Eleven archaeological sites in Hillsborough County, Florida ( FGS: Report of investigations 8 )

9

.1Io--
A.AA

j *, fS

J. CLARENCE SIMPSON, 1910 TO 1952.

: a ~

r

FLORIDA

STATE BOARD OF CONSERVATION

FLORIDA GEOLOGICAL SURVEY

Herman Gunter, Director

REPORT OF INVESTIGATIONS

NO. 8

ELEVEN ARCHAEOLOGICAL SITES
IN HILLSBOROUGH COUNTY, FLORIDA

By
Ripley P. Bullen
Assistant Archaeologist
FLORIDA PARK SERVICE

Published for
THE FLORIDA GEOLOGICAL SURVEY

Tallahassee, 1952

FLORIDA STATE BOARD

OF

CONSERVATION

DAN McCARTY
Governor

R. A. GRAY
Secretary of State

NATHAN MAYO
Commissioner of Agriculture

J. EDWIN LARSON
Treasurer

THOMAS D. BAILEY
Superintendent Public Instruction

CLARENCE M. GAY
Comptroller

RICHARD ERVIN
Attorney General

CHARLIE BEVIS
Supervisor of Conservation

IV

LETTER OF TRANSMITTAL

iloria geological Survei

Zallakassee

April 15, 1953

Mr. Charlie Bevis, Supervisor
Florida State Board of Conservation
Tallahassee, Florida

Dear Mr. Bevis:

During the years of 1935 to 1938, the late J. Clarence Simpson,
formerly of this department, was supervisor of rather comprehen-
,sive archaeological studies in Hillsborough County, Florida. These
studies were a part of the Works Progress Administration program,
and the data collected were never fully studied and reported upon.
Mr. Ripley P. Bullen, now Curator of Social Sciences at the
Florida State Museum, Gainesville, has recently compiled these
data and has made correlations of the various sites. This study is
being published as Report of Investigations No. 8 of the Florida
Geological Survey and it is a memorial to Mr. Simpson and to his
unselfish service to the State.

Respectfully,

Herman Gunter, Director

TABLE OF CONTENTS

Page
Letter of transmittal -- ---. --..................--..-.._----_---- .-- ----..----------- V
Dedication to J. Clarence Simpson ....--...--... --...---...--..----. ---- X
Eleven Archaeological Sites in Hillsborough County, Florida--- 1
Introduction -------.--..........--- --- .--.---- ...--..--...--....--.---.--. ---------. 1
Purpose of report .--.--.---.---- ------------------- 1
History of W.P.A. Projects 690 and 1928 ---..---...._---------- 1
Methods used in excavation .-...---........-----.. .---... --...--- 2
Limitations of data ........ -----.. ..... --....--.---.. --... ------ 3
Culture sequence in Tampa Bay region -.......---.--.---.--.----- 5
Archaeological problems -......------.. --... ---.------...-..-- 6
Excavations ...__---....--- .....--------------------- 7
Thomas mound .......---...... -- --...- -------.------.------------... 7
Cockroach Key ----..---.. ---.------------------.-- 20
Spender mound -..-....----.------------....-..-.-----.-- 25
Cagnini mound ...--..---..-...--..-.------ -------.---.- 26
Branch mound ---.....---.--.----.-....-----..--- .------ 31
Lykes mound ---..--.----... ..--.-.--.-..---..---.. -------- 33
Snavely mounds --.--...----. ---------------------.. 39
Jones mound -..--..... -------... --. -------...-....----.-------- 43
Picnic mound -... --... ------ .....--.---..--- ..----- -------------61
Sellner shell middens -...----....----.---------- --.---------------- 71
Buck Island -..--....--...__..-----------------------.------ 75
Conclusions -----...---. _--.-----.---------------------------.. 79
Bibliography .--...-..-- _...._...------...---..-....-.----------------- 84

VII

ILLUSTRATIONS

Figure Page
Frontispiece-J. Clarence Simpson ._._.._ .......------. ---------. II
1. Map of Hillsborough County and Tampa Bay Region, Florida 4
2. ,Sketch map of the Thomas site ____--------------- 8
3. Thomas mound, excavation plan and profile ----- -----------------. 10
4. Miscellaneous artifacts from Thomas mound -.-------------- ----- 14
5. Miscellaneous sherds from Thomas mound __----------------------- 16
6. Decorated bone dagger from Cockroach Key ----..------------. 21
7. Cagnini mound, excavation plan ---------------. 27
8. Branch mound, excavation plan -------------------------...-- 31
9. Lykes mound, excavation plan and profile ------------- 34
10. Flexed burial at Lykes mound ..........--------------------. 36
11. Snavely mounds, excavation plans __ ---------------- 40
12. Jones mound, sketch map and surface profile ---------- 42
13. Jones mound, excavation plan _--- --------------- 44
14. Excavations at the Jones mound ____-_------------------------- 46
15. Deerhead and birdhead stone pendants from Jones mound ----- 48
16. Duckbill type stone pendants from Jones mound --------- 50
17. Stone pendants, plumet type from Jones mound ------------------- 52
18. Shell pendants and Cassis lip from Jones mound -- ------ 54
19. Miscellaneous artifacts from Jones mound ------------------------. 56
20. Pottery from Jones mound ..--_..... ___---- --------------- 58
21. Picnic mound, excavation plan and profile ------------------------ 62
22. Miscellaneous artifacts from Picnic mound ---------- 66
23. Miscellaneous sherds from Picnic mound -----------------------.-- 68
24. Miscellaneous artifacts from Buck Island -------------------------'- 76

TABLES

Table Page
1. Vertical distribution of burials at Thomas mound -- ----11
2. Vertical distribution of burials at Cockroach Key ------.-------- 22
3. Vertical distribution of burials at Lykes mound ...-------- ------37
4. Vertical distribution of burials at Jones mound .------- -.--------. 47
5. Stone pendants and burial types at Jones mound ------- 49
6. Vertical distribution of burials at Picnic mound ..... .--- ------_ 64

IX

J. CLARENCE SIMPSON

(1910-1952)

To the staff of the Florida Geological Survey and to his many
friends the sudden passing of James Clarence Simpson on March
29, 1952, at his home in Marianna, Florida, came as a great shock.
Courageously and cheerfully he had struggled against ill health
even after the discouraging day his loved ones were told he could
live only a few weeks or months. With renewed determination,
however, he accepted the fighting chance offered'through unwaver-
ing adherence to a very restricted prescribed diet. Thus he added
four to five years to his life span. Through it all he worked untir-
ingly and enthusiastically at his loved work as a naturalist, spend-
ing his last day in the interesting activity of mounting a portion
of the skull and lower jaw of a four-tusked mastodon. Then bid-
ding us a cheery adieu at the close of the day, he went out. That
night he lay down to rest in his home and just slept on past rising
time in the morning.
Clarence was born in 1910 at Micanopy, Alachua County, Flor-
ida, the son of Katie Mathers Simpson and the late Henry H.
Simpson of High Springs. "Bruce," so known to his boyhood and
intimate friends, received training in the High Springs elementary
and high schools graduating in the class of 1929. Through these
early years he manifested unusual interest in natural history,
which interest dominated his later life and resulted in an amazing
fund of knowledge relative to archaeology, geology, the fauna and
flora of his native State and natural history in general.

In 1930 Clarence joined the Florida Geological Survey and re-
mained on the staff almost continuously until his death. Although
denied the advantages of formal training, Clarence was well
equipped for life's endeavors through his keen sense of observation,
inquisitive mind, natural intuitiveness and self-schooling in those
subjects he loved most-archaeology, botany, entomology. He was
always happy in sharing his vast store of knowledge, and in show-
ing his treasured collections of artifacts, vertebrate fossils and ob-
jects of natural history. His enthusiasm for the out-of-doors and
his insatiable desire to explore resulted in an intimate familiarity
with natural history, and a knowledge of the many fossil collecting
localities of Florida. He brought to the attention of scientists such
X

localities as Thomas Farm in Gilchrist County, the Itchtucknee
River, the Santa Fe River for vertebrate fossil collecting and the
many archaeological sites within that area, as well as in other parts
of Florida. His short life was crowded to the last with construc-
tive activities and faithfully performed services.

In 1936 at Tampa, Florida, Clarence was married to Zelma
Harris, and to this union three children were born-Bruce 14,
Genevieve 5, and Jo Ann 3, all of whom survive. He also leaves
his mother, Mrs. Katie Mathers Simpson, one brother Harry Hor-
ton Simpson of High Springs, one sister Mrs. Dorothy Simpson
Baer, Gainesville.
-Herman Gunter.

Reprinted from the Society of Vertebrate Paleontology News Bulletin,
October, 1952.
XI

ELEVEN ARCHAEOLOGICAL SITES IN HILLSBOROUGH
COUNTY, FLORIDA
Ripley P. Bullen

INTRODUCTION

PURPOSE OF REPORT
Between November 2, 1935, and March 31, 1938, an archaeo-
logical survey was made of Hillsborough County, Florida, and ex-
cavations conducted at eleven sites under two Works Progress
Administration projects. Results of these excavations have never
been fully reported although J. Clarence Simpson, field supervisor
of both projects, briefly discussed some of the salient features in
two progress reports (Simpson, 1937; Anonymous- (J. Clarence
Simpson), 1939). Gordon R. Willey, also, has discussed two of
the sites, Thomas and Cockroach Key, in some detail, briefly de-
scribed three and mentioned five others in his Archeology of the
Florida Gulf Coast (Willey, 1949, pp. 113-125, 158-171, 335-9).

The Archaeological Survey of the Florida Board of Parks and
Historic Memorials has become the repository of most of the
original field notes. This places upon it an obligation to report, in
so far as possible, the results of this work. It is believed publica-
tion will salvage knowledge which otherwise would be lost and will
present data not included by Willey on the Thomas mound and
Cockroach Key.

HISTORY OF W. P. A. PROJECTS 690 AND 1928
On November 2, 1935, W.P.A. Archaeological Project No. 690
opened with a personnel of approximately 90, including the super-
visor. This project was written and directed by Vernon Lamme,
then State Archaeologist, for the purpose of investigating Indian
remains in Hillsborough County, Florida. It was sponsored by the
State Archaeologist in cooperation with the Smithsonian Institu-
tion, Washington, D. C., the South Florida Archaeological Research
Society at Miami, and the Florida Historical Society.
On December 11, 1935, the project was reorganized and trans-
ferred by Executive Order to the State Board of Conservation, and
J. Clarence Simpson, of the Florida Geological Survey, was placed
in charge of field activities. The Smithsonian Institution co-
operated by sending Preston Holder to supervise field techniques.

FLORIDA GEOLOGICAL SURVEY

By mutual agreement, half of the material secured was to go to
the Smithsonian Institution and half to the State of Florida. The
Geological Department of the University of Tampa, then under the
direction of Professor Robert F. Webb, cooperated by providing
storage and exhibition space and technical services in the labora-
tory.

Under this arrangement two sites, the Thomas mound and
Cockroach Key, were partially excavated. On April 14, 1936, the
project was closed for lack of funds. Specimens were divided with
the Smithsonian Institution according to agreement and those going
to the State of Florida were deposited in the museum of the Uni-
versity of Tampa as a loan until such time as other provision could
be made.

On May 12, 1936, the project was reopened and continued as
before except for the withdrawal of the Smithsonian Institution.
J. Clarence Simpson, as field supervisor, continued the cataloging
system and field techniques started by Preston Holder. Excava-
tions were conducted at the Spender, Cagnini, Branch, and Lykes
mounds. Work at the last site was terminated on September 29th,
1936, but the supervisor and two clerks were maintained until
November 1, 1936, to get all notes, charts, and other details finished
and filed, when Project 690 was closed.

