Citation
The Terra Ceia site, Manatee County, Florida

Material Information

Title:
The Terra Ceia site, Manatee County, Florida
Series Title:
Florida Anthropological Society. Publications - Florida Anthropological Society ;
Creator:
Bullen, Ripley P
Place of Publication:
Gainesville
Publisher:
Published at the University of Florida
Publication Date:
Language:
English
Edition:
Number 3, 1951
Physical Description:
48 pages, vii plates : illustrations, maps ; 23 cm.

Subjects

Subjects / Keywords:
Antiquities ( fast )
Terra Ceia site (Fla.) ( lcsh )
Antiquities -- Florida ( lcsh )
Florida ( fast )
Genre:
serial ( sobekcm )
bibliography ( marcgt )

Notes

Bibliography:
Includes bibliographical references (pages 46-47).
Statement of Responsibility:
Ripley P. Bullen.

Record Information

Source Institution:
University of Florida
Holding Location:
University of Florida
Rights Management:
Copyright Florida Anthropologist Society, Inc. Permission granted to University of Florida to digitize and display this item for non-profit research and educational purposes. Any reuse of this item in excess of fair use or other copyright exemptions requires permission of the copyright holder.
Resource Identifier:
01936101 ( OCLC )
ocm01936101
Classification:
E78.F6 B85 1951 ( lcc )
913.7591 ( ddc )

Full Text
FLORIDA ANTHROPOLOGICAL SOCIETY
PUBLICATIONS
NUMBER THREE
Site Serrci C^eict Site
Wcmatee (bounty, S^ioricL
RIPLEY P. BULLEN
Published at the University of Florida
GAINESVILLE
1951




3

FLORIDA ANTHROPOLOGICAL SOCIETY
PUBLICATIONS
NUMBER THREE
RIPLEV P. BULLEN
Published at the University of Florida
GAINESVILLE
1951


JOHN M. GOGGIN
Editor
Printed by the University of Florida Duplicating Department
Gainesville, Florida


CONTENTS
The Site PagB
Introduction 5
Boots Point Shell Midden 9
Johnson Burial Mound 10
Walkway 10
Abel Shell Midden 11
Bickel Mound 18
Prine Mound 20
Other Sites on Terra Ceia Island 35
Discussions and Conclusions 36
Bibliography 46


ILLUSTRATIONS
Plates At End
I.Views of the Prine Mound and Abel Midden.
II.Miscellaneous Artifacts from Abel Midden, Taylor
Collection.
III. Miscellaneous Artifacts from Tests in Middens.
IV. Potsherds from Prine Mound.
V.Potsherds from Prine Mound.
VI.Potsherds from Prine Mound.
VII.Potsherds from Prine Mound.
Figures Page
1. Site location map, Terra Ceia Island and vicinity 6
2. Sketch map of Terra Ceia site 8
3. Profiles from Test at Bascoms, Abel midden 12
4. Sketch map of Madira Bickel ceremonial mound 19
5. Profiles from Prine mound 21
6. Excavation plan, Prine mound 23
T ables
1. Test at Bascom's, Abel Shell midden 15
2. Test at Taylors, Abel Shell midden 16
3. Numerical Distribution of Potsherds in Trench E,
by Squares 25
4. Pottery from Prine Mound 30
5. Potsherds from Tests A and B, Perico Island 42
6. Temporal Placement of Sites in the Manatee Region and
Adjacent Areas 44


THE SITE
Introduction
The Terra Ceia site (Fig. 1, 1), consisting of two burial mounds, a
flat-topped temple mound, and extensive shell middens nearly a half mile
in length, has received scant attention from archaeologists. Clarence B.
Moore, who visited the site in 1900, covers his work in three short para
graphs:
In a southerly direction, along the east shore of Tampa Bay,
is Terraceia Island, Manatee Co. On this island (see map), on -
the property of Mr. L. W. Johnson, is a low irregular mound, in
which were found a few flexed burials without artifacts.
About 150 yards N.N.W. from this mound is a shell deposit,
with the usual ridge leading to it.
On property of Dr. L. R. Warren, of Braidenton, Florida, not
far from the other remains on Terraceia Island, is a large ob
long mound running north and south, with the usual graded way
leading to it (Moore, 1900, p. 360, map p. 350).
While Moores map incorrectly locates the site on what is now known
as McGill Island, he is clearly referring to the northern unit of the Terra
Ceia site, the Johnson burial mound, the Boots Point shell midden, and the
walkway between them (Fig. 2, J is the Johnson mound). The oblong mound
is now known as the Bickel mound (Fig. 2, B).
The botanist, John K. Small, gives us the following colorful account of
his 1919 trip to the island:
According to prearranged plans, the following morning found a
motor boat ready to carry us to McGills (Miguel) Bay and Terra Ceia
Island. .
Cactus plants, an abundance of prickly-pears (Opuntia Dillenii
and 0. austrina), came into evidence as soon as we landed on
Terra Ceia Island, but these were not what we sought. While
wandering about in the woods we came upon a cabin, and a girl
who lived there said she thought we would find other cacti further
up on the shore. Also, she volunteered to guide us past the dead
bodies in the trail! We followed. The bodies were there, sure
enough, but only skeletons happily. They were the fossil remains
of aborigines dug from a shell midden in making a drainage ditch.
Terra Ceia Island itself is a vast kitchen-midden, or ancient
artificial shell heap, built up by former inhabitants with their
5


THOMAS MOUND
Fig. 1. Site location map, Terra Ceia Island and vicinity.
6


discarded oyster, clam, and conch shells. The bones we saw,
evidently strongly impregnated with lime, were in a good state
of preservation. The skulls were particularly interesting. Many
of us have been taught that a diet of sea-food, particularly de
velops the brain. Now these ancient people lived largely, per
haps wholly, on fish and shell-fish; but from what we know
of their history they were not intellectual giants. Their diet did
evidently develop their heads, however. The skull-bones of the
specimens we observed varied from one-half an inch to nearly
three-quarters of an inch in thickness.
After this venture in archaeology, we broke into the ham^
mock at several points and soon found our prize, a tall columnar
night-blooming cactus with stout stems growing to a height of
twenty feet (Small, 1921, pp. 35-36).
After finding Smalls account I wrote Mrs. James W. Kissick of Terra
Ceia, a native of the island, who is familiar with and interested in its his
tory. She kindly contacted Mr. A. C. Johnson who has lived beside the In
dian site for over fifty years (son of Mr. L. W. Johnson mentioned by Moore)
and Mrs. E. Cliff Abel, who has lived on the shell midden since 1910. Mr.
Johnson says the only landing or dock on Miguel Bay in 1919 was that of
the Abels. Hence, Small if he came via Miguel Bay as he writes, must
have landed at that dock. The cabin mentioned by Small was then occupied
by a family named Hester who had four children of whom two daughters
were 10 and 12 respectively in 1919. Smalls ditches were scoop trenches
made by shell dealers who had purchased shell from the Johnsons. These
ditches were located on the landward side of the shell ridge and are said
by Mr. Johnson and Mrs. Abel to have turned up many burials.
Dr. M. T. Newman included Terra Ceia in his rapid survey of Manatee
County, made in 1933-1934, but did no work there. He recorded the presence
of the walkway connecting the Johnson mound with the Boots Point shell
midden,^ previously reported by Moore.
Other than these three brief notes, Terra Ceia does not seem to be men
tioned in archaeological literature. Historians have, however, suggested the
site might be the location of Ucita, an Indian village near DeSotos landing
place(Swanton, 1939, pp. 135-137; Bickel, 1942, pp. 70 75, 77, 86). Argu
ments against this hypothesis will be developed later.
Paucity of information in print regarding the site at Terra Ceia does not
correlate with the amount of activity which has gone on there. Both burial
^Willey, 1949, p. 173; Willey refers to the shell midden at Boots Point as the Johnson
shell mound.
7


Fig. 2. Sketch map of Terra Ceia site.
8


mounds and about 80 per cent of the shell middens have been removed. A
misguided treasure seeker spent untold hours, working after mid-night and
only when the moon was right, to dig an extremely large and unproductive
hole in the Bickel mound (Fig. 4). The site has been investigated by num
erous amateur archaeologists, collectors, and curiosity seekers.
Through the generosity of Mr. Karl A. Bickel of Sarasota and Mr. and
Mrs. R. H. Prine of Terra Ceia, ten acres of Terra Ceia was given to the
State of Florida as a historic memorial to preserve the oblong ceremonial
mound and to be known as the Madira Bickel Mound State Monument (Fig.
2)* At that time it was not known that the base of a burial mound was also
on this property. These mounds will be referred to as the Bickel and Prine
mounds respectively (Fig. 2, B and P). They and the extensive Abel shell
midden, an isolated deposit of shells and, probably, a walkway connecting
the Prine mound and the Abel midden, form the Southern part of the Terra
Ceia site (Fig. 2, A is the Abel midden).
Terra Ceia is a low peninsula converted into an island by tidal Terra
Ceia River which connects Terra Ceia Bay with Bishop Harbor, an extension
of Tampa Bay (Fig. 1). The island is low and no portion of it, except Indian
mounds and shell deposits, exceeds an elevation of 10 feet. Completely
fringed by mangroves, except along the eastern shore where there has been
a substantial amount of filling, inner portions of the island are extremely
rich and valuable farm land. The area so used, however, is considerably
limited by innumerable lime sinks (Fig. 2). Land is so productive and so
low that in places farmers have built dykes to keep out sea water. The vast
amount of water has an ameliorating effect on the climate so that Terra Ceia
is practically frost free. Vegetation is consequently more tropical than that
of other land in the region.
Fish and shell fish are, or were, plentiful. The secret of the location of
the large site on the western shore of Terra Ceia Island probably lies in
Miguel Bay. Warm, shallow water of this extensive bay supplied Indians an
abundant source of shellfish.
BOOTS POINT SHELL MIDDEN
Mr. B. A. Johnson, brother of Mr. A. C. Johnson mentioned earlier, says
that originally this shell midden (Fig. 2) covered about 214 acres and was
approximately 20 feet high. Today, the base and a small portion, about 3 feet
high, of the northeastern edge, are all that remain.
Investigation of the edge produced sherds, the tip of a bone awl or pin,
a fragment of cut deer bone, and various deer, turtle, and fish bones includ
ing those of the drumfish. Pottery consisted of 55 sand-tempered plain, 1
9


