi-
fied, although several shapes were the same as from other sites.
The one species identified, Canna flaccida, "Everglades Bandana,"
was also found growing on the mound. Charcoal was found in all
levels, usually in small amounts, and has been preserved for
possible dating use.
The major feature, indirectly produced by human agency, was
the concretionary layer, or layers, occurring across the mound
at various depths. These hardened deposits are a common-
place occurrence on Everglades sites. The concretion in this
Fig. 7. Bone artifacts; a, incised pin; b, broken bipoint;
c, broken turtle bone pin.
c, broken turtle bone pin.
site is a stony, light gray composition, like that of the
Peace Camp site (Mowers and Williams 1972). The concreted layer
was not continuous across the mound. It was of blackish, very
rough, porous appearance and stuffed with bits of food bones,
garfish scales and vertebrae, shell and pottery, charcoal and
ashes. It is "consolidated garbage" in hard, irregular and
Several strata were found and recorded on pit stratification
drawings. Between the layers, a blackish, granular soil, largely
composed of altered ash, was found. It contained bones and arti-
facts with a coating of the "organic concretion" as we call it.
Samples of this concretion were saved for future study. Some con-
tain pottery, others contain shell tools.
One post mold was found and carefully excavated down. It
ended on a flat stone. The Indian usually dug a small hole, put a
flat stone in it and set his post on the stone. The back fill con-
tained a small piece of Strombus celt, making a carbon 14 dating
possible. This was at a depth below the surface corresponding to
the Transitional period. No house-pattern post molds were observed.
It is believed that the natives living on this mound buried
their dead in the nearby burial mound until about 1 A.D. However,
a number of bone clusters were found in this mound which probably
represented casual interments, possibly of persons dying in transit.
These were in Levels 3 and 4. In no case were these prone burials
of entire bodies. With few exceptions, these bone clusters appear
to be bones collected from corpses after long exposure in the open,
and only the larger bones were usually interred. No complete
skull was found.
In pit 0-17, Level 4, was found a mass of 150 or more broken
bones, ranging from fragments of long bones to small phalanges.
This was not a burial even of the most casual type; it was a
jumble of bone bits, some not even human (alligator, turtle, etc.).
The material was in bad condition, with many fractures and dis-
In pit N-18, Level 4, was found a group of four cranial bones
associated with a few St. Johns Plain sherds. In pit N-23, Level
4,was a mandible fragment with two teeth and a total of eight
cranial bones. Pit Q-22, Level 3, yielded a mandible fragment with
one tooth in place, plus four loose teeth nearby. From pit R-20,
Level 3, came a total of six cranial bones.
Level 3 of pit L-16 contained a maxilla fragment, a mandible
fragment and 11 miscellaneous loose teeth, 33 cranial bones, and
ten long bone pieces, all in poor condition. This was not a skull
destroyed by careless digging, but a carefully trowelled excavation
WILLIAMS AND MOWERS
made by experienced workers. The same pit, at Level 4, produced
another mandible fragment plus two teeth and one cranial bone.
These did not come from the same individual as in the level above,
since both were left mandible fragments, nearly the same parts.
In pit Q-21, Level 4, at a depth of 18 to 24 inches, was
found a great mass of human bones, along with some food bones and
potsherds. These were scattered and jumbled, in no sense the
orderly burial of one or more corpses. These were in poor con-
dition,with broken and rotted joint ends, cracked rib bones, etc.,
and inamong them was a quantity of deer bones, turtle shell, fish
vertebrae, rodent jaw bones and other items. As they were troweled
out, these parts were carefully lifted and packed in plastic bags
and small cartons; some were foil wrapped.
Bone fragments that joined were glued together, but the soft,
rotten condition of the bones made restoration almost impossible.
Parts of at least two adults and one small child were recognized.
A few pieces of the skull of a child were glued together. Among
the 59 human teeth found were 17 from children. Incisors were
typically "shovel shaped"; canines were rather small; molars were
moderately worn down; caries not observed. Taurodontism was fre-
quent. A child's incisor was much worn off on the edge. An
inventory of pit Q-21 appears in Table 3.
Pit R-21, Level 3, adjoining Q-21 and part of the same bone
mass, contained one mandible fragment with two teeth, plus four
loose teeth, one cranial bone and a piece of a long bone.