A new W.P.A. project, No. 1928, was approved and work started
January 11, 1937, with a personnel of 34, including the field super-
visor, J. Clarence Simpson. Under this project the Snavely, Jones,
Picnic, Sellner shell midden, and Buck Island mounds were ex-
cavated, additional work was done at the Thomas mound, and two
weeks were spent screening the spoil bank of a new canal for
fossils. Forty-eight Indian sites in the County, including mounds,
midden deposits, and chert quarries were surveyed, recorded by
brief descriptions, and located on large scale maps. During most
of the work at Sellner's and at Buck Island, William G. Southerland,
the foreman, acted as field supervisor while Simpson arranged for
the Buck Island excavations and worked on closing the project
which was completed March 31, 1938.

METHODS USED IN EXCAVATIONS

Holder's field techniques, also used by Simpson, included staking
the area to be excavated into a convenient grid of 4- or 5-foot

REPORT OF INVESTIGATIONS No. 8

squares, recording of contours, taking of profiles, and keeping a
field note book. Burials and important specimens were located
horizontally by reference to the grid and vertically by depths in
the ground. Whenever possible, individual sketches were made of
each burial. All deposits that contained culture material were
screened.

The presence of brow ridges and poor development of genial
tubercles were used by the excavators to indicate the male skeleton.
The large number of male skeletons relative to those of females
reported in the following data is probably the result of this inac-
curate method of sex determination.

At the time these excavations were conducted, pottery types
and their implications were practically unknown in Florida archae-
ology. Consequently, little attention seems to have been paid to
sherds unless they were especially distinctive. In some cases, espec-
ially at the thick Sellner midden, material was removed by arbitrary
zones so that a chronology based on variations in artifacts occur-
ring with depth could be established. Unfortunately, material has
not been preserved over the years in that manner.

LIMITATIONS OF DATA

Many specimens and the data which accompanied them have
become lost in the sixteen years since excavation. We have had to
rely on field note books for the little information which is avail-
able regarding a chronology based on changes in pottery and pro-
jectile points. Fortunately, Simpson appreciated the difference
between Weeden Island and Safety Harbor-then called "decadent
Weeden Island"-pottery decoration and realized that narrow tri-
angular arrow points (sometimes called "bird points") were a
very late phenomena. His comments in these respects, both in the
field notes and in his preliminary reports, have been very helpful.

The data from W.P.A. excavations in Hillsborough County have
been so poorly preserved that we have samples of pottery from
only four of the mounds, and of the great many specimens of
chipped stone which were unearthed only three are available.
Fortunately, Simpson took the unique polished stone pendants and
certain other important specimens, when he returned to the Florida
Geological Survey, and deposited them with that agency or they
would probably have disappeared also. It was only the personal

FLORIDA GEOLOGICAL SURVEY

HILLSBOROUGH CC

I-THOMAS
2-COCKROACH KEY
3-SPENDER
4-CAGNINI
5-BRANCH
6-LYKES

)UNTY SITES

7-SNAVELY
8-JONES
9-PIGNIC
10-SELLNER
I -BUCK ISLAND

-OTHER SITES

12-PERICO ISLAND
13-WEEDEN ISLAND
14-SAFETY HARBOR
15-TERRA CEIA
16-PARRISH

Figure 1.-Map of Hillsborough County and Tampa Bay Region, Florida.

REPORT OF INVESTIGATIONS NO. 8

interest of J. Clarence Simpson that preserved these specimens and
the field notes and makes the writing of this report possible.

CULTURE SEQUENCE IN TAMPA BAY REGION

In this paper the data are described and discussed in terms of
the culture sequence at present applicable to the Tampa Bay Region
(Willey, 1948, 1949; Goggin, 1950; Bullen, 1951). This sequence,
while only relatively recently formalized, is based on work done in
1923-24 at Weeden Island near St. Petersburg, in 1930 at Safety
Harbor, and in 1933-34 at the Parrish mounds and at Perico Island
west of Bradenton. Figure 1 presents a map of the Tampa Bay
region on which these sites as well as those in Hillsborough County
covered by this report are located. The evident proximity of all
of these sites makes the comparison proper.

The culture sequence, slightly modified, follows:

Seminole
1700 A.D.
Historic Safety Harbor
1500 A.D.
Prehistoric Safety Harbor
1400 A.D.
Weeden Island II
1100 A.D.
Weeden Island I
700 A.D.
Perico Island
100 B.C.
Fiber-tempered pottery (?)
2000 B.C.
Preceramic Archaic (?)
?? B.C.
Palaeo-Indian or Folsom-like (?)
8000 B.C. (?)

Dates given for the above sequence should be considered merely
as current estimates, subject to future change. The three earliest
periods are represented in this region only by occasional finds of
diagnostic artifacts and have not, as yet, been found as separate
and complete entities.

In order that readers may be acquainted with artifact com-
plexes representing the periods since the time of Christ, certain
diagnostic traits will be mentioned here.

The division between Historic Safety Harbor and Prehistoric
Safety Harbor is based on the absence of items of European origin.
2

FLORIDA GEOLOGICAL SURVEY

Otherwise, Safety Harbor times are characterized by narrow tri-
angular arrow points, and by the Safety Harbor Incised, Pinellas
Incised, Lake Jackson Plain, and St. Johns Check Stamped pottery
types (Griffin and Bullen, 1950). Temple mounds as well as burial
mounds are found and interments are nearly always of the secon-
dary or bundle type.

As defined at present, Weeden Island refuse containing
sherds from Wakulla Check Stamped vessels but none from Swift
Creek Complicated containers would be representation of Weeden
Island II while the reverse would be considered Weeden Island I
(Willey, 1949). This distinction, established in the northwestern
part of the State, may not be applicable to Hillsborough County
and the Tampa Bay region. It seems best, however, to continue a
division of Weeden Island into two parts as there must have been
an early and a late Weeden Island period even if Wakulla Check
Stamped pottery is not the proper marker.

Important pottery types of the Weeden Island periods include
Weeden Island Plain, Weeden Island and Papys Bayou Punctuated
and Incised, Carrabelle Punctuated and Incised, Hillsborough Shell
Stamped, Dunns Creek Red, and Wakulla and St. Johns Check
Stamped. The last two were also made in the succeeding Safety
Harbor period. Stone axes or celts and plummet-shaped pendants
of shell and stone are also present in Safety Harbor time but they
are more common in the preceding Weeden Island period. Burial
mounds, containing various types of interments, should be found
at a Weeden Island site, but temple mounds should not be associated
with this culture.

The Perico Island period is by far the least known of the ceramic
periods. For purposes of this report it will be considered as an
early pottery making period during which vessels were only oc-
casionally decorated. Sometimes decorated sherds from the Dept-
ford and Swift Creek complexes to the north are found. Burials
are of primary (in the flesh) types.

ARCHAEOLOGICAL PROBLEMS

Most of the material excavated in Hillsborough County by the
W.P.A. projects pertains to the Weeden Island and Safety Harbor
periods. The latter, which is pre-Seminole, represents the culture
of the Timucua Indians who were found in the Tampa Bay region

REPORT OF INVESTIGATIONS NO. 8

by early Spanish explorers. It is the archaeological expression of
the end result of aboriginal influences entering the area from the
northwest and acting upon Indians who lived in a cultural manner
which we refer to as Weeden Island. It is this change from Weeden
Island to Safety Harbor upon which data from the excavations
particularly shed light.

In his recent Archeology of the Florida Gulf Coast, Gordon
R. Willey noted the presence of European trade objects and Safety
Harbor period pottery in burial mounds that but for these objects
would be classified as Weeden Island. He considered such specimens
to be intrusive-late intrusions into or reuse of an older mound.
Frequent repetition of this phenomena suggests rather that Indians
of Hillsborough County continued to use the same burial mounds,
without any hiatus, after their material culture had changed from
that of Weeden Island to that of Safety Harbor.

Data from the W.P.A. excavations in Hillsborough County,
while oftentimes incomplete, bear upon these problems and in-
crease our knowledge of past inhabitants. If publication could
have occurred shortly after excavation, when more of the material
was available, we would know much more about the prehistory of
the Tampa Bay region. It is only by piecing together this scattered,
poorly recorded, and incomplete evidence that an understanding
of the life of and cultural forces at work upon the inhabitants can
be made clear.

THOMAS MOUND

The extensive and complex Thomas site is located on the north
side of a large bayou which joins the Little Manatee River about a
mile and a half west-northwest of Ruskin (Fig. 1). Clarence B.
Moore tested the burial mound at this site in 1900. The first W.P.A.
project excavated part of this mound in December, 1935, and Janu-
ary, 1936, and the second W.P.A. project, in 1937, completed ex-
cavation of the Thomas burial mound and made tests in other parts
of the site. The W.P.A. projects will be referred to respectively
as the first and second visits.

Figure 2 presents a plan of the site made during the second
visit. Originally a large shell deposit or midden bordered the bayou
and residences (F and F) are situated on large remnants of this
shell heap. "E" and "E" are listed as "leveled sand-shell mounds"

ROAD

0

LITTLE

MANATEE RIVER

A-THOMAS BURIAL MOUND
B-SMALL SAND AND MIDDEN

C-LARGE

MOUNDS

SAND AND MIDDEN MOUND

D-LOW SAND MOUND
E-BASAL PORTIONS OF SHELL
F-HOUSES ON SHELL HEAPS

Figure 2.-Sketch map of the Thomas Site (not to scale).

0
L'i
0
0

0
0
cc

HEAP

REPORT OF INVESTIGATIONS No. 8

and represent basal portions of the shell heap. "B" and "B", small
sand and midden mounds, may be the remains of landward exten-
sions of the shell heap. Toward the north are shown three sand
mounds of which "A" is the burial mound and "C" a residential
mound or the beginnings of a temple mound. Data are not available
concerning "D."

The Thomas burial mound (Fig. 2, A) was about 60 feet in
diameter and six feet in height. According to Moore (1900, p.
358) an aboriginal canal connected the southwest side of this
mound with the bayou, 238 feet distant. The canal was 64 feet
wide near the mound and 26 feet wide where it joined the bayou.
Probably this canal started as a borrow pit to supply material
used in the construction of the burial mound. On the plan another
canal, terminating at the end of a sand ridge, is shown to the east.
This second canal is not mentioned by Moore whose activities were
limited to the burial mound. Examination of the site in 1952
strongly suggested both canals to be drainage ditches dug by
early settlers for farming purposes.

Figure 3 is the excavation plan of the second visit and locates
the burials found during that visit as well as indicates the area
excavated during the first visit and by Moore. Well over 400 burials
were removed during the three excavations. Moore (1900, p. 358),
referring to the 112 burials he uncovered, writes:

"The prevailing form of interment was a squatting position, the feet
on a level with the pelvis, the legs against the thighs and these drawn
up against the body. The upper arms were against the sides with the
forearms sometimes parallel to the upper arms and sometimes on the
chest, reaching to the neck. The head was bent over and forced between
the thighs, sometimes to the pelvis. Certain [other] skeletons lay on the
side with'the same general arrangement of the extremities and the skull
pressed over against the knees."

Moore also found a shell cup, a shell bead, a chert spear point
or knife, a small hammerstone, a smoothing stone, a worked fossil
shark's tooth, blue glass beads, two pieces of looking glass, three
stone pendants, and a quantity of sherds. These sherds were us-
ually undecorated but some bore incised and some punctated mark-
ings while one exhibited a stamped design (Moore, 1900, p. 359).
One of the stone pendants represented the head of a bird, but un-
fortunately, the bill was missing (Willey, 1949, p. 123, Fig. 15).

Results of the first W.P.A. visit and, to a lesser extent, those

78 FEET

EXCAVATION PLAN- SECOND VISIT

1S SECOND
! MOUND

DARY PRI
PRIMARY MOUND

PROFILE FROM FIELD

-.CREMATION g.
*-BUNDLE BURIALS 4+
L- FLEXED BURIAL FS

NOTES -NOT TO SCALE

-EXTENDED BURIIALS
* ISOLATED SKULLS
*EUROPEAN DERIVED SPECIMENS

Figure 3.-Thomas mound, excavation plan and profile.