Belle Glade Plain, 12 Pinellas Plain including 2 rim segments with notched
outer lips, and 2 Ft. Walton Incised (PI. Ill, A-B) sherds.^ This collection,
while small, clearly indicates occupancy during Safety Harbor times, circa
1450-1650. Most of the sand-tempered plain sherds were very thick and un
like the others. They may represent local ceramics on a Safety Harbor time
horizon or may date from an earlier period.
JOHNSON BURIAL MOUND
Moore, who worked at this mound (Fig. 2, J) in 1900, mentions it as
a low irregular mound, in which were found a few flexed burials without
artifacts (Moore, 1900, p. 360). Mr. Johnson, who levelled the mound in
1930, says it was about 3 feet high and approximately 60 feet east-west and
90 feet north-south. In doing this work he found hundreds of skeletons but
apparently no grave goods, Only broken up pottery, none as big as ones
hand. The mound was composed of fine, powdery sand and the sherds were
apparently part of the fill. Bones were located in the lower part of the mound
from a depth of about 1% feet downward into the water table on top of hard
pan (clay?). Mr. Johnson says that flexed burials predominated with some
times an extra skull near the knees. This account agrees well with Moores
report.
Mr. Johnson also quotes his father as saying that in 1893, when they
first saw the mound, there was a rectangular depression, about 6 by 8 feet
across, near the top, lined with large conch shells. Apparently this had
disappeared by 1900 when Moore worked at the site.
We did no work at this mound. Examining the present surface failed to
disclose any sherds but pieces of shells were noted within the area of the
mound. While burials have been removed from the western edge of this mound
in recent years, we made no attempt to locate any as the opportunity to do
controlled excavation seemed extremely limited.
WALKWAY
Both Moore and Newman mention the walkway which connected the
Johnson burial mound with the Boots Point midden (Fig. 2). We traced this
walkway for 330 feet in a straight line. Allowing for gaps at the ends we
would estimate the distance between mound and midden to be 420 feet, which
approximates Moores figure of 150 yards.
The walkway was built of typical midden material, black dirt and shells.
In it we found a few thick, sand-tempered sherds like those mentioned earlier
for the'Boots Point midden. The walkway looks like a narrow causeway,
about 10 ieet wide, which extends about 18 inches above the surrounding
n
Willey, 1949, pp. 364-494, gives descriptions of pottery types used in this paper.
10


land. Mr Johnson says that the land around this walkway consisted of sand
or mud flats in 1893. Today it is black mud or muck supporting mangrove
trees.
A small test was dug in the western side of the walkway. This revealed,
at a distance of 15 feet from the center line, 8 inches of mud and mangrove
roots over 1% inches of shell and black dirt below which was mud to an un
determined depth. Evidently 8 inches of mud and mangrove roots have ac
cumulated since the feather edge of the walkway was deposited. This infor
mation is not very helpful at present but it seems worth recording as it might
have significance if a similar test were made 25 or 50 years hence. It also
shows that a substantial portion of the walkway is underground and that it is
larger in volume than present appearance would indicate.
Finding of mud beneath the walkway seems to support the theory that
such walkways were built to jrovide ready, dry access to burial mounds.
Construction of this walkway represents the collecting, moving, and deposi
tion of about 12,000 cubic feet of material. Expenditure of so much labor
implies that burial mounds were visited fairly frequently, more often than if
visits were limited to an occasional burial. Possibly ceremonial processions
visited burial mounds at regular intervals.
ABEL SHELL MIDDEN
This extensive shell ridge is locally referred to as the Abel Mound in
honor of Mr. and Mrs. E. Cliff Abel whose home was the first built upon it
(1910). The ridge or midden originally extended some 1650 feet along the
shore of Miguel Bay and had a height of 12 to 15 feet. Its width varied from
225 to 450 feet judging from surface indications. The bay side was reasonably
regular with landward extensions representing the variations in width (Fig. 2).
Similar landward extensions are illustrated by Walker (1880, p. 420) for a
shell midden at Shaws Point. Could each of them represent the midden of a
separate residence?
About 85 per cent of the Abel midden has been removed for building proj
ects. This includes all the western half and the northern and southern ends.
An idea of its size can be gleamed from the fact that the remaining portion
supports four good-sized cottages with abundant space between them.
Two stratigraphic tests were dug at a location approximately indicated
by T in Figure 2. The first, 5 by 10 feet, was located 38 feet southwest
of the southwest corner of the residence of Mr. and Mrs. Harry L. Bascom,
to whom appreciation is due for permission to excavate in their front yard.
This test was carried downward until, at a depth of 9 feet, it became ex-
11


tremely difficult to throw debris out of the hole. From this elevation a much
smaller hole was dug 3 feet deeper at which depth dirt was mixed with the
shells and the deposit was slightly damp. This was not, however, the bottom
of the midden. The north and part of the west profiles of this test are given
in Figure 3. Specimens are illustrated in Plate III and listed in Table 1.
Fig. 3. Profiles from test at Bascoms, Abel midden.
Profiles indicate four habitation zones in the upper 5 feet (Fig. 3). These
zones could be seen on all four walls of the test. Corresponding peaks in
frequency of specimens occur in Table 1 at depths of 6-12, 24-30, 36-42
inches, and, to a lesser extent, 48-54 inches. At greater depths were whole
shells nearly devoid of other debris except for the zone of l>rown sand at a
depth of 90 inches. This brown sand covered all of the eastern two-thirds of
the test. Specimens found in it are recorded in the 87-93 inch level while the
84-87 inch level also includes specimens from the western third of the test
down to a depth of 96 inches.
The brown sand deposit varied from 54 to 3 inches in thickness and sloped
downwards towards Miguel Bay. In addition to the sherds, Strombus hammer,
and food hones listed, it contained a few shells and some charcoal. Presum
edly, this sand was purposely deposited by Indians. If it represented a house
floor, occupation was not very heavy.
Other evidenoe, suggestive of houses, consisted of three post holes, two
on the south and one on the east wall of the test. One, varying from 3 to 7
inches across, lead down from the surface zone of black dirt and shell. An
other, 5 inches in diameter and 10 inches vertically, lead down from the
second zone of shell and gray dirt. The third, 4 inches in diameter and 9
inches vertically, started at the occupation zone evidenced by the layer of
black-brown dirt and shell. While these post holes all lead downward from
different zones, they are evidence of some type of construction, presumedly
houses, built by Indians on the shell ridge.
12


As the test at Bascoms did not reach the, base of the Abel midden, an
other test was dug about 17 yards further to the southwest on land of Mr. and
Mrs. Ivan Taylor whose permission to excavate is appreciated. Here, approx
imately 9 feet of midden had been previously removed. Consequently, it was
believed this test, 5 by 5 feet, would supplement the one at Bascoms and
continue it to the base. Specimens from this second test are recorded in
Table 2. A picture of the face of the midden near this test, showing occupa
tion zones and deposits of ash, is included in Plate 1 (lower).
The test at Taylors happened to be located where there was a large fire
pit. The excavated portion of this pit measured 3/4 feet horizontally with a
maximum vertical dimension of 2% feet. It was filled with black dirt and
debris, which was divided into an upper and lower half by an ash zone about
5 inches thick. As a result of penetrating this pit a large sample from the
lower part of the midden was secured. Even though most of the specimens
came from a pit they must represent an early period of the midden at this
point due to the 9 feet of shell and debris which previously covered it. Spec
imens shown in Table 2 for a depth of greater than 30 inches were found
lower than the base of this pit.
At the base of the midden were small shells and fine crystals of calcite
(CaCOg).^ The latter are believed to be a secondary deposition derived from
disintegration of shells. The former include shells of Anomalocardia cunei-
meris, Cardita floridana, and Cerithium sp. From the presence of these shells,
of waterwom pebbles formed of fossil bone, and of some sand, mixed with a
marly-mucky deposit, it is believed the midden at this point was built on a
waterlaid beach or tidal flat. As the base of the midden was 14 inches below
water, details of its junction with the underlying deposit could not be ob
served.
It would be methologically incorrect to argue that data from the small
tests at Bascoms and Taylors adequately represent the Abel midden. Prop
er sampling cannot be done but another test a hundred feet to the north could
be made and would be desirable as a check. In the absence of such data it
is necessary to use that from available tests but it should be remembered
that the sample is probably inadequate.
As will be noted from Tables 1 and 2, the vast majority of sherds from
these tests were undecorated and sand-tempered. All of these sherds might
be classified as Glades Plain^ but many are indistinguishable from undec
orated sherds of the Deptford, Santa Rosa-Swift Creek, Weeden Island, and
3
Kindly identified by Mr. Herman Gunter, Director, Florida Geological Survey, Talla
hassee, Florida.
4
Additional information on Glades area pottery types will be found in Goggin, 1950-
13


Safety Harbor complexes. Limestone-tempered sherds from this test contain
some sand-temper, have a lumpy surface, are gray in color, and exhibit
small holes, Some of these holes are suggestive of fibrous tempering, some
may be the result of leaching of fine particles of limestone.
The three miscellaneous incised sherds are decorated with fine, parallel
incised lines. All are similar to Miami Incised but too small for positive
identification. One exhibits four parallel lines between undecorated areas, a
typical Miami Incised motif. The sherd of Biscayne Plain is from the neck
of a vessel shaped like a water bottle.
Sherds from tests at Bascoms and Taylors are not definitive as culture
period markers but they do show differences with depth which have implica
tions of chronology. Goggin, for the Glades area to the south of Terra Ceia,
has postulated an early ceramic period, Glades I (the earliest for that area),
characterized by undecorated, sand-tempered pottery (Goggin, 1950, pp. 244-
245; see also Table 36, p. 233). The lowest zones at the Abel midden would
support Goggins hypothesis. Likewise, levels at a depth of about 6 feet,
which produced incised pottery similar to Miami Incised, would seem to cor
relate with Goggins Glades Ha period to the south (Goggin, 1950, p. 245).
These periods he correlates respectively with the Deptford and Santa Rosa-
Swift Creek periods of the Florida northwest coast (Goggin, 1949, p. 16;
Willey, 1949, Fig. 76). For the Manatee region, where Terra Ceia is located,
Willey (1948, p. 217; 1949, p. 185-193) uses the term Perico Island for the
equivalent of these periods.
Presence of Biscayne Red suggests the 1% to 3/4 foot zone to represent
a Weeden Island time period. Limestone-tempered pottery at the same and
greater depths would not argue otherwise as many limestone-tempered sherds
were found in the nearby Prine mound (Fig. 2, P) associated with Weeden
Island ceramics. The top VA feet of .the midden would then, presumedly,
represent a post-Weeden Island or Safety Harbor time period.
These correlations, while admittedly tenuous, suggest occupation of the
Abel midden over a period of possibly 800 years.^ Even if one is not willing
to accept these correlations, changes in pottery with depth at the Abel mid
den clearly indicate occupancy over a long period of time. This implies a
relatively small population over a long time, rather than a large population
for a relatively short time to account for the vast accumulation of shells. A
moments thought will indicate this must be the case as collection of both
Goggin, 1949, p. 17, and Willey, 1949, opposite p. 574, would place the end of
Glades I at circa 800 A.D. A middle of the period date for Safety Harbor in the
Manatee area would be circa 1600.
14


Table 7. TEST AT BASCOM'S, ABEL SHELL MIDDEN
Depths
Plain,
Pottery
Plain, Misc.
Bis-
Bis-
Other Material
Strom- Other Rocks Frag.
Food
in
sand-
lime
in
cayne
cayne
bus
Shell
bone
bones
inches
tempered
stone
cised
Plain
Red
ham
tools
pins
tempered
mers
0-6
132
1
3
5
58
6-12
526
13
4
10
1
188
12-18
64
1
2
2
3
1
24
18-24
135
1
1
1
1
4
1
58
24-30
340
3
3
5
4
9
1
71
30-36
37
5
3
1
2
2
32
36-42
323
107
2
10
1
12
6
101
42-48
54
23
5
3
14
48-54
64
47
3
2
25
54-60
31
26
1
5
16
60-66
25
12
1
4
3
24
66-72
7
2
1
1
2
7
72-78
25
6
1
1
1
29
78-84
21
6
1
4
16
84-87
16
2
3
2
10
87-93
12
1
3
93-108
42
4
2
2
1
23
108-144
36
25
Totals
1890
244
3
1
6 56 17
69
13
724
shellfish and fire wood must proceed at a rate commensurate with the ability
of nature to make replacements.
Strombus hammers (Pi. Ill, M) were the most common tool at the Abel mid
den.^ Examination of Tables 1 and 2 indicates these tools to have been
found in practically every level of both tests. Columella chisels (Pi. Ill, Q)
also show a long life as two came from the test at Taylors and one from the
24-30 inch level at Bascoms. Four perforated Pecten shells (PI. Ill, H) were
found, two between depths of 6 and 18 inches, one in the 36-42 inch level,
and one in the 66-72 inch level.
For typology of shell tools, see Goggin and Sommer, 1949.
15