This mound appears to have been a habitation site starting
soon after the Everglades area emerged from the sea and continuing
until the present. The aboriginal Indian in the Glades area lived
on the tree islands by necessity. There was no other place high
enough to keep their feet dry and their fires burning. The islands
formed by the outcropping of oolitic limestone and occurring
sporadically throughout the Glades area in groups determined the
living pattern. Therefore, a family could, conceivably, live on
several islands or hammocks in the course of a period of years as
the family grew larger.
The earliest inhabitants did build an occasional shelter on
the mound, as shown by the post hole with a flat stone under its
base. Perhaps the area was a hunting ground for seasonal use by
shore dwellers; big bones were plentiful in the lowest levels.
There is no doubt that the first users of the mound were
people who were frequent visitors to the coast because the lowest
levels showing occupancy were those with the most Strombus celts
and Busycon tools and vessels. They arrived before pottery was
used, perhaps just before pottery was invented, or at least known
in this area. Busycon cooking vessels probably were not used
after pottery came into use.
In the very earliest times building dugouts was probably a con-
tinuing occupation. Numerous large, durable shell celts, adze beveled,
were found in the lowest levels to indicate the industry. This par-
ticular site seems to have been on a waterway that was quite popular.
Many visitors over the centuries found their way to it in their dug-
out canoes as the large collection of what are believed to be non-
indigenous potsherds indicates.
Across the mound, between the bottom sterile sand to Glades series
pottery levels, there is an average of about 25 inches of sandy dirt,
with artifacts contained in it. This dirt could hardly have come
from any other source than human transportation from borrow pits over
many years of habitation. Within this 25 inches of fill are found
all of the fiber tempered and semi-fibered pottery, all of the thick
pinky beige Transitional St. Johns ware, most of the Strombus celts
and Busycon vessels and many other artifacts, as well as animal bones
larger than those of the same species found in higher levels.
Higher levels contain almost all of the muck-ware, the fine
sand-tempered ware, and the St. Johns ware. This includes the in-
cised and check stamped decorated pots, almost all of the sand-tempered
Glades gritty ware, as well as the Belle Glades Plain and the various
unclassified wares. Also, in the higher levels is an increased number
of bone and stone artifacts, sharks teeth and vertebrae, and shell
tools except Strombus celts and cooking vessels.
This culture thus appears to be a continuum from the first-
comers, with no major changes in the culture until pottery came into
general use. Even this change was not so very great since the en-
vironment was generally the same, only wetter, and pottery utensils
merely provided better utilization of food resources. It was a
fairly secure life, if somewhat monotonous.
It was a culture of hunting and gathering, essentially the
Archaic pattern with improvements. Even European contact made little
impression on the culture, so far as this site shows.
The unusual number of kinds of pottery, with the variety of
designs found on the pots, all coming together on this tiny island
in a sea of water and grass, testify to the ceaseless comings and
goings of people over a very long time. We might ask where they
were all moving to and why.
This mound was but a stepping stone, not a destination. These
people needed nothing but the contents of their dug-outs to set up a
home. The quest for food furnished part of the motives for travel,
but there is a possibility that the need for meeting together for
religious, social and ceremonial purposes at various seasons dic-
tated travel across the Everglades to other places. Stone artifacts
were trade items, ultimately derived from far-away places, valuable
but not indispensable since the Indians could live comfortably
WILLIAMS AND MOWERS
This site is but one of a chain of small habitation sites
throughout the Everglades. The artifacts found here are generally
the same ones found on all the other sites. The most notable fea-
tures of Markham Park Mound No. 2 are its very early occupation, com-
parable to the Peace Camp site (Mowers and Williams 1972), and its
accompanying burial mound. These are features which in their details
enlighten us as to the earliest inhabitants of the Everglades.
The members of the Broward County Archaeological Society
wish to thank the Broward County Commission, which requested that
we excavate this site, and the Broward County Parks Department
and its employees who gave us assistance in this excavation.
Appreciation is extended to Dr. William H. Sears, Department of
Anthropology at Florida Atlantic University, and to Dr. Ripley P.
Bullen for identifications; to Gypsy Graves for photographs; to
Robert J. Pearsall for excavation and contour maps; to Joan Lynn
for typing the final manuscript; to Marion Henriquez and Norcott
Henriquez, the editorial committee; and, to Nedra Lexow for
assistance in preparing the manuscript.