I1 I

_-1

._ I ll i

_ ~__ ___ _

REPORT OF INVESTIGATIONS NO. 8 11

TABLE 1.
VERTICAL DISTRIBUTION OF BURIALS AT THOMAS MOUND
Depths in feet
Burial type
0-1 1-2 2-3 3-4 4-5 5-6
First visit
Cremation .. ...... .......... 1
Vertical bundle .. ............... 11 29 14 4 21
Horizontal bundle .......... 3 8 11 7 4J
Flexed ......-.............. ..... .... .. 4
Semi-flexed ----.... ..--.--. 3
Isolated skulls .-......- ... ... 1 1
Second visit
Cremation ..-..1.--. .... -......... 1 13
Vertical bundle .....-... .. .._ 15 40 29 3
Horizontal bundle -.. ___......... 4 8 7 2 74
Flexed _.................... 1...... __.. 1 5 17 21 30
Semi-flexed .---...-...... .. ....-.- ....- 15 2 3 3
Isolated skulls ---..---......____. 2 14 21 23 12
Extended ....-----........--..........._ 7
0One below limestone block.
20f four lowest, one 9 inches above flexed burials, 3 doubtful as to type.
3Listed as "decomposed cremation."
4One badly disturbed, five listed as "decayed bundle burials."
5Badly disturbed.

of the second visit, have been reported by Willey (1949, pp. 115-
125). Both visits uncovered the same types of burials as reported
by Moore, vertical bundle and horizontal flexed interments, with
the additional information that flexed burials were the deeper and,
hence, the earlier type. The superposition is shown in Table 1 for
burials upon which the requisite information is available. As sug-
gested in Table 1, the extended burials appear to have been the
earliest form at this site, followed by flexed and semi-flexed in-
terments which, in turn, were superceded by vertical and horizontal
bundle burials. The habit of interring isolated skulls appears to
occupy no definite position vertically and the number of cremations
are too few for any chronological implications. Excavations during
both' visits found burial pits that had penetrated other burial pits,
showing that these interments were not all made at one time.

Pottery'sherds from both the Weeden Island and Safety Harbor
periods are present in all collections from this site. Except for an
association between check-stamped sherds, Wakulla or St. Johns
Check Stamped, and the cremation at a depth of 30 inches and for

FLORIDA GEOLOGICAL SURVEY

the finding of part of a square red vessel at a depth of 59 inches,
we have no data regarding the vertical distribution of pottery types
found during the second visit. Willey, who had such information
on 1143 of the 7746 sherds found during the first visit, noted no
vertical differences except that pottery of the Safety Harbor period
came from not far below the surface of the mound (Willey, 1949,
p. 119).

During the first visit it was observed that the sherds, while
found throughout most of the mound, were concentrated in two
zones which at the center were 24 to 36 and 42 to 50 inches below
the surface.

A narrow curving deposit of midden material was also noted
at a depth of 42 to 50 inches near the northern edge of the mound
(Fig. 3, Sections 63 to 153 to 184). Subsequently, during the second
visit, this deposit was found to extend southwesterly to Section
202 and northwesterly to the middle of Section 49. Simpson's field
notes suggest this deposit may have marked the edge of a primary
mound.

These slight indications of physical stratigraphy in the Thomas
mound together with the superposition of different types of burials
suggest various building periods or superimposed mounds. On this
point Willey quotes from field notes written during the first visit:

"While suggestions appeared in the field to support a thesis for the
existence of two mounds, one of which was an older underlying mound
about 20 to 30 feet in diameter with a rise of perhaps 2 feet which
had subsequently been covered over by the present mound, it is doubtful
that the evidence recovered will warrant such an hypothesis. The evi-
dence of definite (physical) stratigraphy was disappearingly faint."
(Willey, 1949, p. 116).

That such may, however, have been the case is indicated by two
items found among the field notes of the second visit. One is a
penciled cross-section, presumedly not made to scale, which we have
traced, reduced, and reproduced in Figure 3. This sketch bears the
notation, "Cross Section from N to S beginning in Sec. 151 and
continuing S to Sec. 153. Showing older mound with newer mound
superimposed."

The other is a note which reads, "All burials in Sections 170-171-
172-173-174-185-186-187-188 were uncovered at the same time-
showing Recorded Burials 141-142-143-144-145-147-148 to be on

REPORT OF INVESTIGATIONS NO. 8

by early Spanish explorers. It is the archaeological expression of
the end result of aboriginal influences entering the area from the
northwest and acting upon Indians who lived in a cultural manner
which we refer to as Weeden Island. It is this change from Weeden
Island to Safety Harbor upon which data from the excavations
particularly shed light.

In his recent Archeology of the Florida Gulf Coast, Gordon
R. Willey noted the presence of European trade objects and Safety
Harbor period pottery in burial mounds that but for these objects
would be classified as Weeden Island. He considered such specimens
to be intrusive-late intrusions into or reuse of an older mound.
Frequent repetition of this phenomena suggests rather that Indians
of Hillsborough County continued to use the same burial mounds,
without any hiatus, after their material culture had changed from
that of Weeden Island to that of Safety Harbor.

Data from the W.P.A. excavations in Hillsborough County,
while oftentimes incomplete, bear upon these problems and in-
crease our knowledge of past inhabitants. If publication could
have occurred shortly after excavation, when more of the material
was available, we would know much more about the prehistory of
the Tampa Bay region. It is only by piecing together this scattered,
poorly recorded, and incomplete evidence that an understanding
of the life of and cultural forces at work upon the inhabitants can
be made clear.

THOMAS MOUND

The extensive and complex Thomas site is located on the north
side of a large bayou which joins the Little Manatee River about a
mile and a half west-northwest of Ruskin (Fig. 1). Clarence B.
Moore tested the burial mound at this site in 1900. The first W.P.A.
project excavated part of this mound in December, 1935, and Janu-
ary, 1936, and the second W.P.A. project, in 1937, completed ex-
cavation of the Thomas burial mound and made tests in other parts
of the site. The W.P.A. projects will be referred to respectively
as the first and second visits.

Figure 2 presents a plan of the site made during the second
visit. Originally a large shell deposit or midden bordered the bayou
and residences (F and F) are situated on large remnants of this
shell heap. "E" and "E" are listed as "leveled sand-shell mounds"

REPORT OF INVESTIGATIONS NO. 8

the west side of sloping ground." The sloping ground is further
delineated as 30 inches below the mound surface at the eastern
edge of Sections 171-172 and 54 inches below the mound surface
at the western edge of Sections 186-187. This sloping ground may
have once been the surface of a primary mound. If so the dirt sur-
rounding these burials, all of which were flexed except for one skull
and long bones in a disturbed area, should represent an accretion
to the mound.

Accretional building of the mound is also suggested by the fol-
lowing statement found in an original field notebook, written, ap-
parently, by Jimmie Redding:

"In the NW 1/4 of the plat (as much as excavated) two levels of pot-
tery appear with fair regularity indicating two different periods [of]
occupation [construction] of the mound. The upper most level shows
high frequency of pat[t]ern[ed] pottery and occurs at 24 to 36" at
peak of mound. The lower most level shows more black ware (cooking
pottery?) and occurs 42 to 50"."

To summarize the foregoing,,the Thomas burial mound appears
to have been built by accretion and may have consisted of a pri-
mary and a secondary mound. During its construction burial
habits changed from extended to flexed and then to bundle inter-
ments while pottery increased in percentage of decorated wares
and the Safety Harbor period pottery types were relatively late.

The 1143 sherds, collected during the first visit, stored and cata-
loged in the U.S. National Museum, have been classified by Willey
(1949, pp. 119-120). The collection includes many Weeden Island
types, eight Safety Harbor Incised, one Pinellas Plain, and one
Pinellas Incised. The most common are the Weeden Island Punc-
tated, Incised, and Plain with the St. Johns and Wakulla Check
Stamped being less common.

Some of the pottery collected during the second visit is stored
at the Florida State Museum in Gainesville, having been trans-
ferred there, February 24, 1939, by the Florida Geological Survey.
Dr. John M. Goggin, now of the University of Florida, classified
this collection some years ago and the results were presented by
Willey (1949, p. 121). Changes in and additions to pottery clas-
sifications have occurred since this work was done. Our classifica-
tion of this collection, Florida State Museum Cat. Nos. 76502-76638,
Ace. No. 3422, follows:

FLORIDA GEOLOGICAL SURVEY

4

** c h

C 0

IN geg
IN OHS$ >'^:

^ ''.i~ v ^r^ :: ^^^^^^

';~ 1 W^ -

IS

,, i. rV

.:. "' (
S ,'

;.

' i ;''.*r -^
* '*^^.iqt.w

7;
-I.. .

'

r:,it iii. :
("

;
~- 2r.l
i \;-
~.t(il-;, ,:s
'I
i:~" .
..

itt, i
cr

i
~d ii
~-~:"r, 1
I '

i I
i
~ I:?

'i : -.'

- ~

r .
.
: :
1:ri
I. r : '

;) :(B

''
?il i~ ~

~P
,
..: r; -
:-~
*t i: *Y

1/1

1~

~iC'~
,.:i~rb~iL;I~~ r~-r

REPORT OF INVESTIGATIONS NO. 8

Restored
Safety Harbor Period Sherds Vessels
Safety Harbor Incised -------------------............. .. 8 1
Pinellas Incised -----------------------.---------.... 2 1
Ft. Walton Incised --.--....-----....-.------------ 4
Safety Harbor or Weeden Island Period
Englewood Incised ....---..--..--- ----------...... .... 1
Wakulla Check Stamped ....----------------..... 32 1
St. Johns Check Stamped ......----------------..._ 14
St. Johns Plain ----..----------.....................---- 14
Belle Glade Plain ....-----...._-----------------_. 1
Weeden Island Period
Weeden Island Plain ....-----.---- --.--------- 3
Weeden Island Incised ......------.-..-------------_ 30
Weeden Island Punctated .....----.-----..-...---. 26 1
Weeden Island Zoned Red ......----.----..---_ 2
Carrabelle Incised (one with Pasco paste) --- 2
Carrabelle Punctated ---.......-----------------. 4
Hare Hammock Surface Indented .-----------. 2
Sun City Complicated Stamped -..--.---.----- 1
Hillsborough Shell Stamped ____-..--------.----. 7 1
Swift Creek Complicated Stamped
(Late Variety) ..-----.....----..._..-----...--.--- 3
Tampa Complicated Stamped ....------------- 4 1
Prairie Cord-marked ------- 1
Gainesville Linear Punctated ...--------------- 1
Papys Bayou Punctated (two red painted) 19
Papys Bayou Incised -...---..--...------. --------. 2
Dunns Creek Red -----__...--... ---.----- -------- 14
Pasco Plain --.---...-......-......---....-----.---.---. 20
Miscellaneous
Unclassified complicated stamped--------- 2
Unclassified incised or punctated ..-..---------- 9
Unclassifiable (eroded surface) ---- 1
Plain Red ..------..-...-.---------------------------- 3
Residual ..._-------_ ......-- --------.------ 5
Totals 234 9

These sherds include examples of duck bills as ornamental de-
tails (Fig. 5, j) and several necks of water bottles. Some of the
restored vessels are rather small. The Safety Harbor Incised con-
tainer has an incised band around the neck reminiscent of Swift
Creek Complicated Stamping (Fig. 4, k).

The field notes include pen and ink drawings of sherds, found

Figure 4.-Miscellaneous artifacts from Thomas Mound: a, incised brass
tablet; b, same, incision indicated by ink; c-d, stone pendants, plummet type;
e, fossil shark tooth pendant; f, stone pendant, bird effigy type; g, Weeden
Island Incised sherd; h, silver pendant; i, restored vessel, Prairie Cork Marked,
9% inches across; j, portion of Weeden Island Punctated vessel, 7Y2 inches
across; k, portion of Safety Harbor Incised vessel, 5 inches across. a-f, Florida
Geological Survey, balance Florida State Museum. Florida Park Service and
Florida State Museum photographs.