Table 2. TEST AT TAYLORS, ABEL SHELL MIDDEN
Pottery
Other Material
Depths
Plain,
Plain,
Good-
Strombus
Colum
Frag.
Food
in
sand-
lime
land
hammers
ella
Rocks
bone
bones
inches
tempered
stone-
Plain
chisels
pins
tempered
9* plus
0-6
127
1
12(?)
1
2
1
37
6-16
72
m
2
2
8
16-24
41
1
1
3
12
24-30
34
1
1
2
3
30-36
12
1
5
36-42
4
42-58
15
2(?)
1
3
1
11
Totals
305
1
16(?)
6
2
13
2
76
Other shell tools concentrated in the upper three feet of the midden.
This concentration may be a function of the habitation zones of the upper
levels or may indicate these tools to be relatively late types. The 0-6 inch
level produced a Busycon celt or spoon (PI. Ill, N), a fragment of a
Busycon hammer and of a Fasciolaria hammer; the 6-12 inch level an Oliva
bead (Pi. Ill, J) and a fragment of a Busycon pick; the 24-30 inch level a
perforated Noetia shell; and the 30-36 level a Busycon hammer, Type A
(PI. Ill, P). Four Venus anvils (Pi. Ill, 0) were found between depths of 12
and 30 inches and a perforated Noetia shell between 24 and 30 inches.
Venus anvils, as the name implies, exhibit shallow pits on their outer sur
faces caused by hammering or pounding. One had a chipped edge. In the
6-12 inch level was a large Busycon perversa shell with a 1% inch hole cut
through the wall near its mouth. Such shells are sometimes referred to as
anchors. This one was filled with Modiolus shells.
While concentrating in upper levels, fragments of bone pins or awls were
also found in the lowest zones (Tables 1 and 2). Not included in the tables
is the carved and incised head of a pin (PI. Ill, I). It came from the 36-42
inch level at Bascoms.
A projectile point (Pi. Ill, L) was found in the 6-12 inch level and a
plummet made from a sherd (PI. Ill, K) in the 78-84 inch level, both at
Bascoms. The point is typical of others in local collections from the midden.
Other items include burnt clay in the 6-12 inch level, an antler tine in
16


the 24-30 inch level, burnt clay and the tip of a chert point or knife in the
48-54 inch level, two pieces of ochre in the 60-66 inch level, burnt s4ndy
clay containing shells in the 66-72 inch level, and parts of two deer antlers
in the 84-87 inch level. Various wire nails and the blade of a hacksaw were
found in the top 6 inches.
The vertical distribution of rocks and food bones has been indicated in
Tables 1 and 2. Rocks include eleven pieces of chert as well as fragments
of sandstone, limestone, marl, and fossil bone (Manatee ribs). All had general
distribution vertically. A mineralized astragalus of a camel was found in the
6-12 level at Bascoms.
Food bones comprised those of deer, turtle, and fish (including drumfish
and rays) plus a few bones of unidentified birds and nine of the alligator. The
latter came from the test at Taylors as well as the upper zones in Bascoms.
A crabs claw, a bears tooth, and the jaw of a rabbit complete the inventory.
All manner of shellfish were consumed at Terra Ceia. The collection in
cludes Busycon perversa, Linne; Busycon pyrum, Dillwyn; Fasciolaria gigan-
tea, Kierner; Fasciolaria tupipa, Linne; Fasciolaria distorts, Lamarck; Cardita
floridana, Conrad; Macrocallista nimbosa, Solander; Melongena corona, Gmelin;
Modiolus demissus, Dillwyn; Murex pomurn, Gmelen; Noetia ponderosa, Say;
Ostrea virginica, Gmelin; Ostrea cristata, Born; Pecten irradians, Lamark;
Spisula solidissima similis, Say; Strombus pugilis, Linne; Strombus (pugilis)
alatus, Gmelin; Cardium (Trachycardium) isocardium,hinne; Venus campechi-
ensis, Gmelin; one Vermicularia spirata; Polinices duplicata, Say; Sinum
perspectivium, Say; and a few shells of the land snail Euglandina rosea,
Ferussac. Of these Ostrea cristata, Spisula, and Cardita were noted only in
the lower zones.
Supplementing excavated specimens from the Abel midden is a collection
made by Mrs. Peter Anemeat (daughter of Mr. and Mrs. Abel) and one made by
Mr. Taylor while working around his home. Mrs. Anemeats collection includes
2 Busycon dippers, 3 columella chisels, 2 columella gouges, 7 columella
plummets (5 grooved, 1 knobbed, and 1 double ended), 3 columella hammers,
4 flat sharpening stones of fossil bone, 5 shell celts (4 thin and concave,
like spoons), a double ended bone point, a broken plummet of chert, 33 pro
jectile points, pottery, and a fragment of charred netting with % of an inch
square holes, knotted at the corners. Projectile points, all of chert, include
2 narrow triangular, inches long, like those found at the Safety Harbor
site; one double basal notched, 3 inches long (Griffin and Bullen, 1950, pi. 2);
9 broad, side-notched, varying from % to 1% inches in length (similar to Pi. II,
N-P); 3 leaf-shaped with rounded bases; 4 stemmed with rounded or sloping
tangs; 2 stemmed with barbed blades; 2 miscellaneous stemmed; 1 rhombic-
17


shaped; and 2 corner-notched points.
Pottery in Mrs. Anameats collection includes plain sand-tempered,
Biscayne Check Stamped, Belle Glade Plain, and Pasco Plain sherds.
Sherds in Mr. Taylors collection and from the talus near his home include
11 Biscayne Red, 10 Perico-Pasco Plain (limestone-tempered), 65 plain sand-
tempered, and 3 with contorted paste. All sherds are typical of pottery, to be
mentioned later, from the Prine mound.
Other specimens in Mr. Taylors collection include many Strombus ham
mers, 7 Busycon hammers, type A, 3 shell celts or spoons, 10 columella
plummets (one knobbed, seven grooved, and two expanded center) (PI. II,
A-E), 2 flat sharpening stones, 5 spatulate bone tools (PI. II, G, I),
1 fish hook (PI. II, J), 1 bone awl (Pi. II, H), 9 double-pointed bone pins,
28 fragmentary bone pins, 8 pieces of worked bone, 4 side-notched and 5
stemmed points with expanding tangs (PI. II, M-R).
Mr. T. A. Fetter, who lives at the northern end of the Abel midden has
found several expanding base and side-notched projectile points, plain sand-
tempered pottery, and parts of two burials, while working around his home.
These must have come from lower levels as all except the lowest zones
have been removed in the area where his house is situated.
It is evident from these collections that side-notched points and those
with expanding tangs, Strombus hammers, columella chisels, columella
plummets, and double ended bone points are the most common artifacts
other than pottery from the Abel midden.
BICKEL MOUND
I
The oblong mound mentioned by Moore is now known as the Bickel
Ceremonial Mound and is part of the Madira Bickel Mound State Monument
(Fig. 2, B). This mound was surveyed and the results are presented in
Figure 4. To the southwest of the center of the mound, the large disturbance
made by a misguided treasure seeker will be noted. The smaller crater, near
the center of the top, bears mute testimony of a similar attempt.
The Bickel mound is about 170 feet northeast-southwest and 100 to
115 feet northwest-southeast at its base. Its flat top was originally about
70 feet long by 25 feet wide and situated 20 feet above the surrounding land.
A ramp, the lower part of which was removed years ago, extends west-north-
west and points towards the village on the Abel shell midden.
A small test, 4 feet square and 4 feet deep, was dug in the bottom of
18


w
Fig. 4. Skotch map of Madira Bicksl ceremonial mound.
the centrally located treasure seekers pit to secure a cultural sample from
the mound. This work produced 21 plain sand-tempered sherds, 3 deer, 22
turtle, and 15 fish bones, a fragment of red and yellow ochre, a Strombus
hammer, charcoal, and ash. One sherd had a surface suggesting a brushed
treatment. Examination of the spoil from the large treasure seekers hole
located two more plain, sand-tempered sherds, another Strombus hammer,
and a deer bone.
The Bickel mound was constructed of shell, sand, and village debris.
Mr. B. A. Johnson advises that Mr. Rowell, who dug the large hole in this
mound in 1919-1920, informed him that the mound from top to bottom con
sisted of alternate layers of shell and sand with which was mixed some
pottery and nothing else. No doubt Mr. Rowell would not have mentioned
food bones or shell tools.
Undoubtedly, the Bickel mound was built as a platform for a ceremonial
building a temple, priests or chiefs home as were other mounds of this
19


type in the southeastern part of the country. It is also likely it was built in
a series of stages, not all at one time.
The mound fill, consisting primarily of village debris, probably was
brought from the Abel midden or from the more handy shell midden situated
between the Bickel mound and the Abel midden (Fig. 2). Land on three sides
is low and swampy but no borrow pit is now in evidence.^ Land to the west
is slightly higher and rather sparsely covered with weeds. Many pieces of
marl appear on the surface which is barren of cultural remains. The south
easterly extension of the Abel midden covers the extreme westerly side of
this field while the basal portion of a small shell midden borders it to the
north (Fig. 2). Sand and also village debris may have been scrapped from
this field in the construction of the Bickel mound.
It may also be mentioned in passing that this reasonably hard surfaced,
level field would have been an ideal location for a ball court or for spectators
participating in ceremonies held at the Bickel mound. A similar area, nearly
denuded of cultural remains, was found between Mounds 2 and 4 at the Lake
Jackson site near Tallahassee (Griffin, 1950). As we learn more about these
large temple mound sites, such plaza-like areas may assume greater importance.
PEINE MOUND
The Prine burial mound (Fig. 2, P) is located a short distance north
of the Bickel ceremonial mound on land of the Madira Bickel Mound State
Monument. As has been mentioned earlier, the existance of this burial mound
was not known to the Florida Park Service at the time of the dedication of
the monument.
Mrs. James W. Kissick, as a girl, assisted Mrs. Isaac Craft of Tampa
in uncovering skeletons near the top of this mound around the turn of the
century. These skeletons were not removed but carefully examined and then
reburied. Mrs. Kissick describes the mound as being built of sand with no
shell admixture and as extending about 100 feet north-south and somewhat
less east-west. She remembers all skeletons examined by her to have been
extended on the back and to have had no grave goods.
The Prine mound has been substantially damaged since Mrs. Kissicks
early visits. In 1914 much of it was removed for road fill. A workman who
was present said that many human bones were found in the upper part at
that time. He did not remember the position of the bones or know about
7
This lower elevation is believed to be a natural feature connected with the large
pond to the northeast of the mound (Fig. 2). Nevertheless, it may to some extent
be the result of aboriginal activities.
20


pottery but was certain that no beads were found. Presumedly, these were the
burials examined by Mrs. Kissick.
Subsequently, the present road was built along the eastern side of the
mound. Burials are said to have been found at that time but no information
regarding them is available. About 20 years ago, Mr. J. E. Moore of Sarasota
and Mr. Montague Tallent of Manatee dug over the top of the remaining part
of the mound finding a small amount of pottery.^ Slightly later the remaining
portion was leveled and three small cabins built for field hands. These were
occupied for some years during which numerous small pits were dug by the
inhabitants.
At a still later date, Mr. William C. Chadeayne of Bradenton dug through
the eastern half of the remnant of the Prine mound. He has kindly given the
sherds he recovered to the Florida Park Service and an inventory of them
has been included in Table 4. Mr. Chadeayne found no burials but did not
dig to the base of the mound so that burials may be in the lowest zone of the
area dug over by him, as was found for the western part of the mound.
Today the Prine mound is represented by a low sandy knoll, about 100
feet across, which rises 18 inches above the somewhat swampy surrounding
land. Excavation revealed the mound to be composed of gray sand with which
was mixed a small amount of charcoal and which rested on a thin black-
brown zone. Below was impure marly clay. The gray sand and black-brown
deposits may be seen in Plate I (upper).
Fig. 5. Profiles from Prine mound.
8
Mr. J. E. Moore, Sarasota, Florida personal communication.
21