Grateful thanks is offered to all friends and visitors who
helped dig and screen on occasions. Also to faculty members and
students of the Broward Community College, Stranahan High School
and the Miami-Date Community College, who worked with us on numerous
Finally, our thanks to all those active members who did the
bulk of the work of digging and screening on the mound.
Florida's Prehistory. In Florida From Indian to Space
Age, Vol. 1, Chap XXIII, SouthernPublishing Co.,
Delray Beach, Florida.
1975 A Guide to the Identification of Florida Projectile
Points. Kendall Books, Gainesville, Florida.
Ripley P. & Beilman, Laurence E.
The Nalcrest Site, Lake Weohyakapka, Florida. The Florida
Indian and Spanish Selected Writings. University of
Miami Press, Coral Gables, Florida.
Functional Analyses of Poverty Point Clay Objects. The
Florida Anthropologist 28:57-71.
Strombus Lip Shell Tools of the Tequesta Sub-Area. The
Florida Anthropologist 17:215-220.
Laxson, Dan D.
1970 Seven Sawgrass Middens in Dade and Broward Counties,
Florida. The Florida Anthropologist 23:151-158.
Miami-West Indian Archaeological Society & Broward County Archae-
1975 Arch Creek Site, Dade County, Florida. The Florida
Mowers, Bert & Williams, Wilma B.
1972 The Peace Camp Site. The Florida Anthropologist
1951 A Survey of Indian River Archaeology, Florida.
University Press, New Haven, Connecticut.
A New Concept for Busycon Shell Receptacles. The Florida
Willey, Gordon R.
1949 Archeology of the Florida Gulf Coast. Smithsonian
Institution Miscellaneous Collections 113. Washington,
DISTRIBUTION OF PALEO-INDIAN PROJECTILES IN FLORIDA
Ben I. Waller and James Dunbar
Florida has a tremendous Paleo-Indian cultural heritage.
Hundreds of the lanceolate, Paleo-Indian projectile points have
been recovered throughout the state. When this paper was conceived,
it was to be no more than a map showing the distributional density
ofPaleo-Indian points in Florida. But as the sites were plotted
on a map of Florida a unique picture began to unfold and it became
apparent that several patterns exist regarding the distribution
of Paleo-Indians archeological remains (see Fig. 1).
In our analysis we have chosen to disregard any site where
doubts exist, regarding actual recovery of artifacts, e.g., a beach
on the northeast coast of Florida from which two Paleo points were
recovered. These two projectiles were found in sand which had
been redeposited from an area in Central Florida near the St. Johns
River. We have also disregarded controversial sites where uncon-
firmed reports have indicated that Paleo-Indian materials were
found but which could not be confirmed.
Florida is a state criss-crossed with streams and rivers,
pockmarked with thousands of lakes, springs, and sinks and surrounded
by coastline. As the location of each Paleo-Indian point was pin-
pointed on the map it became increasingly evident that the sites
were located with such a high degree of frequency along rivers,
lakes, and coastal areas that we began to think of Paleo-Indians as
being "aquatically oriented." When sites in areas where no water
presently existed were visited, it became apparent that in the
past the sites had been adjacent to lakes and springs which
today were dry or non-flowing.
A preponderance of sites, too dense in distribution to record
on the map, were located along the Santa Fe River. These sites
appear to extend along the river, from the upper areas to the mouth
where it flows into the Suwanee River. The Santa Fe River, there-
fore, becomes the focal point of attention for the Paleo-Indians in
the state. Hundreds of points, tools and artifacts have been re-
covered from the river including many bone tools. Fossilized animal
remains have been collected in abundance in portions of the river
thought to be kill sites. At least some of these animals were
associated with people. Cutting marks have been found on some of
the bones of the Ice Age animals, probably a result of butchering.
This and the tremendous number of projectiles suggests that people
have killed now extinct Pleistocene animals at the river crossings
of game trails thousands of years ago. The bone must have been cut
when it was fresh because fossilized bone is nearly impossible to
work due to its brittle or hard nature. The possible association of
Paleo-Indian tools and now extinct Pleistocene faunal remains is an
The Florida Anthropologist, vol. 30, no. 2, June 1977
important clue in locating and understanding the distribution of
Paleo-Indian sites along rivers.
One dramatic thing about the distribution is the lack of
siteseast of the St. Johns River toward the Atlantic Ocean. We
have no explanation for this since Pleistocene fossil deposits
have been found in various areas along the eastern seaboard.