15

FLORIDA GEOLOGICAL SURVEY

A 8

O 0

0 1 ~g

SINCHI

0.

1:1
S : .. .<.-';,,

., i: 1
I Ii
K r

1 :
.
~: ,
C
1 ''

' -- '~:
:r
1

REPORT OF INVESTIGATIONS No. 8

during both visits. These illustrations include designs of Kaith
Incised, St. Petersburg Incised, Oklawaha Plain, and Sarasota
Incised not listed above.

Other specimens from both W.P.A. excavations are listed by
Willey (1949, pp. 122-4). We have added a few from the field
notes. Stone specimens include three narrow, triangular arrow
points (Safety Harbor type), twelve stemmed points, two drills,
one scraper, two abrading stones, several tubular steatite beads, and
seven pendants. The pendants comprise one steatite and four
quartz (Fig. 4, c-d) plummet type pendants, one fragmentary
pendant, the beak of a bird-effigy pendant (Willey, 1949, p. 123,
Fig. 15), and a nearly complete duck-effigy pendant (Fig. 4, f).
To this list should be added the bird-effigy pendant found by Moore
(Willey, 1949, p. 123, Fig. 15).

Shell tools and artifacts were represented by various fragments
including worked columellae and clam shells, a Busycon cup or
bowl, Venus shell anvils and seven shell beads. A worked shark's
tooth (Fig. 4, e) was also found.

European contact is indicated by glass beads and two fragments
of looking glass found by Moore and by about 200 glass beads,
a triangular piece of sheet copper, measuring three inches on each
side, a tubular silver bead, a brass pendant or tablet (Fig. 4, a-b)
and a larger silver pendant (Fig. 4, h) found during the second
visit. The smaller pendant is also illustrated by Willey (1949, p.
124, Fig. 16), but it is not certain that figure 4 a-b is the right or
left of Willey's figure. If figure 4 a-b is the same as the left of
Willey's figure 16, it would correct the outline and the incised draw-
ing; if figure 4 a-b is the same as the right of Willey's figure, it
would add the design and a hole for suspension.

During the second visit, a stone bead, 50 mm. long and 14 mm.
in diameter, and a small quartz pendant was found associated with
vertical bundle burials, and an egg-shaped abrading stone and a
drill iere recorded with flexed burials.

Figure 5.-Miscellaneous sherds from Thomas Mound: a, Pinellas Incised;
b, c, Safety Harbor Incised; d, Hare Hammock Surface-Indented; e, Ft. Walton
Incised; f, Hillsborough Shell Stamped; g, Englewood Incised; h, Tampa Com-
plicated Stamped; i, St. Johns Check Stamped; j, Weeden Island Plain; k-m, q,
Weeden Island Incised; n, Weeden Island Zoned Red; o, p, r, Weeden Island
Punctated. Florida State Museum photographs.

17

FLORIDA GEOLOGICAL SURVEY

Skeletal material was extremely poorly preserved and very
little about physical types at the Thomas site could be determined.
Apparently, they were mesocephalic to brachycephalic with, in
some instances, pronounced prognathism and heavy supraorbital
ridges. Field designations of age and sex simply imply a normal
distribution in these respects except that data from both visits in-
dicate nearly three times as many men as women. In several cases
an infant's skull was buried with an adult suggesting difficulties
of child-birth may have been a factor in the surplus of males.

While back-filling the Thomas mound during the second visit,
five exploratory trenches were dug in other parts of the site. At
Mound C (Fig. 2) some midden material and a sherd were found on
the north side and a midden area, five by three feet, composed of
oyster shells, fish bones, and dark sand was present on the east side.
This low sand mound probably was the site of an Indian home,
possibly the beginning of a temple mound. Small midden areas
B and B produced an arrow point and a rough abrader. Results of
tests at D are not given in the field notes.

SUMMARY

The Thomas site was a complex unit with extensive shell de-
posits, midden areas, a large burial mound, and two smaller mounds.

At the burial mound vertical and horizontal bundle burials over-
lay primary flexed interments. Pottery, while tending to concen-
trate at depths of 30 and 46 inches, was found throughout the
tumulus. Most of this pottery is typical of the Weeden Island
period, but some sherds, found near the top of the mound, represent
the Safety Harbor period. Stylistically some of the Weeden Island
pottery foreshadows later Safety Harbor ceramics. Projectile
points pertain to both the Safety Harbor and Weeden Island
periods. No pottery or other artifact attributable to pre-Weeden
Island times was uncovered.

The evidence suggests that the Thomas mound was used over
a long period of time and additions to the structure were made
periodically. Flexed burials of the lower zone were probably in-
terred during a Weeden Island I period or between 700 and 1100
A.D. The bulk of the pottery and most of the bundle burials are
Weeden Island II in date, about 1100 to 1400 A.D. People were still
living at the site and making additions to and burying in the mound

REPORT OF INVESTIGATIONS NO. 8

in Safety Harbor times to judge from the pottery and triangular
projectile points of that period.

The metal and glass found by Moore and during the second
W.P.A. visit raises the question as to whether the community was
still a going concern after the discovery of America by Columbus.
Specimens are of types used by Indians in Florida during early
post-Columbian times but differ from those known for the later
Leon-Jefferson (Spanish mission) period of north Florida.

While depths, associations, and other details regarding Euro-
pean derived materials from the Thomas mound are not available,
field notes indicate their horizontal distribution. Moore's beads
must have come from the area he dug, presumedly the southern
part. All these locations, each separated by about 10 feet, have
been shown on the excavation plan (Fig. 3) by the letter "E."
This distribution supports a theory of continued occupation of the
site in post-Columbian times as opposed to an intrusive interment
after the site had been abandoned.

Thus it would seem that the Thomas site was occupied and its
burial mound used from Weeden Island I times into the contact
part of the Safety Harbor period. This would mean a life of about
900 years or from about 700 A.D. to 1600 A.D. A relatively small
but stable community over this period would readily account for
the extensive remains.

A somewhat different temporal interpretation is possible which,
however, might omit a Weeden Island I period. In a preliminary
unpublished report on the second visit Simpson wrote, probably
during the winter of 1937-8:
"Potsherds were fairly abundant in both levels. Level 1 [the lower]
contained less pottery than Level 2 and pottery from Level 1 was pre-
dominantly plain and crude, largely cooking ware bearing heavy soot
accumulations.
"Pottery from Level 2 was of a superior workmanship and preser-
vation. The ware itself was in all respects better. Most of the sherds
from this level could be classed as Weeden Island type with a good deal
of degenerate Weeden Island or Safety Harbor type pottery. A great
deal of common check stamp ware was found at this level. Portions
of level 2 are unquestionably of comparatively late post-Columbian
age although the greater part of this level may be early post-Columbian
or late pre-Columbian."
The above suggests a much higher percentage of Safety Harbor
ceramics than is indicated by the surviving collections and a

FLORIDA GEOLOGICAL SURVEY

greater difference between sherds from the two zones than is in-
dicated elsewhere. Ceramically, the plain pottery from the lower
zone would correlate well with the Perico Island period. If such
were the case most of the flexed burials would be of that period.
This would suggest the mound to have been used from shortly after
the time of Christ, about 100 A.D., until about 1600 A.D. As this
length of time is extremely long, the shorter time span is favored
and the lower zone with flexed burials and a few decorated sherds
is assumed to represent a Weeden Island I period.

From the fragmentary data available, we get but little idea of
the life of the inhabitants of the Thomas site except that they col-
lected shellfish and animals for food, made pottery vessels and
stone and bone tools, died, and were buried. For some reason, bird
heads, especially ducks, were prominent during Weeden Island
times as an art form. Duck bills are frequently represented on clay
vessels and stone pendants carved in the form of bird heads are
sometimes found.

The copper and silver ceremonial tablets represent an entirely
different art form, technique, and concept. They must date from
the Safety Harbor period as they are made of post-Columbian ma-
terials. This artistic difference may reflect a different psychologi-
cal orientation during Safety Harbor as opposed to Weeden Island
times, possibly correlating with much greater dependence on agri-
culture.

COCKROACH KEY

Cockroach Key is a small island on the eastern shore of Tampa
Bay about three miles west of Sun City (Fig. 1). Formed entirely
of shells and midden deposits left by Indians, the island is about
600 feet long, 200 feet wide, and reaches an elevation of thirty-five
feet above mean high tide.

One feature of the site is an oval burial mound, about 50 by 70
feet, with a height of 10 feet above surrounding shell deposits. This
mound was tested by Moore (1900, pp. 359-360), who illustrated
his work with an excellent picture that shows the burial mound
cleared for excavation and the higher shell ridge towering behind
it.

W.P.A. Project 690 spent ten weeks excavating the eastern

20

REPORT OF INVESTIGATIONS NO. 8

half of this mound and testing four other parts of the site by
means of extensive exploratory trenches, see Willey (1949, pp. 158-
172) and the reader is referred to Willey's account for a map of the
site, a profile of the burial mound, lists of artifacts, fauna, mollusca,
and other details. The writer's discussion of the site is limited to
the burial mound and is based on the reworked original field notes.
It supplements information already presented by Willey.

Figure 6.-Decorated dagger of deer bone, 8 inches long, from an explora-
tory trench at Cockroach Key. Florida Geological Survey.

As excavated the burial mound was found to consist of three
structures. The base was merely a low elevation, about 21/2 feet in
height, of the horizontally stratified shell midden. Over this rise
an irregularly stratified mound was built of shells, sand, and midden
material, adding about three feet in height. Then a foot or more
of midden accumulated which was capped by a layer of white sand,
several inches thick. The final addition of a tertiary mound, built
of clean shells, brought the burial mound to a total height of
about 10 feet.

Depths are available for all burials and, for those removed
after the multiple structure of the mound was ascertained, a com-
ment was given regarding their relationship to the secondary
mound. Unfortunately, however, elevations of the tops of burial
pit shafts were not noted. Importance of this point is indicated
by seven instances where later burials are recorded as cutting
through earlier interments and three other cases where burials are
mentioned as being in pits.

All burials, except a few omitted because of disturbance or other
reasons, are tabulated in Table 2 against depths below the surface.
These burials have been divided into groups, those from the mound
proper and those from the more or less level area immediately
southeast of the mound (below the 5-foot contour; Willey, 1949,
map 12, area "A"). Included as flexed burials are over fifteen in-
terments of children. For those listed as "Baby or Child" the type
3

21

FLORIDA GEOLOGICAL SURVEY

VERTICAL DISTRIBUTION

TABLE 2.
OF BURIALS AT COCKROACH KEY

Depths in feet
Burial type o-, ) '81 6-7 I 7-8
0-1 1-2 2-3 3-4 4-5 5-6 6-7 7-8
In mound proper
Vertical bundle .. ---2 --.-.- .-. 2 1
Horizontal bundle ---...-- 3 2
Flexed ............ 5 7 17 7 4 1 1
Semi-flexed ....-.....-.....___ __ 2 4 2 1
Isolated skulls .........71....11. 7 1 1 __
Extended-- .....---------------____ 1 1?
Baby or child ............. 7 5 19 32 12 1 2
In level area
Horizontal bundle ..--.... 4 41 1 1
Flexed -3 7 25_~2--_...... 3 7 2 2 2
Semi-flexed ....- __ 2 2 2 1 22
Isolated skulls ..1......... 1 1_ 2 1
Extended -..._ -. ..... -- ...1 ... 1
Baby or child --......-.. 4 1

'Includes three group bundle
6 femurs.

burials; one with 30, one with 10 and one with

'In one case depth due to inclusion of Moore's spoil in vertical measurement.

of burial was not otherwise specified. Similar analyses using only
adult burials or those related to the top of the secondary mound
(when possible) did not produce significantly different results.