Over part of the excavated area, particularly to the south, the top 6-9
inches was heavily impregnated with charcoal (Fig. 5). This zone, and small
pits leading down from it, contained numerous wire nails, other metal, pieces
of glass and of recent ceramics, battery electrodes, and a United States
twenty-five cent piece (dated 1905) as well as sherds of Indian pottery. This
admixture is the result of 20th century occupation by field hands.
Otherwise, vertical zones in the gray sand fill of the mound could not be
delineated. This sand became lighter in color with depth until, just above
the black-brown zone, it became brownish. On the average, gray sand was 30
inches thick and contained pottery and occasionally shells from the present
surface downward to a depth of 24 inches. Bundle burials were found in this
pottery producing zone while flexed burials occurred at the base of gray sand
in the black-brown zone and were separated from the pottery by sterile gray
sand. Presence of primary extended burials in the upper, previously removed,
part of the mound, as stated by Mrs. Kissick, would indicate three superim
posed modes of burial at the Prine mound.
The black-brown zone below the mound consisted of two parts. The upper
portion, actually the base of the mound, was composed of brown sand, semi-
cemented by iron salts deposited by rain water seeping through the mound.
This transition zone became darker with depth. The lower portion, or black-
brown zone proper, was extremely compact, composed of marly clay-sand,
and contained root casts, and numerous beach pebbles of fossil bone in
cluding horse teeth and fragments of turtle shells. The suggestion on the
profiles (Fig. 5) that this black-brown zone continued to the south and under
lay the surrounding swampy land was confirmed by tests. This zone, at least
in part, represents an estuarine deposit.
The underlying clayey deposit was not penetrated to any great extent
during excavation, ft was mottled gray in color, composed of marly impure
clay, and contained many old root casts leading downward from the black-
brown zone. It did not contain pebbles or fossil bone. Surface of this clayey
deposit exhibited a continuous pattern of octagonal cracks, filled with black-
brown material, suggesting it to have at one time been a dry surface.
Trenches were dug along the 5 and -1 lines (not shown on the excavation
plan) to cut the southern edge of the mound and uncover details of construc
tion. The resultant profiles show the edge of the mound below the 5A and
-1-D stakes (Fig. 5). The normal profile of the surrounding swampy land (Fig.
5, towards the left) consisted of a superior black mucky deposit, rich in
vegetable material, over dark gray sand which rested on the black-brown and
mottled clayey deposits already described.
22


Fig. 6. Excavation plan, Prine mound


With the above information a reconstruction of events at the site of the
Prine mound may be attempted. After deposition of the clayey material and,
probably, after the development of its cracked surface, the area was subject-;
ed to tidal action. Beach worn fossil bone was deposited and a black estua
rine or tidal marsh deposit formed. Plants, possibly mangrove trees, with
roots extending downward into the marly clay became established. Sand of
marine or, more likely, of aeolian origin collected to form the dark gray sand
stratum. Thickening of the dark gray sand at the edge of the mound (Fig. 5,
between stakes -5A and 5a) suggests the presence of a sand dune, later en
larged by Indians to form the mound. Subsequently, the present swampy
plant growth developed and the black surface zone was formed.
It will be noticed on the profiles (Fig. 5) that the dark gray sand is sub
stantially thicker below stake -5B than below stake -1-F. Indians may have
removed part of the dark gray sand, particularly to the southwest, in building
the mound. If so, the black mucky deposit below the stake -5B must have ac
cumulated since the first construction of the mound.
There is no evidence of a prepared mound base. It would appear, below
stake 5A, that the black, swampy stratum had been cut in building the
mound. However, if a sand dune were originally present, as has been sug
gested, this cut would merely represent the accumulation of the black
deposit on the edge of the dune. The dashed vertical line shown at this
point represents a transition zone, not a sharp change in color or texture,
fitting, it seems to me, the dune hypothesis.
Indians either utilized a convenient sand dune for the early flexed
burials or built the core of the mound of sterile sand. Later additions were
made of sand containing sherds and an occasional shell.
Only two structural features were encountered during excavation of the
Prine mound. One was a posthole, 6 inches in diameter, which lead down
ward 7 inches from the disturbed top zone of square 5B. It is probably of
20th century origin. The other may have represented a ceremonial fire. The
excavated portion suggested a saucer-shaped deposit of charcoal, 4 inches
thick near the center and tapering outward a foot from a point 2 feet below
stake 3L.
Specimens from the mound fill, other than skeletal material and sherds,
included fragments of shells, chert, and limestone, and a hammerstone. The
latter, of impure cherty limestone and 2% inches in diameter, resembled a
battered ball. It came from square 51 between depths of 18 and 24 inches
9
The upper gray deposit at this point represents material deposited in leveling and
spreading the mound when cabins were built.
24


but was not associated with a burial. Two large pieces of limestone, 1014 by 9
by 4 and 8 by 6 by 4 inches, were found in square 7E. Both were in the lower
part of the gray sand, more than 2 feet below the present surface. A few
other fragments of limestone were also in the fill.
One Dinocardium robustum, 1 Fasciolaria tulipa, 5 Strombus, 50 oyster,
30 clam, and over 100 fragments of Busycon shells were found in the top
6 inches. Many of the oyster and some of the clam shells were obviously
recent and may be assigned to 20th century occupation. Five oyster and 10
clam shells were found in each of two lower levels while 53 fragments of
Busycon were in the 6-12 inch, 23 in the 12-18 inch, and 11 in the 18-24 inch
levels. These shells were not in pits leading downward from the surface
zone nor associated with nails, etc. Fragments of Busycon shells were fre
quently found in pockets of brown dirt.
Fourteen fragments of chert occurred, ten in the upper 12 inches and the
balance at greater depths. Bits of charcoal and, rarely, scraps of food bones
should also be mentioned.
Sherds at the Prine mound do not represent a ceremonial pottery cache.
This is indicated by both their horizontal and vertical distribution. Horizontal
ly, quantities have been indicated on the excavation plan (Fig. 6). Vertically,
they concentrated near the present surface towards the center of the mound.
As the periphery was approached this concentration tended to occur at greater
depths. This situation was best presented by the E trench from which the
Table 3 has been constructed.
Table 3. NUMERICAL DISTRIBUTION OF POTSHERDS IN TRENCH E,
BY SQUARES
Depth in
inches
2E
3E
4E
5E
6E
7E
8E
0-6
0
43
253
121
26
31
6-12
0
Not
3
91
134
34
86
12-18
0
0
34
87
79
29
18-24
0
dug
0
7
3
190
21
24 Plus
0
0
0
0
54
0
Appare
:ntly the
central
core of the mound
was devoid of sherds.
Sherd
concentration tends to follow the slope of an old surface. Sherds in upper
levels at the periphery (7E and 8E in this case) arrived there as the result
of the 20th century spreading of the mound, coming from higher elevations
no longer present (in the above example from over squares 2E-4E, theoreti
cally).
25


Such a situation could occur if the original mound was a natural sand
dune or if the original mound was built of sterile sand, while subsequent
additions came from a village area. An alternative possibility, that the mound
was used by Indians at some time as a habitation area, does not seem likely.
Occupation as intense as would be required by the vast amount of pottery
present would have left an easily recognized and vertically concentrated
midden deposit.
From the foregoing discussion, it seems evident the fill of the Prine
mound over its original barren core came from a village area. The relatively
small amount of shell is inconsistent with the Abel midden or the nearer and
smaller shell midden as the source of this material. Some of the sherds are
eroded and waterworn, suggesting that beach sand adjacent to the Abel mid
den may have been transported to build the mound. Possibly the barrenness
of the field to the southwest, mentioned in connection with the Bickel mound
may indicate the source of building material.
It is also possible that the sand came from land adjacent to the Prine
mound while sherds and shells were brought from the Abel midden. Such a
theory would, in part, explain the scarcity of decorated pottery at the midden.
It would imply innumerable ceremonial visits over a long period of time.
Although the horizontal distribution of pottery does not support this theory,
it must be considered a possibility as Mrs. Abel and others say there used
to he a walkway connecting the Prine mound with the Abel midden. Apparent
ly, it was not straight like the one at Boots Point but twisted to connect
otherwise isolated small deposits of shell. While no trace of this walkway
can be found today, its approximate location has been suggested on the
site map (Fig. 2).
Pottery from the Prine mound is listed in Table 4 and illustrated in
Plates IV to VII. Pasco Complicated Stamped is the only new pottery type
included in the list. This name has been used for limestone-tempered sherds
otherwise the same as Tampa Complicated Stamped.
The two matching cob-marked sherds (Pi. IV, G) were found 40 feet apart.
They have the smooth, black interior found on some Fort Walton period ves
sels of northwest Florida.
Gainesville Linear Punctated sherds (Pi. IV, H) are all, apparently, from
one vessel made of chalky (Biscayne) paste although none of them match.
They came from the same squares and levels as some of the Pinellas Incised
sherds.
Weeden Island Incised sherds include one typical of the Florida North-
26


west Coast (PI. V, B) and everal from a vessel having incised arcs pendant
to a punctated line (PI. V, C-D). The outturned rim is similarly incised and
bears the large punctate hallmark of the Weeden Island potter. Similarly,
only three sherds are classic Weeden Island Punctated (Pi. V, G-I), while
one vessel bears neatly made punctated arcades above single large puncta-
tions. (PI. V, J). Such designs seem to presage the later Safety Harbor type,
Pinellas Incised A. Several Weeden Island Punctated sherds are from wide,
shallow, basin-shaped vessels with punctated ears or rim lugs and a punctat
ed line on the inside of the vessel a short distance below the rim (PI. V, H,
K). Papys Bayou Punctated sherds bear typical Weeden Island decoration
(Pi. V, L-M).
Check size on Wakulla Check Stamped sherds ranged from medium coarse
to fine (7 to 11 per inch). On Biscayne Check Stamped they were large, 4 to
7 per inch. Pasco Check Stamped was intermediate in this report, 6 to 9 per
inch. That checks were applied both perpendicular to and at an angle with
the rim is shown in the illustrated Biscayne Check Stamped Sherd (Pi. VI, L).
Carrabelle Punctated and Carrabelle Incised sherds represent one vessel
in each case plus an individual sherd of another. The punctated vessel was a
shallow, flattened globular bowl about 6 inches in diameter. (PI. V, N).
Matching sherds of Carrabelle Incised formed the neck of a water bottle (PI.
VI, A). This is not typical Carrabelle Incised and these sherds may be of an
unnamed type of the Safety Harbor or Englewood complexes. They occurred
in squares 6B and 6C between depths of 6 and 18 inches while in squares
4K and 5K they were in the top 6 inches with Pinellas Incised sherds.
Ruskin Dentate Stamped sherds exhibit marks which are nearly square,
about 3mm, wide,^ and made with a tool having six or seven teeth. Ves
sels were large, shallow, simple bowls, 10 to 12 inches wide and 5 to 6
inches deep. One was a globular bowl, 7 inches in diameter and 4% inches
deep. Included under Ruskin Dentate Stamped for the 6-12 inch level are
twelve sherds of a small, limestone-tempered vessel, 5 inches in diameter
and 4 inches deep, bearing irregularly placed rows of closely spaced semi
lunar punctations (PI. VII, A).
One Hillsborough Shell Stamped vessel, decorated by impressing with
the back of a Cardium (cockle) shell, has been partially restored. It proved
to be about 8 inches in diameter by 6 inches deep and to have a small rim
These sherds are the same as some illustrated by Willey as Ruskin Dentate Stamped
(Willey, 1949, PI. 37, a) but he does not include such large punctations in his
definition.
* ^Willey, 1949, p. 441, suggests a tool with four or five teeth.
27