Central Florida is built on a unique limestone ridge known
as the Ocala Limerock Ridge. In this region there are literally
hundreds of springs and sinkholes, many of which were used inten-
sively by Paleo-Indians and by more recent Indian groups. The
abundance of early sites in Central Florida compared to Gulf Coast
sites suggests that the movement of people was down the Ocala
Limerock Ridge to the West Coast rather than from the Gulf Coast
to Central Florida. However, many Gulf Coast sites may be inun-
From these data several summary statements can be made:
(1) habitation areas of the Paleo-Indians appear to be along water-
ways throughout Central, North and Northwest Florida; (2) the
Indians, for some reason, did not inhabit the east coast of Florida
in thesame proportions as they did the central ridge and west coast
of the state; and, (3) the tremendous amount of cultural material
present in sites located throughout Central Florida and extending
toward the Gulf West Coast seems to indicate that Paleo-Indians
came to Florida in fairly large numbers.
Fig. 1. Paleo-Indian sites
represented by dots.
SUBSISTENCE SYSTEMS IN THE SOUTHEAST
Elizabeth S. Wing
Although animal remains associated with prehistoric arche-
ological sites may represent only a portion of the subsistence base
of the sites' occupants, they do, nevertheless, indicate human use
of natural resources. In order to gain insight into the patterns
of human use of animal resources and the relative importance of
cultural and environmental factors in this utilization, it is in-
structive to compare faunal assemblages from several standpoints.
In this presentation I have chosen to discuss two methods of com-
parison: sets of the most important species represented in each
faunal assemblage and the variety of species identified in each
This analysis is based on the faunal remains identified from
43 prehistoric sites predominantly in Florida, but also including a
few from Alabama, Georgia, North Carolina and South Carolina. The
sites are, furthermore, located in a variety of habitats and were
occupied in a time range from 2000 B.C. to the historic period. The
samples vary in size from 23 to 1094 minimum numbers of individuals,
but are, on the average, fairly large--215 individuals.
In making this analysis, I have made two basic assumptions.
The first of these is that the faunal samples are not greatly biased
either in the preservation of the remains or in their excavation--in
other words, that the sample approximates the composition of the fauna
used. The second assumption is that the relative abundance of indi-
vidual animals represented reflect their relative importance. This
importance may not be in terms of pounds of useable meat, but, in
general, in time and effort required in procurement.
It is possible to gain insight into the hunting and fishing
strategies of prehistoric southeastern peoples by two approaches to
the analysis and comparison of faunal assemblages. One method is
to define constellations of resources (Tables 1 and 2) or groups of
species that constitute 10 per cent or more of each faunal assem-
blage. The major resources are the predominant animals selected by
the hunters of fishermen, and are, therefore, reflected as important
features of the faunal sample. The minor resources, important in
one or more sites, reflect local dependence on some abundant resource.
In combination with this method, the statistical techniques for
determining species diversity and equitability provide a measure of
the variability in a faunal sample and degree of specialization in
the procurement of animal resources. Species diversity formula is
H' = -EpiLogePi
p is the percentage of individuals of the i species
identified (MacArthur and MacArthur 1961)
Equitability formula is: E = HYH max
H max is the natural logarithm of the number of
observed species (Sheldon 1969)
The Florida Anthropologist, vol. 30, no. 2, June 1977
82 SUBSISTENCE SYSTEMS
Table 1. Constellations of Resources
I. Inland Sites
Terrapene box turtle
Kinosternon musk turtle
Chrysemys pond turtle
Gopherus gopher turtle
Chelydra snapping turtle
II. Coastal Ridge
Chrysemys pond turtle
Trionyx soft-shelled turtle
III. Sea Turtle Harvesting
Cheloniidae sea turtle
Gopherus gopher turtle
Chrysemys pond turtle
IV. Coastal and Specialized Fishermen
Bagre sea catfish
Arius sea catfish
Mergus sea duck
Sciaenops red drum
8 Le 107
Colby Fish Camp
Neuse River, NC
Warren Wilson, NC
Tamiami Trail III
Cumberland Is., Ga.