As shown in Table 2, the bundle burials in the mound proper,
and to a lesser extent in the level area, were, on the average, shal-
lower in depth than the flexed or semi-flexed interments. This
agrees with burial sequence data presented for the Thomas mound
in Table 1. It is also evident, relatively speaking, that flexed burials
were more numerous at Cockroach Key and bundle burials more
prevalent at Thomas. If this burial sequence has implications of
chronology Cockroach Key should be the older site.

Age groupings of skeletal remains, as entered in the field note-
book, are: Baby to 5 years, 70; child, 25; youth, 10; young adult,
50; middle aged adult, 1; unknown, 28. An excess number of the
very young and a lack of old people is obvious. Willey (1949, p.
163) has suggested an epidemic disease might have caused the high
infant mortality. While the data indicate more male than female
burials the ratio (4 to 3) is closer to what might normally be ex-
pected than at the Thomas mound.

22

REPORT OF INVESTIGATIONS No. 8

Two skulls from Cockroach Key are at the Florida State Mu-
seum in Gainesville (Cat. No. 76792, Acc. No. 3422, transferred
by the Florida Geological Survey, Feb. 24, 1939). Both have shovel-
shaped incisor teeth, deep grooves behind mastoid processes, and
exhibit pronounced prognathism. One is the long-faced skull of a
rugged, middle aged to old man with well-developed but not heavy
supraorbital ridges. This skull is 187 mm. long, 145 mm. wide, and
has a cephallic index of 77.5 per cent. The other is that of a young,
adult female whose third molars are not erupted. It has a length of
167 mm., a width of 132 mm., and a cephallic index of 79 per cent.
With upturned nasal bones, heavy and foreshortened mandibles,
these skulls are remarkably similar to three described by Willey
(1949, p. 117) from the Thomas mound although the two from
Cockroach Key do not have particularly low foreheads, a charac-
teristic of the Thomas mound specimens.
Willey lists 340 plain and 18 St. Johns Check Stamped sherds
from the burial mound at Cockroach Key but had only 93 avail-
able for classification. Of this sample he noted that 39 out of 40
Belle Glade Plain, 18 St. Johns Check Stamped, and 23 that he clas-
sified as Pinellas Plain (a Safety Harbor period type), came from
a depth of 30 to 36 inches and, hence, were included in the fill of
the tertiary mound (Willey, 1949, p. 167).
The inclusion of St. Johns Check Stamped sherds in the fill of
the tertiary mound, by definition, dates its construction as Weeden
Island II (or later). Similarly, the bundle burials from this zone
are also dated as of the Weeden Island II period. This agrees nicely
with data from the Thomas mound where bundle burials were
correlated with the Weeden Island II period.
The only exception to the above at Cockroach Key was Burial
92, found at a depth of 6-7 feet in the mound proper. This burial
was of the vertical bundle type where long bones were used to line
a pit after which other bones and the skull were placed between
the long bones. As a pit burial, it is obviously intrusive but the ele-
vatio.n from which the pit was dug is not known.
For the level area, the notebooks mention check-stamped pot-
tery in only one case, a flexed burial at a depth of 30 inches. This
information agrees with data given above and indicates that the
bundle burials of the level area to have also been Weeden Island
II in date. Quite evidently, however, some of the flexed burials
were also interred as late as Weeden Island II times.

23

FLORIDA GEOLOGICAL SURVEY

The field notes mention red pottery in four cases; twice for
flexed burials in the mound proper at depths of 24 and 40 to 47
inches, and twice for the level area with a horizontal bundle burial
between depths of 12 to 20 inches and with a flexed burial at 54
inches. No red painted sherds were included in the sample classi-
fied by Willey. We do not know if the red pottery was painted red
or colored red by overfiring in an oxydizing flame. There is the
possibility these red sherds may have been Dunns Creek Red which,
below the check stamped sherds, would suggest a Weeden Island
I date for lower zones.

Other sherds mentioned in the field notes were all of a black,
sand-tempered ware. They were recorded for depths of 12 to 47
inches in the mound proper and between depths of six and 60 inches
in the level area. Red ochre is mentioned for Burial No. 120, a
semi-flexed interment which intruded into another burial at a depth
of 96 inches in the level area; and for Burial No. 108, a group of
four skulls, randomly arranged with other bones at a depth of 30
inches in the mound proper.

One comment in the field notes, written March 13, 1936, when
the work was about half finished, should be mentioned. It reads:

"The pottery up to date, with the exception of one piece, is crude
and plain. No soot is found on the outside of any of this pottery but
numbers of stones of roughly % lb. weight have been found showing
indications of having been heated and [they] may have been used for
bringing water to a boil in these vessels."

Another comment not specifically dated is also of interest:

"It seems likely that the fish were scaled before cooking rather than
being cooked or roasted whole, because of the deposits of almost pure
fish scales found through the midden. The long bones of the mammals
and larger birds were all split for extraction of the marrow."

SUMMARY

In the burial mound and adjacent level area at Cockroach Key,
bundle burials and St. Johns Check Stamped, Belle Glade Plain,
and Pinellas-like plain sherds were limited in vertical distribution
to upper zones. Flexed interments and plain, sand-tempered sherds
were present in both the upper and lower zones.

The Pinellas-like plain sherds did not have notched lips or
handles, such as is frequently found on Pinellas or Lake Jackson

24

REPORT OF INVESTIGATIONS No. 8

Plain vessels (Willey, 1949, p. 164). For this reason and because
similar Pinellas-like plain sherds were found in lower zones of the
Prine burial mound at Terra Ceia (Fig. 1) in an unquestionably
Weeden Island context (Bullen, 1951, p. 28), we do not believe
such pottery can be taken as evidence of Indians living at Cock-
roach Key during the Safety Harbor period.

We would conclude Indians started living at Cockroach Key at
least as early as Weeden Island I times, more likely during the pre-
ceding Perico Island period (because of. crude, plain pottery from
lower zones) or during the early centuries of the Christian era.
Habitation continued into the Weeden Island II period about 1100-
1400 A.D., as evidenced by bundle burials and St. Johns Check
Stamped pottery.

The stratigraphy at Cockroach Key is similar to and supports
the interpretation of the stratigraphy at the Thomas burial mound
except that the occupation started earlier at Cockroach Key and
continued later at the Thomas site. Reasons for the abandonment
of Cockroach Key are not evident but it is suggested that increas-
ing interest in agriculture may have been one factor. Farming on
the key would have been impossible while the nearby mainland
consists of mangroves and marsh, that could not be farmed either.

During a long period of about a thousand years, Cockroach Key
was inhabited by a group of Indians who might have been fisher-
men. Their sustenance consisted of fish and shellfish supplemented
by game. Judging from the high rate of infant mortality and from
some pathology among adults, they may have suffered from dietary
deficiencies.

SPENDER MOUND

The Spender mound, located about five miles southeast of
Riverview, is situated between Fishhawk and Bell creeks which
join the Alafia River about two miles north of the site (Fig. 1).
Fornierly, this mound had a cap of shells, approximately two feet
thick, which was removed five years prior to excavation.

Excavations consisted of an east-west trench, 124 feet long,
and a north-south trench, 140 feet long. These trenches, which
intersected in the center of the mound and were laid out along its
major and minor axes, were four feet wide.

25

FLORIDA GEOLOGICAL SURVEY

This work uncovered two sawed cypress boards and two small
fireplaces, one containing charred fragments of a board, a little
south of the center of the mound. These were believed to be as-
sociated with a disturbance or exploratory trench, other evidence
of which was also found, dug by unknown investigators since the
removal of the shell cap in 1931.
The mound consisted of sand containing occasional fragments
of charcoal, some chips of chert, one fragment of worked chert,
and three small, undecorated sherds. Chert chips had random dis-
tribution at various depths greater than two feet. The worked
chert was found at a depth of three and one-half feet which was
also the average depth of the three sherds.

The Spender mound is believed to be of the domiciliary type
which has been investigated by Moore and others with very meager
results. Presumedly, it was built as a platform for a home. The
shell cap, removed prior to excavation, may have been a midden
deposit. If so, pottery and other specimens would have been re-
moved with these shells. The few chips and sherds uncovered were
undoubtedly at or near the original surface of the ground upon
which the mound was built.

The Spender mound was located in an area suitable for agri-
culture. That it once supported an aboriginal farmhouse is a
reasonable hypothesis. Specimens, upon which an estimate of the
date of construction might be based, were, unfortunately, not found.

CAGNINI MOUND

The Cagnini burial mound, located about ten miles north of the
center of the city of Tampa, was situated on a sandy ridge on the
north side of a grassy pond about a half-mile west of Cypress
Creek, which drains into the Hillsborough River (Fig. 1). The
mound, with a height of two and one-half feet and major and
minor axes of about 100 and 80 feet respectively, contained ninety-
four burials as shown on the excavation plan (Fig. 7).
While built of unstratified sand, the Cagnini mound contained
various features. In Sections 116, 227, and 243, three decayed tree
stumps, 12 and 14 inches in diameter, were found at depths of
82 and 72 inches. One stump was that of a pine tree, two were
charred at the top while another had completely decayed above
the water table. Roots of other trees were also found.

26

REPORT OF INVESTIGATIONS NO. 8

Depths recorded for these stumps would place them well below
the elevation of the surface of the sand ridge on which the mound
had been built. The trees may have grown in natural depressions
or the stumps may have continued to burn underground below the
original surface. Simpson (fieldnotes) interpreted these burnt
stumps as evidence of subsurface preparation of the area by Indians
prior to construction of the mound. This would seem to be a rea-
sonable theory.

A charred area or fireplace, in Sections 222, 223, 240, and 241,
covered an irregular area, 48 by 108 inches in extent, at a depth of

1 .... .. .... 90 FT.-
1_-'

- URN BURIAL -HORIZONTAL BUNDLE BURIALS

Q CREMATIONS

-f--ISOLATED SKULLS
CONTOUR INTERVAL 6 INCHES

Figure 7.-Cagnini mound, excavation plan.

27

FLORIDA GEOLOGICAL SURVEY

24 inches. Located at the periphery of the mound, this fireplace
was lower than the surface of the surrounding land. It is described
as; "Fairly heavy deposits of charcoal and ashes in hard packed
layer, approximately 6 inches thick." Another fireplace was found
in Section 148 at a depth of 48 inches or, again, substantially below
apparent mound base. It formed a shallow pit as the deposit con-
sisted of, "Ashes in hard packed layer, 3 feet across, approximately
12 inches in thickness at center."

Two other possible fireplaces were noted. One, in Section 77,
covered an area 12 inches across at a depth of 60 inches and was
below mound base. It was called a deposit of charcoal fragments,
pottery, and fragments of chert, all showing evidence of having
been burned. The other, in Section 138 and also at a depth of 60
inches, is listed as a deposit of charcoal, pottery, bone, and chert
fragments, sandstone, and "ironized" (sic) bone closely packed
in an area 5 by 15 inches in extent.

The elevations of the fireplaces and of the charred stumps were
lower than that of the surrounding land. It is likely Indians pre-
pared a subsurface base for the Cagnini mound, similar to those
mentioned by Moore and one other recently excavated (Bullen,
1949).

It is also possible the mound was built over an older habitation
area. A cache of large fragments of chert in Section 32 at a depth
of 24 inches, parts of pottery vessels in Sections 40, 41, and 147 at
20 inches, in Section 187 at 42 inches, and in Section 208 at 36
inches might be so interpreted. Included in the body of the mound
were; in Section 113, a deposit of deer bones which covered an
area 12 inches across at a depth of 24 inches and; in Section 98, at
a depth of 24 inches, sherds and charcoal plus a pine knot, found
just below the sherds.