adorno and, apparently, a prefired kill hole. The adorno is illustrated
(Pi. VII, D) although the eroded surface of this part of the.vessel does not
show the shell markings, the opposite rim segment was not found. St. An
drews Complicated Stamped sherds indicated a small bowl with slightly
converging orfice about 8 inches in diameter (PL V, 0).
Three unclassified groups of sherds have been called Unique Incised A,
B and C. Unique Incised A sherds, heavily tempered with quartz sand, may
represent the head of an owl (Pi. IV, S). Incision is deep, bold, and apparent
ly done when the clay was fairly dry. In paste, technique, and motif this ves
sel is reminescent of Point Washington Incised. Moore (1901, p. 483) illus
trates a vessel from the cemetery near Point Washington with a flat triangular
rim projection. While differing in details it forms the top of a head as does
the similar area on the specimen from the Prine mound. The two sherds of
Unique Incised A came from square 3K, between depths of 12 and 24 inches
and below a sherd of Pinellas Incised B.
Unique Incised B, sand-tempered and apparently from the neck of a water
bottle, is suggestive of Englewood Incised (PI. IV, T).
Unique Incised C sherds, also sand-tempered, are from a small, poorly
made, poorly finished vessel (Pi. IV, U). The carelessly applied but fine in
cised lines suggest the work might be that of a child.
Parts of four small undecorated vessels, 2% to 3% inches deep, have
what is usually referred to as prefired kill holes. Sherds in the collection
indicate the water bottle shape to have been a fairly common form of vessel
at Terra Ceia.
Plain, sand-tempered sherds from the Prine mound include those made
of paste similar to that found at the Safety Harbor site, many Weeden Island
sites, and in the Glades area (Glades Plain). It does not seem proper (or
profitable in this case) to try to divide these plain wares. Plain, contorted
paste sherds duplicate many such sherds from the Safety Harbor site. They
do not, however, represent such a horizon at the Prine mound because they
increase numerically with depth, both absolutely and relatively. Here they
must be considered poorly made utilitarian ware of the Weeden Island time
period. Under Perico-Pasco Plain have been included limestone-tempered
sherds whose content of crushed rock ranges from a small to a very large
amount.
Most sherds from the Prine mound show that vessels were built by coil
ing. In some cases coils were very wide and the method should probably
be called segmental building instead of coiling. Exceptions are miniature
28


vessels with prefired kill holes, which were hand molded, and Belle
Glade Plain and Safety Harbor Complex vessels for which the evidence is
inconclusive.
Where available some indication of the size and shape of vessels has
been given for decorated wares. The bulk of the pottery consisted of culi
nary wares, plain or check stamped vessels. These were simple bowls
ranging from 7 to 17 inches in diameter. Most were fairly large shallow af
fairs about 12 inches in diameter. An exception was a deep, narrow jar
(Wakulla Check Stamped) 6 inches in diameter and about 10 inches deep.
Sherds from the Prine mound are listed in Table 4 by depths below the
present surface. This surface is the end result of various activities which
have gone on at the site and is used as a convenient reference for vertical
distribution. It has been mentioned earlier that refuse bearing pits of the
20th century were encountered. Bundle burials represent aboriginal pitting,
at least to some extent. In spite of these disturbances, the vertical distri
bution of pottery agrees with the latter part of the ceramic chronology of the
Northwest Coast of Florida Weeden Island I, Weeden Island II, and Safety
Harbor, from earlier to later.
Sherds of the Safety Harbor complex are limited, practically, to the top
6 inches. Those of the Englewood Complex concentrate in the upper 12
inches except for Lemon Bay Incised. While the latter may be a relatively
earlier pottery type, evidence for such a conclusion would have to come
from elsewhere as two of the three Lemon Bay Incised sherds were in the
fill of the group bundle burial.
Weeden Island sherds while abundant in the top level were also found
in large quantities at greater depths. These sherds are typical of the Weed
en Island II period.
Interestingly, Swift Creek, Sun City, Old Bay, Crooked River, and St.
Andrews Complicated Stamped, West Florida Cord Marked, Carrabelle
Punctated, and Keith Incised sherds were only found at lower depths. Five
of the twelve Swift Creek Complicated Stamped sherds are of the Late
Variety with Weeden Island type rims, typical of Weeden Island I. All are
types of the Weeden Island I period except for Sun City and Old Bay which
are usually considered Weeden Island II (Willey, 1949, p. 437).
It appears, therefore, that the pottery producing portion of the Prine
mound was started in very late Weeden Island I times, that most of it was
deposited during Weeden Island II, and that additions were still being made
to it during the Safety Harbor time period. This would suggest the Prine
29


Table 4. POTTERY FROM PRINE MOUND
Safety Harbor Complex
Lake Jackson Plain
Lake Jackson Type Adorno
Pinellas P lain
Pinel las Incised B
Pine I las Incised C
Cob-marked
Safety Harbor Incised
Englewood Complex
Englewood Incised
Sarasota Incised
Lemon Bay Incised'
Weeden Island Complex
St. Petersburg Incised (?)
Weeden Island Incised
Weeden Island Punctated
Weeden Island Plain
Weeden Island Plain, Pasco paste
Pa pys Bayou Plain
Papys Bayou Punctated
Gainesville Linear Punctated
Carrabelle Punctated
Carrabelle Incised (?)
Keith Incised
Wakulla Check Stamped
Pasco Check Stamped
Biscayne Check Stamped
Tampa Complicated Stamped
Pasco Complicated Stamped ^
Swift Creek Complicated Stamped
Sun City Complicated Stamped
Old Bay Complicated Stamped
Crooked River Complicated Stamped
St. Andrews Complicated Stamped
West Flcrida Cord Marked
Thomas Simple Stamped
Rusk in Dentate Stamped
Ruskin Linear Punctated
Hillsborough Shell Stamped
Santa Rosa-Swift Creek Complex
Alligator Bayou Stamped (?)
Glades Complex
Miami Incised
Bel le Glade Plain
Depth below surface, inches Not From Totals
by Cha-
0-6 6-12 12-18 18-24 24-30 level dea-
yne
1
1
4
5
3
1 1
21 4
1
1
4
5
3
2
26
12
12
1
5a
2
32
1
2a
la
1
3
7
2
1
2a
12
2
2
3
7
3
8
2
1
2
2
18
4
11
13
11
3
2
8
52
1
2
4
6
4
3
20
7C
2b
8C
6
4
1
12
28
4
2
6
1
4
1
6
4
2
1
i
7
72
75
i
65
16
7
12
5
252
14
19
21
8
4
5
4
75
28
17
3
7
1
1
1
58
21
12
13
10
1
5
62
17
10
8
5
1
1
1
43
3
6
2
1
12
2
1
1
4
1
1
1
1
12*
1
13
16b
9b
1^
2
6
5
1
1
52
14
19
13
9
2
12
69
369
1
25
24
45*
4
18
2
152
1
1
i
1
32
40
9
5
7
93
Miscel laneous
Unique Incised A
Unique Incised B
Unique Incised C
Miscellaneous incised
Miscellaneous punctated
Unnamed bold check-stamped
Indeterminate stamped
Scored surface
Smooth plain
Plain red
Biscayne Red
Biscayne Roughened
Biscayne Plain
Limestone-tempered chalky Plain
Perico-Pasco Plain
Plain, Sand-tempered
Plain, Contorted Paste
Totals
2
6
3b
o
1
2
0
14
10
6
12
5
4
37
1
1
2
33
32
28
14
2
5
5
119
5
2
1
3
11
3
3
2
2
11
21
8
6
3
2
1
1
1
22
12
28
14
6
3
6
69
6
2
5
1
14
153
191
131
89
23
43
30
660
1
1
5
7
162
145
154
134
76
61
59
791
1557
1805
1569
1062
320
387
167
6867
136
187
223
180
85
34
23
868
2420
2698
2376
1660
551
614
332
10651
Footnotes, a. In fill of Burial 10; b, One limestone-tempered; C, Two limestone-
tempered; d. Late variety; e. From one vessel; f. Twenty-nine from one
vessel; g. Eight limestone-tempered.
30


19
mound to have been used for burials over a possible 350 years.-1-4
Such extended use sounds exaggerated at first thought. However, it is
not inconsistent with the theory, presented earlier, that the vast Abel
midden represents the accumulation of a relatively small population over a
period of 800 years. Nevertheless, there is nothing in the data which would
prevent the mound having been built in late Weeden Island II times. The
minimum estimate on that basis, assuming continued use, could not, I believe,
be compressed into less than 100 years. The indications are that well over a
hundred individuals of assorted ages were buried in the Prine mound. Assum
ing a constant population and a 1 per cent mortality, the resulting population
would be over 100 people. This would seem to be entirely too large for the
natural resources even if some allowance were made for agriculture. I am
inclined, therefore, towards the longer time range.
Temporal relationships between the Prine mound and the Abel midden
are tenuous because definitively decorated pottery was not found in the mid
den. However, they can be correlated with a fair degree of probability. About
ten per cent of all sherds listed in Table 4 are limestone-tempered. Lime
stone-tempering as a trait occurred in an intermediate position, stratigraphi-
cally, at the Abel midden (Tables 1 and 2). Biscayne Red was the most com
mon pottery type at the Prine mound except for Wakulla Check Stamped,
Ruskin Dentate Stamped, Hillsborough Shell Stamped, and plain sherds,
while it was found in the Abel midden between depths of 18 and 42 inches.
Weeden Island pottery at the Prine mound, presumedly, equates with these
portions of the Abel section.
Location and position of burials at the Prine mound is indicated on the
excavation plan (Fig. 6). Vertically, bundle burials were in the pottery pro
ducing zone while most flexed burials were at the base of the mound, below
pottery and partially in the black-brown stratum. Both types of burials were
unaccompanied by grave goods. Skeletal material was in very poor condition.
Many skulls were warped by ground pressure. Bones of bundle burials were
substantially eroded by ground acids as were the higher parts of the lower
flexed burials. Bones of flexed burials, and to a lesser extent those of bun
dle burials, had a black-brown surface which was hard but brittle and fre
quently cracked in situ. It is thought tannic acid was the chief agency in
producing this case hardened effect. In the following brief comments,
positions and directions of flexed skeletons are omitted as that information
may be secured from the excavation plan.
Burial 1 Skull, fragmentary, eroded, thick (1 cm.) with small mastoids.
No teeth or long bones. In pottery zone.
12
For period dates see Willey, 1949, Fig. 76; Goggin, 1949, p. 16.
31


Burial 2
Burial 3
Burial 4
Burial 5
Burial 6
Burial 7
Burial 8
Burial 9A
Burial 9B
Burial 10
Burial 10A
Probably flexed. Eroded, small, undeformed, roundheaded
skull with small mastoids. Probably female. Long bones
above skull of No. 3. In lower part of pottery zone.
Flexed. Skull thick and massive with medium-sized brow
ridges and some suture closure. Mandible very large, much
tooth wear, lower left molars lost in life. Probably middle
aged male. Partially in black-brown zone and in pit scooped
out of clay.
Flexed. Thick skull with small mastoids, some suture closure.
Length 176 mm., width 137mm., C.I. 78%. Excessive tooth
wear. Femur large but not particularly muscled, femur head
38-40 mm. diameter. Height estimated as 5 feet. Sacrosciatic
notch broad and shallow. An old female. In black-brown zone.
Possibly flexed. Fragmentary skull with small mastoids and
a few long bones in dark area, 18 by 24 inches, in lower part
of pottery zone. Probably a child.
Flexed. Thick skull, small mastoids. Adult. Partially in
black-brown zone.
Bundle. Fragments of skull and long bones. Mandible large
with excessive tooth wear. In black-brown zone.
Tightly flexed. Skull warped but apparently mesocephalic.
Much suture closure. Extreme tooth wear. Sex indeterminate.
In black-brown zone.
Bundle (?). Skull, fragmentary, much eroded. No teeth or
long bones. Child. In pottery zone.
Bundle. Skull and two long bones. Much eroded. In pottery
zone.
Group of bundle burials. Many more bones present than shown
on excavation plan. Bones began 18 inches below surface and
occupied a zone 6-8 inches thick except for lower ends of
vertical long bones and pelves which extended deeper and,
sometimes, penetrated black-brown zone. Sherds in fill in
clude 4 Englewood Incised, 2 Lemon Bay Incised, 3 St. Pe
tersburg Incised (?), 1 Unique Incised B, 1 Biscayne Plain,
and 14 Plain, sand-tempered. Recognizable subdivisions of
this group of burials follows.
Vertical bundle. Articulated femurs and pelvis placed in
bottom of pit with femurs extending upwards. Other long bones
placed horizontally between femurs with skull on top between
distal ends of femurs.
32