Period Date I
Ft. Walton +AD 1200
Pre-Cades Pond AD 200-400
St. Johns I +0
Fiber Tempered-St. Johns 1500 BC-AD 500
Early Woodland Kellog
Trans. Glades III
500 BC-AD 1
1000 BC-AD 1000
300 BC-AD 1200
300 BC-AD 1715
300 BC-AD 500
300 BC-AD 1500
300 BC-AD 300
Buck, 8 OK II
Fig Island, SC
Glades I 300 BC-AD 500
Weeden Island AD 300-1200
Post-Trans.-Deptford 500 BC-AD 1200
Weeden Is.-Safety HarborAD 300-1200
Glades I-III AD 300-1200
St. Johns I 600 BC-AD 300
Deptford-St. Johns IIB
St. Johns II -Historic
Early Safety Harbor
600 BC- AD 1500
1380 +200 BC
600 BC-AD 1200
300 BC-AD 300
Sites tabulated according to resource constellation.
The constellation of resources are of two basic types, one
with deer and pond turtle predominant and the other with marine
animals predominant. Each of these two sets can be further divided,
resulting in four constellations of resources that can be distin-
guished. Deer and pond turtle which are common to two sets are modi-
fied by the addition of box turtle and musk turtle in one case, and
raccoon, otter, and soft-shelled turtle in the other. Likewise, the
group in which marine animals predominate, forms two sets, one com-
posed of deer, sea turtle, shark, and sheephead, and the other in-
cludes two genera of sea catfish, jack, and sheephead. As may be
supposed, these constellations reflect regionally abundant resources,
with the first two sets representing the most important resources of
inland sites and the second two sets representing coastal resource
procurement. While the constellation of resources do reflect regional
abundances, at the same time they indicate strong selection for pre-
ferred species, as clearly suggested by the overall importance of deer
and turtles, in general, but especially pond turtles.
By examining the sites and their constellation of resources, we
may gain an understanding of the effect of these two factors, availa-
bility and selection, in the subsistence system. Constellation I,
composed of deer, box turtle, musk turtle, and pond turtle, includes
all inland sites except the Fort Center site. This site is excluded
because it has such high diversity and equitability indexes that no
species represented 10 per cent of the faunal sample. Constellation
II, with deer, raccoon, otter, pond turtle, and soft-shelled turtle,
is associated with coastal ridge sites. It appeared that this group
might have the closest correlation with a particular region, the
Miami Ridge; however, the Cumberland Island site fits perfectly in
this group, suggesting that this constellation is typical of a broader
environmental zone, a zone which, although close to the sea and marine
resources, also provides abundant land and freshwater resources which
were extensively used. The constellation of resources of most coastal
sites include sea catfishes, jack, and sheephead, which were caught by
generalized or specialized fishermen. Material from some coastal
sites show a different emphasis, one largely based on, or including
sea turtle harvesting as an important aspect of a fishing economy,
here considered Constellation III. These sites are located on both
the Atlantic and Gulf Coasts, and are representative of several cul-
tural periods. Furthermore, some of the sea turtle harvesting was
off the grazing banks (Wing 1963) and undoubtedly some was off the
beaches during the egg laying season. Of the four groups, this shows
the least regional integrity, although it must clearly coincide with
sea turtle and shark abundance.
In order to have a clear picture of resource utilization, one
must consider not only the four or five most important animals
selected, but also the variety of resources used. This is especially
true as the overwhelming majority of faunal assemblages suggest that
a very diverse segment of the resources available were, in fact, used.
When the equitability and diversity indices are plotted for each
faunal assemblage (Fig. 1) one can see that the values for all, ex-
cept for a small group of sites, are very high. The group of three
sites which have low equitability and diversity indexes are set
widely apart from the other sites and indicate an economy based on
%% Cvl oWM
- I Inland
- II Coastal Ridge
v m Sea Turtle Harvesting
*KM ... **'
- ]Y Atlantic Coast
....... Gulf Coast
..-.. 2 Specialized Fishing
Fig. 1. Equitability and diversity indices.
*.Nu G SnaPRp ***..
Wils *LP CFC IC .SH* GM*'..
\ *NeR vBL -"--iJI 202 Bilns
^ *8Le .*' I *Mc,..-''
." o 0 OSI v*w
Table 3. Summarized species diversity data.