All 94 burials at the Cagnini mound were of secondary types.
These included 1 urn burial, 2 cremations, 58 horizontal bundle
burials, and 33 interments of isolated skulls. Skeletal material was
in extremely poor condition but field observations indicated a dis-
tinctly roundheaded group. Interments are listed as one child, three
youths, thirteen male adults, one female adult, seventy-two adults
of unknown sex, and four unknown as to age or sex. One humerus
exhibited a healed fracture. One mandible contained an impacted
third molar.

28

REPORT OF INVESTIGATIONS No. 8

Burials were concentrated in the eastern, southeastern and
southern parts of the mound while interments of isolated skulls
tended to be more centrally located than bundle burials (Fig..7).
The bundle and isolated skull burials were found between depths
of 6 and 42 inches but of the two types, the bundle burials had the
shallower average depth. Forty bundle burials were found above
and eighteen below a depth of 2 feet while only nine isolated skulls
were found above and 24 below that depth.

Both cremations were found at a depth of 30 inches and the
urn burial at a depth of 12 inches. The latter consisted of bones
of a child placed in a large, wide-mouthed, undecorated pottery
vessel of about four gallons capacity. It was a disarticulated or
bundle burial as the skull was found below the long bones.

Artifacts from the mound included five arrow points; four spear
points; seven drills, three turtleback scrapers, six small scrapers;
and 158 chips, all of chert; 32 sandstone abraders; seven stone,
pendants (plummet type; two sandstone, one coral, one quartz,
and three of imported stone) ; an Oliva shell bead; a shell awl (?);
an arrow point made of the tooth of a shark; 213 sherds; and three
pottery vessels. Of these artifacts a stone drill, quartz pendant,
sandstone abrader, and in five instances sherds are mentioned in
the fieldnotes as associated with burials. Other lists add a stone
celt, a triangular arrow point (Safety Harbor type), and a shell
dipper to the inventory.

The fieldnotes record undecorated and somewhat crude pottery.
Color is given as dark brown in one case, light tan in another. One
nearly complete vessel is described as, "almost perfectly round in
shape," and was about 15 inches in diameter. Another is given as
six inches in diameter and two inches deep. In a preliminary un-
published report Simpson' wrote:

Pottery in the Cagnini mound consisted, with only one exception,
of simple spherical pots and plain shallow bowls. One small sherd had
been ornamented on the rim with a series of shallow triangular pits in-
cised [punctated] after the vessel had been fired [dried]. Fugitive red
slip had been used on the spherical pots. This ware was smooth and un-
tempered. The simple bowls were of a peculiar pitted ware of poor
quality later found in the upper levels of both the Jones and the Lykes
mounds. Caches [deposits] of potsherds containing parts of several
vessels were found at various points during excavation."

'Manuscript on file in Archaeological Survey of Florida Park Service office,
Gainesville, Florida.

FLORIDA GEOLOGICAL SURVEY

The rim sherd with the triangular punctations is undoubtedly
representative of Weeden Island Punctated. The red-slipped, un-
tempered vessels are unquestionable Dunns Creek Red. Both are
pottery types of the Weeden Island period. The simple bowls of
pitted ware are Pasco Plain, a limestone-tempered ware made over
a long period of time, particularly in the area to the north of Tampa
Bay.

SUMMARY

Indians built the Cagnini burial mound either upon a prepared,
subsurface base or upon a previously occupied ground surface. In
either case, fill of the mound contained pottery, chips of chert, and
other specimens derived from an Indian village area.

All interments were of secondary types, predominantly hori-
zontal bundle burials plus many consisting only of isolated skulls.
As isolated skulls were deeper, on the average, this form of inter-
ment may have been decreasing in favor during the period of use
of the mound. Excavators also found two cremations and a burial
of a child in a pottery vessel.

Specimens from the fill of the mound and usually considered
typical of the Weeden Island period, include one sherd of Weeden
Island Punctated, many sherds of Dunns Creek Red, a stone celt,
a shell dipper, and seven stone pendants (plummet type). The
stone pendants and celts may be either Weeden Island or Safety
Harbor in date. The one narrow triangular arrow point is sug-
gestive of Safety Harbor while burial in a pottery vessel is remi-
niscent of the Fort Walton period of northwest Florida. Crema-
tions also appear to be a late trait as they occur in Parrish Mounds
2 and 3 (Fig. 1) with articles of European manufacture (Willey,
1949, pp. 146-156; particularly p. 153).

Construction and use of the Cagnini burial mound occurred
during the Weeden Island II period, about 1100-1400 A.D. Due to
the lack of flexed burials and to inclusion of cremations and a
burial in a pottery vessel, the mound should probably be considered
late Weeden Island II, possibly 1300-1400 A.D., with use extending
into early Safety Harbor times.

30

REPORT OF INVESTIGATIONS NO. 8

BRANCH MOUND

The Branch mound was located on the east side of Cypress
Creek about six miles northeast of the Cagnini mound and sixteen
miles north of the center of the city of Tampa (Fig. 1). Details
regarding the site, other than contours shown on the excavation
plan (Fig. 8), are not available.

- BURIALS E GLASS BEADS

CONTOUR INTERVAL 6 INCHES

Figure 8.-Branch mound, excavation plan.

31

FLORIDA GEOLOGICAL SURVEY

The mound was circular with a diameter of 50 feet and a maxi-
mum height of two feet but, in general, it formed a more or less
level plateau only 18 inches above the surrounding land. These
dimensions are suggestive of a domiciliary mound or platform for
a house, but as neither postholes nor other features suggested such
occupation, it must be assumed the Branch mound was built for
burial purposes.

Six burials, four of adults and two of youths, were uncovered
between depths of two and twelve inches. These are listed in the
field notes as two horizontal bundle, one isolated skull, one crema-
tion, and two semi-flexed interments. Skeletal material is described
as "very fragmentary" in all cases.

Sketches are available for the isolated skull, one of the bundle,
and one of the semi-flexed burials. The sketch of the latter
shows a skull, shoulder blades, humeri, and thoracic vertebrae
in anatomical order but no lower arm bones, no pelvic
bones, nor any bones of the legs or feet. If a complete body
had been interred some evidence of pelvis and lower extremities
should have been found. While arrangement of the bones present
suggests a semi-flexed burial, incompleteness of the sketch raises
a doubt. The other semi-flexed burial is mentioned as too frag-
mentary for sketch or accurate description.

Horizontal locations of these burials (Fig. 8) suggest a planned
burial pattern or, at least, that knowledge of locations of previous
burials was available when later interments were made.

The sand fill of the mound, evidently derived from a village
area, produced a drill, fourteen whole or fragmentary arrow points,
a sandstone abrader, three scrapers, a piece of burnt chert, at least
seventeen sherds plus portions of a pottery vessel, and five whole or
fragmentary small glass beads. Pottery is not described but both
plain and decorated sherds are listed. One, at least, must pertain to
the Safety Harbor period as it is described as a section of rim
with decoration and a lug handle. Two of the glass beads, one whole
and one broken, are listed as associated with an isolated skull and
with a horizontal bundle burial, respectively. The other glass beads
were found at or near the surface.

SUMMARY

The small Branch mound contained six burials arranged in a

32

REPORT OF INVESTIGATIONS NO. 8

semi-circle. All burials were at shallow depths and, apparently,
interred over a very short period of time. Fill of the mound con-
tained arrow points, pottery and, near the surface, small blue glass
beads. Types of arrow points and pottery are not known except that
one sherd pertains to the Safety Harbor period.

To agree with data from sites discussed earlier, flexed burials
should be approximately Weeden Island I in date. The two semi-
flexed interments listed for the Branch mound were very frag-
mentary and we do feel they may be considered unquestionably as
of that type.

Hence, in spite of these burials, we believe the Branch mound
was built and used for burial purposes near the middle of the
Safety Harbor period, about 1550 A.D. The cremation, decorated
rim sherd with lug handle, and glass beads all point towards such
a date. As glass beads were only found at or near the surface, the
mound may have been built. a short time prior to the introduction
of such beads as trade goods to Indians of Florida.

LYKES MOUND

The Lykes mound, located behind the Lykes Brothers packing
house in the Oak Park section of Tampa, was about a half-mile
northeast of the head of McKay Bay and a mile north of the mouth
of Palm River (Fig. 1). A north-south profile and an excavation
plan showing contours, locations of burials, extent of shell and of
ash areas, and a borrow pit are shown on Figure 9.

The available data permit a reasonably accurate reconstruction
of the building of the Lykes mound. Judging from the profile
(Fig. 9), there was a slight rise of ground at the spot chosen for
the mound (this assumes a surface elevation lower than at present).
Into this rise a shallow pit or fireplace was dug (between Sections
189 and 321). Iirt from this pit was placed towards the north,
forming a low and small pile of debris (between Sections 101 and
211). The pit was then filled with ashes and charcoal (probably
by building a fire in it for some time). This fill (mixed with sand)
extended beyond the limits of the pit to form a mound base cap-
ping both the pit and the dirt thrown out from it. The extent of
this ash and charcoal zone is shown on the excavation plan (Fig.
9).

33

100 FEET

)8 110 1It 154 Ir 196 120 242

* -BUNDLE SURIALS
* -FLEXEOD URIALS
+-ISOLATED SKULLS
M-MULE BONES
CONTOUR IN

II D -EDGE OF SHELL DEPOSIT
.... EDGE OF CHARCOAL & ASH DEPOSIT
F FIREPLACES
H. HORSE 0 N E'S
TERVAL- I FOOT

EXCAVATION PLAN

HORIZONTAL
V CRITICAL

SCALE FEET
SCALE P FEET

PROFILE

Figure 9.-Lykes mound, excavation plan and profile.

to*

:b

REPORT OF INVESTIGATIONS NO. 8

While the pit was, undoubtedly, used as a large fireplace for
the production of ashes and charcoal, supplementary fireplaces
were also used. One in Section 172 was the base of a charred stump.
Another, in Section 259, contained portions of charred logs, four
to six inches in diameter and 18 inches in length. A third, in sec-
tion 192, was three feet in diameter and was filled with ashes and
burnt sand.

These fireplaces were found at depths of 15, 38, and 39 inches,
respectively. Changing these depths below surface to depths below
the top of the mound, by adding to them the indicated contour dif-
ferences, shows that these fireplaces were located approximately 62
to 76 inches below the top of the mound. Ash in the central pit was
found between depths of 64 and 76 inches.

There seems to be no doubt but that these fireplaces and the
central pit functioned as charcoal generators to produce the basal
zone of charcoal and ash impregnated sand. The supplementary
fireplaces were all located near the eastern edge of the charcoal and
ash deposit. Similar fireplaces, damaged when the borrow pit was
dug, may have located towards the west.

After the installation of this charcoal zone, a slight concavity
existed in the surface over the central pit. This declivity was filled
with material taken from a shell midden. Midden material (chiefly
shells of a small clam) was deposited to a thickness -of about 18
inches to form a core or small primary mound. Sand, presumably
from the borrow pit shown just to the southwest of the mound
(Fig. 9), was then used to build the mound proper and to cover
both the prepared base and the deposit of shells and midden ma-
terial.

Building of the mound proper did not occur all at one time.
A flexed burial was found in the shell and midden layer while an-
other was in the top of this layer. They probably represent original
interments over which part of the sand was deposited. In these
cases shallow depressions or basin-shaped pits were scooped out of
the surface of the shell-midden deposit. The remains were laid in
such depressions and covered over with dark sand heaped up like
a small mound. Unopened clam shells were then scattered over the
surface of these small individual mounds.

The other flexed burials were treated similarly but were at a

35

36 FLORIDA GEOLOGICAL SURVEY

Figure 10.-Flexed burial in shell and midden deposit at Lykes mound.

REPORT OF INVESTIGATIONS No. 8

slightly higher elevation, being in the lower part of the overlying
sand. A dark humic-like stain was noted in the soil immediately
surrounding flexed burials. This was caused by decay of animal
matter plus, quite likely, clothing or bindings required to keep
the bodies in a closely flexed position.