Burial 10B Horizontal bundle. Pile of long bones with skull placed
near one end. Cervical vertebrae under skull. Sherd of Bis-
cayne Plain under long bones.
Burial IOC Vertical bundle. Pelvis in bottom of pit with articulated
femurs extending upward at angle of 30 from vertical.
Other long bones placed parallel to femurs on western side
of pit. Other bones and fragments of skull on top in pit.
Burial 10D Vertical bundle. Long bones placed on eastern side of small
pit, sloping about 45 from vertical. Fragments of mandible,
maxilla, and other bones in pit over parts of long bones.
Tibia on top.
Burial 10E Vertical (?) bundle. Eroded skull with vertically placed long
bone below it. Other long bones to west sloping downward to
towards south.
Burial 10F Badly decayed skull, face downward. No mandible.
Burial 10G Horizontal bundle. Long bones placed horizontally in pit under
decayed skull. More long bones placed horizontally over skull.
No teeth.
Burial 10H Decayed skull. No mandible. No teeth. Probably child.
Burial 101 Fragmentary skull with vertically placed long bone. Unique
Incised B sherd. Burial not completely excavated.
Burial 10J Vertical bundle. Two long bones placed on each side of pit,
sloping about 30 from vertical. Other bones placed between
sloping bones with skull on top. No pelvis found.
Burial 10K Fragmentary skull including maxilla.
Burial 10L Spare mandible. May belong to Burial 10K.
Burial 10M Thick skull cap, found with opening upward. Isolated from
other bones of Burial 10 and at a higher elevation.
Burial 11 Badly decayed, isolated skull of an adult. In pottery zone.
Burial 12 Bundle. Eroded fragments of skull, mandible, and long
bones. One bicuspid without tooth wear. Probably youth.
In pottery zone.
Burial 13 Bundle. Long bones including femur articulated with
innominate. In pottery zone. Burial 11 may represent skull
displaced from this burial.
Burial 14 Bundle. Fragmentary skull with scattered long bones to east.
In pottery zone.
Burial 15 Flexed. Skull not found. In black-brown zone.
Burial 16 Flexed. Skull thick, prominent brow ridges, small mastoids,
33


sutures well closed. Mandible eaten away for area % inch
across below lower second bicuspid, probably result of ex
treme pyorrhea. Old adult, sex indeterminate. Head and
shoulders in lower part gray sand, pelvis in black-brown
zone. See Plate I.
Burial 17 Flexed (?). Mandible but no skull present. Hand and lower
arm bones under leg bones. In black-brown zone.
Burial 18 Two eroded long bones surrounded by brown dirt which
covered an area 10 by 15 inches. In pottery zone.
Burial 19 Eroded long bone. In pottery zone.
Summarizing the above data we have evidence of the burial of at least
twenty-seven individuals in the excavated portion of the Prine mound. Seven
are represented by flexed burials partially in the black-brown zone, two by
flexed burials in the lower part of the pottery zone, and the balance by bun
dle burials in the pottery zone, including eleven in the bunched bundle burial.
Condition of bones prevents much information regarding age and sex but the
great majority of interments are those of adults, some of them well along in
years. A gracile, meso- or brachycephalic population is represented. In keep
ing with Floridian skeletal material in general, skull caps are relatively
thick.
The group bundle burial may be presumed to represent bodies collected
from a carnal house where they had been temporarily left, as was done in
early historic times, and then buried at one time (Bourne, 1904, pp. 28-29).
Evidently bodies were not exposed a sufficient length of time for all liga
ments to decay as some articulated bones were found among these burials.
Due to lack of traceable burial shafts we were unable to determine sur
faces at times of inhumation. Obviously flexed skeletons at the base of the
mound were buried before the deposition of the sand mantle containing pot
tery of the Weeden Island II period. It is also evident that the bundle burials
were inhumed during or after installation of this mantle. While Safety Harbor
Incised sherds had general distribution horizontally, sherds of Pinellas In
cised were limited to squares 3K-5K and those of Lake Jackson Plain to
square 5L. This pottery, presumedly the latest at the mound, was over the
scattered bundle burials, Burials 11-14. It is believed, therefore they rep
resent the most recent burials of the excavated area.
Following this reasoning, it may be suggested that the flexed burials
were interred during pre-Weeden Island or Weeden Island I times, and the
bundle burials during Weeden Island II times plus a few of the Safety Harbor
period. Prone burials in the top of the mound, mentioned previously as having
34


been found by Mrs. Kissick, would then be of late Safety Harbor or later
historic times. As historic Safety Harbor mounds in the area, Parrish Mounds
1-3, contained only bundle burials (Willey, 1949, pp. 142-145), it may be pre
sumed that these prone burials were interred in post-Safety Harbor times.
OTHER SITES ON TERRA CEIA ISLAND
The location of Indian sites on Terra Ceia has been indicated on the
regional map (Fig. 1). Of these sites 2-6 are small shell deposits along
the shore or beside bayous. A few plain sherds were found at these sites
including, at site 5, a sherd from a Spanish olive jar.
Site 7 is the location of the Kennedy mound, removed for fill about 40
years ago. Mrs. Kissick, who told us of this mound, advises that it was
built of shell and sand and originally was about 16 feet high. Shell and
midden debris now cover an area, 100 feet across, where this mound stood.
This appears to have been a ceremonial mound, similar to the Madira Bickel
mound, and there should have been an Indian village near by.
0
35


DISCUSSION AND CONCLUSIONS
The large Indian site at Terra Ceia has been considered by some to be
Ucita, the Indian village to which DeSoto moved his forces immediately
after landing in 1539 and where he left a rear guard of over 100 men when
he marched north. Both Elvas and Garcilaso place this town as two leagues
(5.26 miles) from the landing place. If the landing was made at Shaws Point
the distance to Terra Ceia would be about right but it is separated from
Shaws Point by the Manatee River, Sneads Island, and Terra Ceia Bay
(Fig. 1). It is interesting to note, in passing, that two leagues is also the
distance from Shaws Point to the present city of Bradenton.
Swanton appears to confuse the issue when he interprets Ranjels ac
count of DeSotos attempt to locate this village to mean the army had to
travel around the Manatee River to reach Ucita. Ranjel writes: On Trinity
Sunday, June 1, 1539, this army marched by land towards the village ....
and they lost their bearings somewhat .... Thereupon the Governor went
ahead with some horsemen, but since they were unfamiliar with the land
they wearied the horses following deer and floundering in the streams and
swamps for twelve leagues (31.5 miles) till they found themselves opposite
the village on the other side of the roadstead of the harbour, which they
could not pass over (Bourne, 1904, vol. 2, p. 55).
This clearly was a scouting.expedition and not the march of the army.
That DeSoto finally saw Ucita across water does not mean he had to go
around that water to get to the village from his original starting place.
Swanton is incorrect in using this quotation to require Ucita to be located
2 leagues from the landing place but to require a trip of 12 leagues before
it could be reached by land.-^
DeSoto left a rear guard of 86 to 120 men (depending on which narrative
one follows) at Ucita. As a minimum of 26 and a maximum of 40 cavalry are
mentioned, we must assume at least that many horses were'also left at
Ucita. To this force must be added the sailors of three small vessels. Ac
cording to Elvas these men were left with provisions for two years. They
lived at Ucita for five months after which the soldiers joined DeSoto and the
boats sailed northward.
Garcilaso writes that when this camp was broken up, supplies the
Spaniards couldnt take with them were given to Indians. He lists: Five
hundred quintals of cassava bread, great quantities of cape cloaks, sack
coats, jackets, breeches, long trousers, all types of leg and foot coverings,
13
Swanton, 1939, pp. 127-138; Wilkinson, 1947, No. 36 of Dec. 25, 1947, has already
pointed out this error.
36


shoes, buskins, and hempen sandles. Of armament equipment, there were
many vests of armour (cuirass), round bucklers (shields), pikes, lances and
steel helmets. The governor was rich, so that he had brought along a super
abundance of all these things. There was also a large amount of naval
stores, such as sails, tackle, rosin, tow, tallow, ropes, pannier baskets,
hampers, anchors, cables, many iron tools (or material), and much steel
equipment (Lewis and Wilkinson, N.D.).
The question of possible Spanish objects found at the site has been
discussed with Mrs. Abel, Mrs. Kissick, Mr. Johnson, and Mr. T. Ralph
Robinson of Terra Ceia, with Mr. Tallent of Manatee and Mr. Chadeayne
of Bradenton, who have collected Indian artifacts for many years from the
area. The sum total of objects found at or near the site, which might conceiv
ably represent occupation by Spaniards, consists of a Spanish bronze medal
lion (?) found by Mr. Abel (now lost), a small barrel filled with sails found
by Mr. Johnsons father, a piece of horse hardware plowed up years ago,
and one sherd of a Spanish olive jar. There are also rumors of a small brass
cannon found by a tourist and of a sword-like piece of metal found by a Mr.
Ford, both in the Abel midden. These finds are not well substantiated.
The great amount of midden and mound removal at Terra Ceia would' have
uncovered abundant evidence of Spanish occupation if DeSoto left 100 men
there for five months with any such vast stores as are mentioned in the
narratives. Terra Ceia is extremely low and extra high tides, such as
accompany storms, sometimes inundate considerable portions of it. It is not
the type of place a capable commander like DeSoto would pick for his head
quarters taking into consideration the presence of many horses and pigs.
Nor does the Elvas description of Ucita agree well with the situation at
Terra Ceia (Bourne, 1904, vol. 1, pp. 23-24). It seems conclusive that the
large site at Terra Ceia is not the Ucita of the DeSoto narratives.
Data has been presented indicating the Terra Ceia site to have been
occupied over a long period of time, possibly 800 years, by a relatively
small number of Indians. During these years large shell middens accumulated
on the shore of Miguel Bay where homes were located. The Johnson and
Prine mounds were built and used for mortuary purposes, and, relatively late,
the Bickel mound was constructed for ceremonial purposes. Probably the
burial mounds grew by accretion and the ceremonial mound was built in a
succession of stages.
In 1948 Willey (1948, Table 12, p. 216) set up a chronological scheme
for the Manatee region of Florida, in which Terra Ceia is located, showing
an early Perico Island period followed by Weeden Island and then by Engle
wood and Safety Harbor periods. The Englewood site by seriation and from
37