A. Organized according to constellation of resources
Number of Range of species Mean
sites diversity 1
I. Inland 10 2.07-2.87 2.56
II. Coastal Ridge 10 2.47-3.03 2.70
III. Sea Turtle Harvest 6 2.48-3.20 2.73
IV. Coastal, generalized 13 2.70-3.33 3.03
specialized 3 1.53-1.70 1.63
B. Organized according to time period
I. Early Formative 3 1.67-3.27 2.65
II. Sedentism 11 1.70-3.03 2.64
600 BC-300 AD
III. Agricultural Development 10 1.53-3.12 2.67
300 1200 AD
IV. Late Prehistoric and
Historic 9 2.07-3.33 2.72
specialized fishing. All other sites, both inland and coastal, or
early and late, show a dependence on a wide variety of animal species.
When sites are grouped according to their constellations of resources,
some general trends may be seen (Table 3, A). Most inland and coastal
ridge sites have less diverse faunas than coastal sites. The
coastal sites, when separated geographically into Atlantic and Gulf
Coast sites show the greatest range of diversity. The turtle har-
vesting group shows the least uniformity, reflecting, perhaps, its
lack of regional unity. General as these trends may be, they do show
inherent differences that are not seen when the sites are grouped
according to the time period during which they were occupied and
selecting animal resources (Table 3, B).
As one might expect, both cultural and environmental factors
play a role in resource utilization. Cultural selection of presumably
preferred animals is clearly reflected in the constellation of re-
sources. Deer, for example, is almost universally the most important
resource. Turtles, likewise, contributed in a significant way to
prehistoric diets. The resources available near each site determines
the predominant faunal elements, particularly the minor resources.
The species diversity of the faunal assemblage seems to be linked
more closely to the regional fauna than to any particular cultural
development, at least in the time period being considered. On the
other hand, the low diversity and equitability indices of the rela-
tively rare faunal assemblages, which appear to reflect specialized
fishing communities, are culturally influenced.
I am most grateful to the many archaeologists who have provided
the faunal samples they excavated for us to identify and analyze.
Thanks are also due particularly to Kent Ainslie and Stephen Cumbaa,
who identified some of these samples. Some of this work was
supported by grant G-17948 of the National Science Foundation.
MacArthur, R.H., and J.W. MacArthur
1981 On Bird Species Diversity. Ecology 42:595-598.
Sheldon, Andrew L.
1969 Equitability Indices: Dependence on the Species Count.
1963 Vertebrate Remains from the Wash Island Site. Florida
Back issues of the FLORIDA ANTHROPOLOGIST, if available,
may be purchased from the Treasurer at $6.00 per volume ($1.50
for single and $3.00 for double numbers). Volumes 25:2, pt. 2
(Publication No. 6) and 28:3, pt. 2 (Publication No. 7) are $2.00
each. Volumes 1; 2:3-4; 5:1-2; 6:1, 4; 7; 8; 9; 10; 11:1,2,4;
13:1; 17:1; 18:1; 20:1-2; 24:1; 25:3-4; and 26:1 are out of print.
Copies of the FLORIDA ANTHROPOLOGICAL SOCIETY PUBLICATIONS,
a monograph series, may be purchased as follows: Numbers 5, 6,
and 7 at $2.00 each, Number 8 at $3.00. Numbers 1, 2, 3, and 4
are out of print.
Orders with remittances for any of the above should be sent
to the Treasurer whose address is given on the inside of the front
cover. The Society has no resale discount for dealers or resale
INFORMATION FOR AUTHORS
THE FLORIDA ANTHROPOLOGIST published original papers in all
subfields of anthropology with an emphasis on archaeology. Contri-
butions from allied disciplines are acceptable when concerned with
anthropological problems. The journal's geographical scope, in
order of importance, includes Florida, Southeastern United States,
Caribbean, and related regions.
Manuscripts should be double-spaced and typed on one side
only of an 8 by 11 inch sheet leaving at least a 1 inch margin
on all sides. Contributors are referred to recent back issues of
the FLORIDA ANTHROPOLOGIST 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 "Reference Cited"
section, at the end of an article, do not abbreviate names of
series or journals. Books need full title, publisher's name, and
All illustrative material--line drawings (in India ink) and
photographs--should be included in one numbered series of "Figures"
suitable for a maximum reduction to 6 by 8 inches. Parts of
figures are designated by lower case letters. Figures must be
accompanied by brief descriptive captions. All artwork should be
camera-ready, excepting captions. Figures 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.
Authors should submit the original and one copy of their