Vertical location and this special treatment of flexed burials,
small individual mounds surrounded by clam shells, implies that
at some time in the history of construction of the Lykes mound the
sand mantle was relatively thin. Subsequently, more material was
taken from the borrow pit and the mound was heightened. In this
additional material were found bundle burials. The resultant mound
was about five feet in height with a northwest-southeast major
axis of 110 feet and a northeast-southwest minor axis of 60 feet.

Horizontally, burials were concentrated in the southeastern
quadrant of the Lykes mound. Vertical distribution of burial types
is given in Table 3. These twenty-six interments represent about
thirty-four individuals who were listed in the field notes as two
children, one female youth, three youths of unknown sex, four adult
males, two adult females, and twenty-two adults of unknown sex.
The number of children is smaller, relatively, than at the Cock-
roach Key site.

TABLE 3.
VERTICAL DISTRIBUTION OF BURIALS AT LYKES MOUND
Depths in feet
Burial type 0-1 1-2 2-3 3-4 4-5
Horizontal bundle .......... ...-..........._ 9 3 2 1
Group horizontal bundle -.....-..... 3 1
Isolated skulls .......12--...---.----. 2.___ ____1 2
Tightly flexed _--_-_- ..----_.. 1 1 2

As indicated in Table 3, bundle burials were concentrated in
upper zones and flexed burials in lower zones. The data from the
Lykes mound agree with data from the Thomas mound and Cock-
roach'Key to indicate a period during the life of these mounds when
burials were flexed followed by a period when interments were
predominantly of bundle types.

Sherds are mentioned as associated in the fill of four of the
bundle burials and the shallowest flexed burial, otherwise, grave
goods were not found. Pottery is not described but that in the
4

37

FLORIDA GEOLOGICAL SURVEY

upper or bundle burial zone is mentioned as undecorated and rather
crude. Some sherds of the plain pitted ware (Pasco Plain) such
as that found at the Cagnini mound, were noted. One decorated
sherd is listed but neither the description nor location are given.

Artifacts included five arrow points, two scrapers, three worked
fragments of chert, two sandstone abraders, a piece of a celt, and
a fragmentary shell dipper. Small concentrations or caches (basket
loads?) of closed clam shells and of small quartz pebbles are also
mentioned in the field notes. The former might be offerings of food
but the latter were probably natural inclusions in the fill.

Two other finds should be mentioned. One was the skeleton of
a mule in Section 280 at a depth of 48 inches together with an old
fashioned iron horse or mule shoe. The other, in Section 245 at a
depth of 9 inches, was the fragmentary skeleton of a horse. It may
be presumed both of these buried animals were intrusive and that
their interments occurred a long time after the abandonment of
the mound by Indians. Locations of these finds are given on the
excavation plan (Fig. 9).

SUMMARY

The Lykes burial mound was built of sand over a prepared base,
of charcoal impregnated sand. The association of the base with
large fireplaces probably denotes a purification or sanctification
ceremony. Such prepared bases have been found under several
burial mounds of the Weeden Island period (Moore, 1902, pp. 130-
1; Bullen, 1949).

In agreement with data presented earlier for other sites, flexed
burials at the Lykes mound underlay bundle interments. Unfortu-
nately, we have no information regarding types of either pottery
or projectile points found in the mound fill. By correlation with
other sites, it may be suggested that the Lykes mound was built
in late Weeden Island I times, for flexed burials, and added to and
used, for bundle burials, during the early part of the Weeden Island
II period, about 1100-1300 A.D.

Data from the Lykes mound strongly support that from other
sites in respect to the succession of burial habits while the smaller
number of interments for such a large mound implies it was used
only a relatively short period of time.

38

REPORT OF INVESTIGATIONS NO. 8

SNAVELY MOUNDS

The two Snavely mounds were located, about three-quarters of
a mile apart, on a sand ridge that borders the swamp along the
south side of the Hillsborough River about a mile north-northwest
of Thonotosassa (Fig. 1). Approximate peripheries of the mounds
and of their more or less level tops, as well as extents of excavated
areas and of discolored ash zones are given in Figure 11. Both
mounds were about three feet in height.

Excavation, carried to a depth of a foot below the elevation of
the surrounding land, was more complete in the case of the first
mound, known as Snavely Mound A. Only two features were noted,
a badly decomposed horizontal bundle burial, in Section 71 at a
depth of 18 inches, and a discolored zone composed of ashes and
charcoal. No additional data are available regarding the bundle
burial.

The discolored zone of Mound A, which did not exceed a thick-
ness of three inches, was uncovered a few inches below the surface.
This zone covered about half of the top of the mound (Fig. 11).
Edges are mentioned as having been fairly distinct. This dis-
colored zone, as well as that found in Mound B, presumedly repre-
sented a house floor even though it was not, apparently, associated
with postholes.

A surprisingly large number of chert spalls and chipped tools
were found in Mound A, including 84 arrow points, 11 knives or
spear points, 139 scrapers, 74 chipped blanks, six drills, a broken
celt, three hammerstones, and three abraders. This list undoubted-
ly, contains many duplications as most specimens were listed as
broken. Several of the arrow points were narrow and triangular,
typical of the Safety Harbor period.

Forty-six locations of pottery were recorded in the notebooks
and these represent a much greater number of sherds. Vessel frag-
ments seem to have been small, as if trampled upon. In four cases,
they are mentioned as "black." One heavy or thick sherd was found
at a depth of 48 inches but otherwise the vertical position was not
given for either pottery or stone specimens. Horizontally, the
stone artifacts were two or three times as frequent in the level
portion of the mound as in the sloping sides. The reverse would
seem to have been true of pottery.

39

IO1 FIlt

MOUND A

MOUND B

** HORIZONTAL BUNDLE BURIAL
.... EDIO OF DISOOLOREO ASH AREAS
iti* PERIPINRIKl OF BASIS AND TOPS OF MOUNDS

Figure 11.-Snavely mounds, excavation plans.

1 I_

'I

REPORT OF INVESTIGATIONS No. 8

Snavely Mound B was tested by means of an "L"-shaped trench
and two supplementary tests (Fig. 11). This mound was similar
to Mound A in that it also contained a discolored ash and charcoal
zone at a shallow depth below the surface. While Mound B was
about half as large as Mound A, the discolored zones of both mounds
were nearly the same size (Fig. 11).

Artifacts were not plentiful in the excavated portion of Mound
B but one small and one large arrow point, and a chipped blank
were found.

SUMMARY

Both Snavely mounds appear to be good examples of domiciliary
mounds or low flat-topped mounds built as foundations for abo-
riginal residences. Oval zones, discolored with ashes and charcoal
and found slightly below the surface, presumedly represent house
floors. Unfortunately, postholes were not found so that shapes of
the houses cannot be determined.
Large quantities of chips, broken artifacts, and chipped blanks
suggest Mound A may have been the home of a manufacturer of
chipped tools. This assumption would be supported if such speci-
mens were found near the surface of the mound, a point not men-
tioned in the field notes. A chert quarry, found by Simpson during
the excavations, a short distance from the site was probably the
source of the stone worked at this mound.
In contradistinction to all of the burial mounds, previously
described, these residential mounds had a charcoal and ash zone
near the surface, as opposed to a basal location, the artifacts of
chipped stone were relatively numerous and burials either absent
or extremely rare.
Due to the inclusion in Snavely Mound A of narrow triangular
arrow points, we believe it to have been built during the Safety
Harbor period, about 1400-1550 A.D. Description of pottery as
"black," as well as the apparent lack of decoration, are apparently
characteristic of this period, to judge from sherds found at the
Safety Harbor site (Griffin and Bullen, 1950). The presence of a
bundle burial would also agree with this dating.
As the Safety .Harbor period was an.agricultural era and the
Snavely mounds were located near arable land, it is likely they sup-
ported Indian farmhouses.

41

42 FLORIDA GEOLOGICAL SURVEY

"" %
N N o "1 i* I t11111 'I)I11I jh iI" / ^/

S- -
S s d -

-_ N -.

-- 70 FEET ---- > -

N ,. -- "-.
II- / "

5 C ,,

-- ..
S -., ,0 -

-- ^ lll r.-
S::
'00 wo-

HI,.,NOICITES CANE I, N SLOPE
-
PL A N

Figure --udsk-c mp ndsu
'N
-- -

~~E~ EB =
'I *'11'/IIj( \h\ Itr 5

'~p ~

z r r

"I)," NO O CL
illIDCTSCHNEI LP

Xr

VERTIAL SALE. XAGGRATE

Fiur 1.Joesmon, sktc ma n ufaepoie

REPORT OF INVESTIGATIONS NO. 8

JONES MOUND

The Jones burial mound was located near the east bank of
Pemberton Creek about a mile southeast of Lake Thonotosassa
(Fig. 1). As shown in Figure 12, the site consists of a burial mound
(A), about 70 feet in diameter and three feet in height, situated
between the open ends of a horseshoe-shaped ridge (B, D, C), about
a foot in height. Just outside the ridge, on the north and south
sides opposite the mound, were two borrow pits (F and G), two
and a half and one and a half feet deep respectively. Area E was
level and many sherds were noted on its surface. The whole com-
plex faced east.

Excavation of the burial mound was nearly completed when it
was decided to dig a trench and two small tests in the eastern ends
of the horseshoe-shaped ridge. Data regarding these two tests are
not available other than a list of artifacts uncovered. The list in-
cludes arrow points, other stone tools, and both plain and decorated
pottery and indicates that the ridge, or at least its eastern ends,
was built of sand taken from a habitation area.

For the burial mound we have a cross-section taken near the
center of the mound (Fig. 12, lower). This section shows an ir-
regular humic stratum extending across the mound at about the
elevation of the surface of the surrounding land. The humic zone
was about a foot higher near the center than at the edges of the
mound.

Simpson refers to this structural feature as a, "very dim broken
humic layer." The field notes state, "Mound built up of yellowish-
brown sand. Base level on east side 36" below surface" and "Base
level burned and calcified calcinedd]. Found at depth of 34" in
Section 29" (see excavation plan, Fig. 13, for section locations).
A large fireplace, about six feet in diameter, was found at a depth
of 40 inches below the junctions of Sections 109, 110, 123, and
124. The various depths at which evidence of fire was found,
when' referred to the top of the mound, form a zone five to six
feet below the top.

These data indicate a subsurface prepared base, similar to
those discussed for the Lykes and Cagnini mounds. If this is cor-
rect, the humic zone mentioned earlier must represent either a
period of abandonment or infrequent use, during which humic ma-

43

44

I I I I I I '
237 236 235 234 233 232 251, 1300 229

ao *I 0 *O 1
w a ** *

210 '

196 n
too
[aA

CF
I

*L E

.'~~ A.* I
* I 009
I.,,.,.

0 6

' 0

J _

m- I

8

* *

*
4-I

43

12 II 10
A i ~ i

9 B 7 6 5 4 3 2
__ I i I I ,

- CREMATION
- BUNDLE BURIALS
* FLEXED BURIALS
- ISOLATED SKULLS

B-BURIALS OF INFANTS
I INDETERMINATE BURIALS
F FIREPLACES
... EDGES OF CHARCOAL AREAS

Figure 13.-Jones mound, excavation plan.

FLORIDA GEOLOGICAL SURVEY

W
-70 FEET

238

224

1 2 I

168

154

140

126

112

98

84

70

56

42

28

14
I:

15

I-

ldF;

ix

I I I
228 227 226 225

211

197
UNEXCAVATEO
183

169

155

141

127

++ 113
*99
99

I

REPORT OF INVESTIGATIONS NO. 8

trial accumulated naturally, or a stage in mound construction
during which humic material was deposited on the mound.