stylistic considerations was considered intermediate between Weeden Island
and Safety Harbor. In his more recent presentation, the same sequence is
given with the additional suggestion that Weeden Island might be split into
an early and a late phase, as in northwest Florida (Willey, 1949, Fig. 76).
Data from Terra Ceia fits this arrangement well. At the Prine mound
most of the decorated pottery was typical of the Weeden Island II period, as
defined, with some Weeden Island I types in lower levels and Englewood-
Safety Harbor ceramics in superior zones. Tests at the Abel shell midden
uncovered a lower zone characterized by plain sand-tempered pottery plus,
towards the upper portion, some limestone-tempered sherds. This lower
zone equates with Willeys Perico period. As a sample it is rather close to
the ceramic content of the Perico burial mound where 482 out of 496 sherds
were plain and sand-tempered, Glades Plain (Willey, 1949, p. 177). The
Miami Incised (?) sherds would argue similarly. Higher zones supplying
Biscayne Red and many limestone-tempered sherds are Weeden Island by
correlation with the Prine mound. Presumedly, the superior levels without
limestone-tempered pottery represent the Safety Harbor time period. Certainly
this time period is represented by the notched lip and Ft. Walton Incised
sherds from the Boots Point midden.
Locations of important sites near Terra Ceia have been indicated on the
regional map (Fig. 1). Shell deposits on McGill Island, to the south of Terra
Ceia, are not extensive but those at Sneads Island and Shaws Point are com
parable. At Sneads Island is a ceremonial mound of the same construction
and about the same size as the Bickel mound. Sherds from Shaws Point in
dicate occupancy to have occurred over the same time periods as at Terra
Ceia and probably to have continued later (Willey, 1949, p. 341).
The Prine burial mound is similar to other Weeden Island mounds of the
Tampa Bay area. At the famous Weeden Island mound itself, near St. Peters
burg, flexed burials and a few plain sherds came from below an old surface.
At shallower depths were extended burials while the bulk of the mound pro
duced bundle burials and a ceramic inventory, including Safety Harbor and
Englewood complex sherds, close to that from the Prine mound. Skulls were
meso- or brachycephalic (Willey, 1949, pp. 107-111).
At the Thomas mound (Fig. 1) bundle burials overlay flexed burials and
skulls were meso- or brachycephalic. The ceramic inventory, as might be
expected from its nearer location, is closer to that from the Prine mound al
though Englewood sherds were, apparently not present (Willey, 1949, pp.
116-121).
At the Englewood mound, 55 miles south of Terra Ceia, most interments
38


were bundle burials. Sixteen flexed burials (5% of the total) are included in a
list of burials from underthe original surface, in the primary mound, and in the
secondary mound. Presumedly they came from lower zones as at other sites.
Only 509 sherds out of the thousands found were available for Willeys analy
sis. This reduced inventory contains a relatively small amount of Weeden Is
land pottery types and a substantial increase in Sarasota Incised and Lemon
Bay Incised when compared with the Prine mound but in respect to Engle
wood Incised, Papys Bayou Punctated, Safety Harbor Incised, and the Bis-
cayne series the absolute numbers are comparable. Pinellas Incised, present
at the mounds previously mentioned, was not, apparently, found at Englewood.
These differences are probably regional in import. Interestingly, dozens of
conch shells, whole and broken, were found throughout the mound (Willey,
1949, pp. 127-134). This feature was repeatedly noted at the Prine mound.
At Weeden Island, Thomas, and Englewood there was physical stratigraphy
but, unfortunately, excavated material was not carefully referred to physical
zones. At the Thomas (Willey, 1949, p. 119) and Prine mounds, Safety Harbor
period sherds were definitely limited to superficial depths. Presumedly, this
was also the case at Weeden Island and Englewood. As the flexed burials
at Weeden Island were below the mound base and at Prine below the pottery
zone, they may have been pre-Weeden Island interments. The large number
of bundle burials in fill containing quantities of Weeden Island sherds are
unquestionably of that period and not later intrusions. Some of these burials
are Safety Harbor in date but, again, not intrusive. The Thomas mound with
silver ornaments and 200 glass beads appears to be the latest in end date.
Granting continued use of this mound into the Safety Harbor period, as shown
by the pottery, these specimens do not necessarily mean intrusive burials
but may show continued use into the historic portion of the Safety Harbor
period.
If due allowance is made for regional variation, these burial mounds,
Weeden Island, Thomas, Prine, and Englewood, all represent approximately
the same time span. They were all started during or before the Weeden Island
period and were still being used after the introduction of Safety Harbor Pot
tery types. This implies the passing of a considerable period of time.
Concomitant with a long period of use of these mounds is the theory
they grew by accretion and that the remains are those of a relatively small
population. The large number and random arrangement of burials supports
the first postulation and similarity in physical types the second. This also
implies no sudden influx of new people, in this area at least, with the com
ing of Safety Harbor times.
Tests in the Abel midden give us our first stratigraphic data for the
39


southern portion of the Tampa Bay area. Cockroach Key and Perico Island
(Fig. 1) are large nearby sites of a comparable nature. The burial mound at
Cockroach Key was found to have physical stratigraphy but few of the data
were recorded accordingly. Most burials were flexed while relatively few
(less than 14%) were of the bundle type. Pottery from this mound included
317 plain sherds (Belle Glade and Glades Plain), 18 Biscayne Check Stamp
ed, and 23 classified as Pinellas Plain although lacking diagnostic features.
The latter, without notched lips or rim projections, cannot be distinguished
from sherds of contorted paste as found at the Prine mound and so lose their
diagnostic value as an indicator of Safety-Harbor times. These Pinellas-like
sherds, 39 of the 40 Belle Glade Plain sherds, and all of the Biscayne Check
Stamped sherds came from the tertiary or superior mound and are therefore re
latively late at the site (Willey, 1949, pp. 162-172). Examination of the pot
tery tabulation for the Prine mound (Table 4) will show an appreciable quanti
ty of Belle Glade Plain, Pinellas-like Plain (contorted paste), and Biscayne
Check Stamped sherds. It would appear that the tertiary mound at Cockroach
Key should be equated with part of the Prine mound and so be Weeden Island
in date, probably relatively early Weeden Island II. Presumedly the few bun
dle burials were in this tertiary mound but, unfortunately, data does not cover
this point. The flexed burials of the lower zone at the Prine mound would then
equate with the more extensive secondary burial mound at Cockroach Key.
Other tests at Cockroach Key produced plain sherds, a great many shell
hammers, shell tools, and bone points, etc. This inventory is very similar to
that from the Abel midden. One would expect some limestone-tempered pot
tery to be present. As only 99 sherds were available for Willeys analysis
out of 1221 found at Cockroach Key there is a good chance limestone-tempered
sherds may have been among those discarded.
On May 26, 1950, I was able to visit Cockroach Key and make a small
surface collection. Time did not permit a careful examination but one lime
stone-tempered and 41 sand-tempered (Glades Plain) sherds were found.
This would support the theory that limestone-tempered pottery was used at
Cockroach Key. In contra-distinction to most shell middens on the Central
Gulf Coast, the one at Cockroach Key is built predominantly of oyster shells.
Willey believes occupation of Cockroach Key to have started in Glades II
(or Glades I) and to have continued into Glades III or, in terms of the Mana
tee area, Perico Island and Weeden Island. This would appear to be correct.
Cockroach Key and Terra Ceia are comparable sites co-existing during much
of their time range. No doubt Terra Ceia continued to be inhabited after Cock
roach Key was abandoned.
The site at Perico Island is described by Willey from Newmans field
40


notes of 1934 as a large shell midden connected by walkway to a burial
mound built of sand, shell, and some midden debris with, a little to the
south, another but smaller midden and connected walkway. A cemetery
area was also discovered. Part of the smaller midden is said to have been
removed for commercial purposes. Newman excavated the burial mound, the
cemetery area, and a trench through the smaller midden.
This arrangement, shell middens connected by walkways to burial mounds,
sounds similar to that found at Terra Ceia. However, a visit to the site on
May 25, 1950, disclosed some discrepencies in the above description. It ap
pears that a considerable amount of the larger and northern midden has been
removed for commercial purposes. This may have occurred since 1934. There
is some evidence of shell removal from the southern midden but this would
not appear to have been extensive.
A more or less circular area, slightly dome-shaped and sparsely covered
with weeds, was presumed to be the site of the burial mound. Between this
field and the main (northern) midden is a narrow shell ridge but it is entirely
unlike walkways to be seen at Terra Ceia and Shaws Point. It appears,
rather, to be merely a narrow extension of the shell midden. A 5*foot test
hole, to be mentioned later as Perico Test A, was made in this narrow ex
tension of the northern shell midden. A similar test, to be known as Perico
Test B, was made in the southern shell midden about 10 feet west of its
central line and 20 feet north of an old trench, presumedly Newmans test.
Newmans work disclosed only flexed burials in the mound and in the
cemetery. This would suggest occupation of the site to have occurred earlier
than Weeden Island, equal to or earlier than the flexed burials found in the
lower parts of Weeden Island mounds mentioned earlier.
Pottery from the Perico Island burial mound was predominantly plain and
sand-tempered with eight limestone-tempered (Perico Plain), three Biscayne
Plain, two Deptford Bold Check Stamped and one early complicated stamped
sherds. This inventory is reminescent of that from the upper portion of the
lower part of the Abel midden (pre-Weeden Island or Perico Island). Theoret
ical considerations indicate Deptford and early complicated stamped sherds
should be found at the later site if a sufficient volume were excavated.
Unfortunately, no tests were made by Newman in the large midden con
nected to the burial mound. The smaller midden produced Glades Plain (sand-
tempered), Perico Plain (limestone-tempered), Perico Incised and Perico
Linear Punctated, Biscayne Plain, St. Johns Incised (?), and fiber-tempered
sherds (Willey, 1949, pp. 173-182). Although this midden was 5 feet thick,
material was not kept by arbitrary or stratigraphic zones and no data is
41


available from the site regarding chronological relationships between these
pottery types.
It was hoped that our tests (1950) would clarify this situation but un
fortunately no incised or fiber-tempered pottery was found. Perico Test A
disclosed a homogeneous deposit of black dirt, charcoal, sand and shells,
5 feet 8 inches thick resting on sterile gray sand. Shells, presumedly as a
result of trampling, were mixed with the gray sand in the top 6 to 8 inches of
of the lower deposit. Venus, Busycon, Spisula, Fasciolaria, Pecten, Strom-
bus and, rarely, Ostrea shells were in the midden. Relatively more oysters
were found near the bottom.
At Test B the deposit was composed homogeneously, of clean crushed
shells. At a depth of 3 feet water was encountered so that we were unable
to reach the base of the deposit or a depth much over 4 feet. All sherds from
this test, except three, were rounded and waterworn.
Sherds from these tests are tabulated below. Included in a small surface
collection was an eroded Gulf Check Stamped sherd.
Table 5. POTSHERDS FROM TESTS A AND B,
PERICO ISLAND
Test A
Test B
Depths in feet
0-1
1-2
2-3
3-4
4-5
0-1
1-2
2-3
below
three
Plain, Lake Jackson paste
Biscayne Plain
Belle Glade Plain
Pasco Plain
2
1
1
1
1
2
Perico Plain
Smooth, Plain
16
1
1
1
1
Glades Plain
8
22
11
1
2
15
64
35
These tests are inconclusive but they do show limestone-tempered pot
tery to be present in intermediate or early levels. Lack of fiber-tempered or
Perico Incised sherds in test B may be explained by the fact we were unable
to reach the bottom of the deposit.
Some of the pottery from the 1934 excavation is obviously early, as fiber-
tempered sherds, undoubtedly, represent the earliest pottery of the Manatee
region as similar sherds do elsewhere in the southeast. The question is,
Did the limestone-tempered Perico pottery follow immediately thereafter?
Due to similarity of incised and punctated designs on a few of these sherds
to those found on some fiber-tempered pottery of the east coast, this may be
the case. If so, a period followed when decoration on pottery was extremely
rare, to judge from the lower zones at the Abel midden and from the 1950 tests
at Perico Island. Such a situation would have analogies with the east coast
42


where fiber-tempered decorations were carried over onto early chalky pottery
(St. Johns Incised) but were shortly abandoned and for the long St.Johns I
period local pottery was undecorated (Goggin, 1949, p. 24).
Willey (1949, p. 181) takes the contrary view and speculates that the
Perico Series (limestone-tempered) types are later innovations at the Perico
site, and that a horizon of virtually 100 per cent Glades Plain probably pre
cedes it. Lack of limestone-tempered pottery in the lowest zones at the
Abel midden strongly supports Willey. However, the similarity in designs on
Perico Incised and incised fiber-tempered is not explained. Lack of lime
stone-tempered sherds in the lower zones at the Abel midden may be the
fault of sampling. At Johns Island, 80 miles north of Terra Ceia, limestone-
tempered pottery with decoration similar to that found on fiber-tempered
pottery (Pasco Incised) was associated with St. Johns Incised in the lowest
zone (Bullen and Bullen, 1950). Such data would support the earlier dating
for part of the Perico Island site and suggest that more work at the Abel
midden might disclose limestone-tempered pottery in the lower zones. Our
inconclusive tests at Perico Island may be used to support either hypothesis
but favor the first one.
Analysis of data from Terra Ceia and certain other sites in the Tampa
Bay area strongly suggests that most large sites were occupied over a
fairly long period of time. In most cases this time seems to have spanned
more than one archaeological period. Table 6 is an attempt to illustrate the
situation in accordance with our present knowledge. No attempt has been
made to include sites other than those mentioned earlier except for the
Parrish mounds (Willey, 1949, pp. 142-158).
A question mark will be found after Weeden Island I. This is because
no clear indication of such a period (Swift Creek Complicated Stamped but
no Wakulla Check Stamped pottery) has, so far, been discovered in the Mana
tee region. Omission of this period might improve the visual representation
of Table 6.
Sequence of burial types has been suggested in the table and has been
a prime consideration in its construction. Prone burials reported for the
eastern side of the shell ridge at Terra Ceia and those from the top of the
Prine mound have been disregarded. The former may represent old burials
of the early part of the Perico Island period, inhumed prior to the construc
tion of the burial mounds. Those from the top of the Prine mound are not so
easily disposed of. Assuming the correctness of Mrs. Kissicks memory,
they must represent very late interments. As such prone burials are not re
ported for the other mounds it seems proper, in this case, to consider the
possibility that they may be intrusive.
43


Vertical location of the ends of bar-graphs in Table 6 have been deter
mined by certain considerations. Locations in the prehistoric Safety Harbor
period reflect the presence and absence of Safety Harbor Incised and Pinel
las Incised.^ Lack of the later at Englewood may be regional. Presence of
nine sherds of Papys Bayou Punctated at Parrish No. 3 has extended the
line for this site downward into the Weeden Island period. This may be in
correct.
Table 6. TEMPORAL PLACEMENT OF SITES IN THE MANATEE REGION
AND ADJACENT AREAS
v
a
10
O
C
G
a
o £
.2 -5
E G
G
m
in h £
o
.£2
T3 O O I
§ z z ;
s
Burial
Form
i m fc fc s o e
B
fcOHSHHCUOnizltceO
Indian cultures of the Manatee area, as we know them, started with a
maritime economy. Sites were located adjacent to shallow water teeming
with shellfish which formed the major component of their food supply,
abundantly supplemented with fish, turtles, mammals, and birds. Remains
of their material culture include pottery, shell tools and bone points (pins).
From this base, the culture gradually changed to what we speak of as Safety
Harbor through the intermediate stage we refer to as Weeden Island. This
change was brought about by external influences, all of which came from
further north. Evidences of these changes are seen in the succession of
pottery types and modification of burial habits. Communication to the east
is indicated by Belle Glade pottery. The basic maritime economy seems to
have been modified but little. Eventually ceremonial mounds were introduced.
These mounds imply a change in social organization and may reflect a greater
dependence on agriculture. Fully historic burial mounds, such as Parrish 1,
2, and 3, are located at substantial distances inland from the gulf. This ap
parent seeking of good agricultural land during the Safety Harbor period is
similar to the situation on the Northwest Gulf Coast where contemporaneous
14
Pinellas Incised,
on the average, is the more recent.
44


Fort Walton period sites are usually located inland.
The last pre-Seminole Indian occupation of the Manatee region probably
occurred during or shortly after the Spanish Mission period of north Florida.
While only one Spanish olive jar sherd has been found at the Abel midden and
another at a small site on the south shore of Terra Ceia, pottery of the Spanish
Mission period is known from one locality on Sneads Island (Fig. 1). There
Mr. William C. Chadeayne of Bradenton collected 1 green-glazed, 30 plain
Spanish olive jar sherds, 3 Mission Red Filmed, 8 Miller Plain, 1 sherd with
an indented rim strip, and 7 sherds of a semi-chalky paste. This assemblage
indicates occupation during or after the period of Spanish Missions in North
Florida, circa 1633-1704. Indian refugees from these missions may have intro
duced the trait of prone burials reported by Mrs. Kissick for the top of the
Prine mound.
45


BIBLIOGRAPHY
Bickel, Karl A.
1942. The Mangrove Coast. New York.
Bourne, Edward G.
1904. Narratives of the Career of Hernando deSoto. 2 vols. New York.
Bullen, Adelaide K. and Ripley P.
1950. The Johns Island Site, Hernando County, Florida, American
Antiquity, vol. 16, pp. 23-45. Menasha, Wise.
Goggin, John M.
1949. Cultural Traditions in Florida Prehistory, in The Florida
Indian and his Neighbors (John W. Griffin, ed.). Winter Park, Fla.
1950. Stratigraphic Tests in the Everglades National Park, American
Antiquity, vol. 15, pp. 228-246. Menasha, Wise.
Goggin, John M., and Frank H. Sommer, III
1949. Excavations on Upper Matecumbe Key, Florida, Yale Univers
ity Publications in Anthropology, no. 41, New Haven, Cnn*
Griffin, John W.
1950. Test Excavations at the Lake Jackson Site, American Antiquity,
vol. 16, pp. 99-112, Menasha, Wise.
Griffin, John W. and Ripley P. Bullen
1950. The Safety Harbor Site, Pinellas County, Florida, Florida
Anthropological Society, P ublications, no. 2. Gainesville, Fla.
Lewis, Bonita B., and Warren H. Wilkinson (translators)
N.D. La Florida of the Inca, History of the Adelantado Hernando de
Soto, Governor and Captain General of the Kingdom of Florida
and of other Heroic Spaniards and Indian Cavaliers. Typed
manuscript in Florida Park Service files.
Moore, Clarence B.
1900. Certain Antiquities of the Florida West-Coast, Journal of
the Academy of Natural Sciences of Philadelphia, N. S., vol. XI,
pp. 350-394. Philadelphia, Pa.
1901. Certain Aboriginal Remains of the Northwest Florida Coast,
Journal of the Academy of Natural Sciences of Philadelphia,
N. S., vol. XI, pp. 420-497. Philadelphia, Pa.
46


Small, John K.
1921. Old Trails and New Discoveries, Journal of the New York
B otanical Garden, vol. 22, pp. 25-40, 49-64. New York.
Smith, Hale G.
1948. Two Historical Archaeological Periods in Florida, American
Antiquity, vol. 13, pp. 313-319. Menasha, Wise.
Swanton, John R.
19 39. Final Report of the United States DeSoto Expedition Commission.
House Document no. 71, 76th Congress, 1st Session. Washington.
Walker, S. T.
1880. "Report on the Shell Heaps of Tampa Bay, Florida, Annual
Report of the Smithsonian Institution for 1879,pp. 413-422.
Washington.
Wilkinson, Warren H.
1947. The DeSoto Expedition in Florida, The American Eagle and
1948. Horticultural Review, vol. 42, no. 29 to vol. 43, no. 5. Estero,
Florida.
Willey, Gordon R.
1948. Culture Sequence in the Manatee Region of West Florida,
American Antiquity, vol. 13, pp. 209-218. Menasha, Wise.
1949. Archeology of the Florida Gulf Coast, Smithsonian Miscel
laneous Collections, vol. 113. Washington.
47


EXPLANATION OF PLATES
Plate I. Views of The Prine Mound and Abel Midden.
Upper, Burial 16, Prine burial mound; Lower, portion of face of Abel
midden.
Plate II. Miscellaneous Artifacts from Abel Midden, Taylor Collection.
A-E, shell pendants; F, worked columella; G, I, bone tools, spatulate
form; H, bone awl; J, bone fish hook; K, bone pendant; L, shell chisel; M-R,
chert projectile points.
Plate III. Miscellaneous Artifacts from Tests in Middens.
A, B, from Boots Point; balance from Abel midden.
A, B, sherds, Ft. Walton Incised; C-G, fragments of bone pins; H, perfo
rated Pecten shell; I, carved head of bone pin; J, Oliva shell bead; K,
pendant made from sherd; L, projectile point of chert; M, Strombus hammer;
N, shell spoon or concave variety (Busycon) celt; 0, Venus anvil; P, Busycon
hammer; Q, shell chisel.
Plate IV. Potsherds from Prine Mound.
A, Pinellas Plain; B, Lake Jackson Plain; C, Pinellas Incised B; D,
Pinellas Incised C; E-F, Safety Harbor Incised; G, cob-marked; H, Gainesville
Linear Punctated; I, adorno, Lake Jackson Type; J-P, Englewood Incised;
Q-R, Lemon Bay Incised; S, Unique Incised A; T, Unique Incised B; U,
Unique Incised C.
Plate V. Potsherds from Prine Mound.
A, St. Petersburg Incised (?); B-D, Weeden Island Incised; E-R, Engle
wood or Weeden Island Incised; G-K, Weeden Island Punctated; L-M, Papys
Bayou Punctated; N, Carrabelle Punctated; 0, St. Andrews Complicated
Stamped.
Plate VI. Potsherds from Prine Mound.
A, Carrabelle Incised; B-D, Swift Creek Complicated Stamped (Late
Variety); E-H, Tampa Complicated Stamped; I, Sun City Complicated Stamped;
J, Wakulla Check Stamped; K, Pasco Check Stamped; L, Biscayne Check
Stamped.
Plate VII. Potsherds from Prine Mound.
A, Ruskin Dentate-like Stamped; B, Ruskin Dentate Stamped; C-E, Hills
borough Shell Stamped (D, eroded surface); F-G, Thomas Simple Stamped;
H, Weeden Island Plain (notched rim); I-J, Weeden Island Plain, Pasco paste;
K, Biscayne Plain.
48


PUBLICATIONS, FLORIDA ANTHROPOLOGICAL SOCIETY: 3
PLATE
Views of the Prine Mound and Abel Midden


PUBLICATIONS, FLORIDA
ANTHROPOLOGICAL SOCIETY: 3
PLATE II
Miscellaneous Artifacts from Abel Midden, Taylor Collection


PUBLICATIONS, FLORIDA ANTHROPOLOGICAL SOCIETY: 3
PLATE III
Miscellaneous Artifacts from Tests in Midden:


PUBLICATIONS, FLORIDA ANTHROPOLOGICAL SOCIETY: 3
PLATE IV
Potsherds from Prine Mound


PUBLICATIONS, FLORIDA ANTHROPOLOGICAL SOCIETY: 3
PLATE V
Potsherds from Prine Mound


PUBLICATIONS, FLORIDA ANTHROPOLOGICAL SOCIETY: 3
PLATE VI
Potsherds from Prine Mound


PUBLICATIONS, FLORIDA ANTHROPOLOGICAL SOCIETY: 3
PLATE VII
Potsherds from Prirte Mound






PUBLICATIONS
FLORIDA ANTHROPOLOGICAL SOCIETY
Publications of the Society appear at irregular intervals as funds permit.
These are provided by the Sustaining Memberships and include, in some in
stances, special gifts or subsidies provided by the author. Publications are
distributed to all members of the Society on issuance. Additional copies may
be secured from the treasurer, as follows:
1. Two Archeological Sites in Brevard County, Florida. Hale
G. Smith. 31 pp., 2 figs., 4 pi. 1949. Price $0.50.
2. The Safety Harbor Site, Pinellas County, Florida. John W.
Griffin and Ripley P. Bullen. 42 pp., 2 figs. 4 pis. 1950.
Price $0.50.
3. The Terra Ceia Site, Manatee County, Florida. Ripley P.
Bullen. 48 pp., 6 figs., 7 pis. 1951. Price $0.50.
Address orders to the Treasurer, Florida Anthropological Society,
Florida Park Service, Seagle Building, Gainesville, Florida