Two features similar to house floors with associated fireplaces
but no postholes were found in the southwestern quadrant of the
mound (Fig. 13, F and F indicate the fireplaces). The first or more
westerly floor was six inches thick and composed of black "oily
looking," midden-like soil, hard-packed and mixed with ashes and
charcoal. At its northern margin this floor was 12 inches beneath
the surface. Due to its location near the edge of the mound, its
southern margin was at a shallower depth. The fireplace at the
northern end of the first floor consisted of a deposit of charcoal
and ashes containing burned shell and bone fragments, at places
eight inches thick. The top of the charcoal was, apparently, a few
inches lower than the elevation of the floor.

The second floor was likewise composed of hard-packed, more
or less "sticky," soil mixed with ashes and charcoal. It had an
average thickness of six inches and was found at a depth of about
four inches. Details regarding the fireplace at the southern end of
this floor are not available. As the second floor was found at a
shallower depth, although nearer the center of the mound, it may
be the later of the two floors.

A flexed burial (Fig. 13, No. 56) was inhumed partly in and
partly below the second floor. The shallowest part of this burial
was at a depth of only four inches as was the surface of the floor.
For this reason and because the skull had not become filled with dirt
while the bones were relatively well preserved, it is believed this
burial was the last interment in the Jones mound.

The vertical distribution of 179 burials is given in Table 4.
Most interments were very tightly flexed with arms and legs folded
and bones of both upper and lower portions of limbs parallel to
the vetebral column, knees and elbows close to the chests. Fre-
quently, the vertebral columns are shown in burial sketches as
curved or bent at an angle. In eight cases flexure was so extreme
pelves were lying close to heads and vertebral columns were "U"-
shaped. Only four interments could be classed as semi-flexed.

Flexed burials occurred both singly and in groups. Four bundle
burials formed a group interment. Indeterminate burials in eight
cases consisted of skulls and arm bones, skulls and vertebrae, or

45

46 FLORIDA GEOLOGICAL SURVEY

Figure 14.-Excavations at the Jones mound.

REPORT OF INVESTIGATIONS NO. 8

skulls, vertebrae and arm bones. In a few cases only teeth or only
teeth and other bones were reported. Some of these may represent
bundle burials.

TABLE 4.
VERTICAL DISTRIBUTION OF BURIALS AT JONES MOUND

Burial type Depths in feet
Burial type
0-1 1-2 2-3 3-4 4-5 5-6 6-7
Cremation ..----......--.-.----.---------- 1
Horizontal bundle ---- ..-..- 42
Isolated skulls -__------ ---__. 4 7 1 2
Flexed _.....-_...........-_.- -- 1. 1 16 273 28 34 23 6
Infant or young child .------------ 1 4 5 3
Indeterminate ---... --- ...-.-....____ 64 2 3 1
1In second house floor.
2One, effect of fire around right eye, charcoal in front of face.
3One, at 24 inches, some of long bones burnt.
4One, includes burnt bones.

Examination of Table 4 will disclose that a cremation, other
burials showing evidence of fire, and bundle burials were all found
at shallow depths, not over 24 inches below the surface. Similarly,
isolated skulls and indeterminate burials, which may include some
of the bundle type, concentrated at relatively shallow depths. These
data agree with those from the Thomas, Cockroach, Cagnini
mounds, in that bundle burials and cremations are, on the average,
found at shallower depths than flexed burials, thus implying chang-
ing burial habits during the useful life of the mound.

Use of the mound for burials over a period of time is also in-
dicated by fifteen cases in which one burial intruded into or through
another. Three other instances of disturbances, of an unspecified
type, are mentioned in the field notes.

Skeletal material was in very poor state of preservation. Field
designations of age indicate ten babies, nine children, fourteen
youths, 123 adults, twenty-four old adults, and eight not specified.
This would seem to represent a normal population distribution. As
at the Thomas mound, however, adult males outnumbered females
three to one. In a few cases skulls of babies suggested, by their
location, childbirth may have been a contributing factor to the
death of female adults.

Many of those interred in the Jones mound were well supplied
with burial goods (Figs. 15-20). Data from individual graves have

47

48 FLORIDA GEOLOGICAL SURVEY

been studied to see if any generalizations could be formulated in

correlating specific types of objects with age or sex. Results were

iot very satisfactory.

e i'": 1 o w
M' 11.,

"L': Y .:
'. ,: ; 1. .' ,' .V .

;,. 15S
... l
., .
',;, .I '

... ,. ..
, -;- v"., .

^1
:..

0lalog

II~i
.4C` '.~

Figure 15.-Deer-head and bird-head stone pendants from Jones mound.
Florida Geological Survey; Florida Park Service photographs.

r
ih':
5~. r I,
""

REPORT OF INVESTIGATIONS NO. 8

In 22 burials where red ochre was present, six skeletons were
babies, eight were adults of unknown sex, eight were male adults,
and one was a female adult with unborn child. This slightly favors
a correlation between the presence of red ochre and burials of males
but there appears to be no correlation with age. All of the babies
ahd one old adult were in ochre stained sand or completely sur-
rounded by ochre. In certain cases piles of sand had been heaped
over burials and ochre sprinkled over the piles. In the case of
thirteen adults, ochre covered the skull, was around the skull, or
on and around the skull. Only a small amount of ochre was found
with isolated skulls. From these data, if any conclusions can be
drawn, ochre appears to be associated with the head.

Beads (Fig. 19, e-h) may be correlated with female burials.
In thirteen instances beads were associated with two babies, a child,
an eleven or twelve year old boy, seven female adults, one male
adult, and an indeterminate burial. In all seven cases where loca-
tions of beads are given, they were around the neck or in the
region of the neck, in front and below the jaw.

Twenty burials were supplied with pendants (Figs. 15-18),
made either of stone or of shell. They were found between depths
of 18 and 68 inches with the vast majority between depths of 24
and 40 inches. Due to the importance of these specimens, Table
5 has been prepared to indicate their distribution by burials.

TABLE 5.
STONE PENDANTS AND BURIAL TYPES AT THE JONES MOUND
Number of pen- Duckbill or
Burial type Number of dants except birdhead
burials duckbill type pendants
Male, adult, flexed ................ 9 14 2
Female, adult, flexed .--..----- 4 6 3
Adult, flexed, sex unknown,!Ul.i 2 5
Isolated skull, sex unknown 3 6 1
Indeterminate .--------- 2 14 3

Pendants were supplied only in cases of adult burials. Appar-
ently, they are more apt to be found with males except for duck-
bill or birdhead type pendants which, while equally divided between
the sexes, were relatively more frequent with females. Due to the
method of sexing skeletons in the field, this suggestion is extremely
tentative. Pendants were located at necks or chests and so, pre-
sumedly, were suspended from the neck in life.

49

50 FLORIDA GEOLOGICAL SURVEY

Three stone celts were found, all associated with isolated skulls.

All were between depths of 18 and 28 inches and had the shallowest

distribution of any type of burial offering. In one case, Burial 147,

. . . . ..
, I ,,l .. ... .:

~. ;7-l~c`: '".
r
!
A~ :
"' '" "' i:; r ;
:.I :.~: J ,:
\
,.... `YLLL-~P, ii ; ii ~ I~ ~i; -~I

!j g;
: ?. .- :r,. ~I*~i~~
r ;
O I t -,; i *-": Pi'
)YOk~) ..

Figure 16.-Duckbill type stone pendants from Jones mound. Florida Geo-
logical Survey; Florida Park Service photographs.

'`' '~
'
-?
fli r
~ue,... .c.

, '-

Al,

if
'I

REPORT OF INVESTIGATIONS No. 8

a celt was associated with pendants and other specimens. In the
other two instances, celts were the only burial offering present.
A few of the outstanding burials from the Jones mound are
described below:

Burial 36-Adult, flexed burial, sex unknown, in Section 105 at
depth of 68 inches, with long shell pendant just below jaw and three
plummet-type limestone pendants (Fig. 17, j) under head and
neck.

Burial 63-Flexed burial of child, 3 to 6 years of age, in Section
119 at depth of 72 inches. Pelvic bones surrounded by fresh water
mussel shells.

Burial 82-Aged female adult, burial consisted of skull and humeri
in anatomical order. Double string of 25 graduated shell beads
(Fig. 19, f) around neck and on chest.

Burial 125-Aged female adult, flexed, in Section 162 at depth of
40 inches with 125 small shell beads (Fig. 19, e) in front and below
jaw. Beads, apparently, in a pile.

Burial 130 and 132-Group burial of male adult, female adult, and
two to three-year old child, in Section 174 at depth of 48 inches.
Adults flexed and facing each other, eight inches apart, with child
between at chest of female. Duckbill pendant (Fig. 16, e) under
right side and cone-shaped pendant under center of jaw, two ob-
long stone and shell pendants in vicinity of neck of male. Large
quantities of small and medium-sized shell beads (Fig. 19, e and f)
about child and neck of female. Bird head pendant (similar to
Fig. 15, d) and two shell pendants between skull of child and jaw
of female.

Burial 135-Adult, female, flexed burial in Section 161 at depth
of 40 inches with small round pendant (Fig. 17, c) under jaw.
Burial 136-Adult, female, flexed burial in Section 175 at depth of
36 inches. Duckbill pendant (Fig. 16, a), eroded limestone pendant,
oblong shell pendant and long, awl-like object at neck.
Burial 141-Adult, male, flexed burial in Section 190 at depth of
30 inches. Parts of two or more undecorated clay vessels over
pelvis and along left side. Plummet-type stone pendant (Fig. 17,
d), another of limestone; and three eroded shell pendants in vicinity
of head and shoulders. Two Busycon dippers near by.

51

I...

: --.:.... ..~..
-- ----
'''
i! i:; r''' .:!'r I-1
:, .. ,19r
~e~r~ ..., ..n!-
~~il t~Pk'~s~s~i;9;H~~I~;yj~;l~&6~i; ,Ir
E
e'i~~' 1~9~~08i~aYn~Y1YB~3 ~~1PCG~aUa~r;SRi"~L~~~~
16'
:~ t'
..
-Yh~~~I~fi!
e
LP

.. i'A-

IljI

S I.

..
I ,l.

M

IN g

a
L;r I. II
'I 1'

,

F

.1

:1

]I
:~ t
I-
'
;: ~~: ;b

REPORT OF INVESTIGATIONS No. 8

Burial 146-Adult, female, flexed burial in Section 189 at depth of
30 inches. One duckbill pendant. (Fig. 16, c), two other stone
pendants (Fig. 17, e), two shell pendants, and quantity of shell
beads present.

Burial 147-Adult burial(?) consisting chiefly of teeth in Section
202 at depth of 18 to 24 inches. Associated objects were seven pen-
dants of imported stone (Fig. 15, b and d; Fig. 16, b; Fig. 17, a, f, h,
and k), two coral and two limestone plummet-type pendants, 10
shell objects including pendants (Fig. 18, a-g), and a large stone
celt. The celt, not illustrated, had parallel sides. These specimens
were in a cache or group of caches at or near where the chest of the
burial should have been.

Burial 149-Represented by teeth in Section 189 at depth of 28 to
30 inches. A duckbill pendant (Fig. 16, d) is listed as associated.

Burial 163-Adult, male, flexed burial in Section 217 at depth of
18 inches. First molars on right sides of both upper and lower jaws
worn down to alveolar processes leaving a symmetrical hole when
jaws were closed. Associated objects included a paraquet head
(Fig. 15, c), another stone, and two shell pendants as well as a
miniature, punctated, clay vessel (Fig. 19, a).
Burial 174-Adult, female, flexed burial in Section 230 at depth of
24 inches. A shell pendant, three perforated fossil shark's teeth
(Fig. 19, b-d), five awl-like shell objects, a chert scraper, and a
quantity of medium-sized shell beads (Fig. 19, f) were present. Ac-
cording to the burial sketch, beads were arranged in three or four
strings around the neck and shoulders with the shark's teeth in
the second or third string. Three of the awl-like objects were
over the beads and over the left shoulder (the skeleton lay on its
right side) behind the ear and head.

Pottery and artifacts of stone and of shell, not associated with
burials, were found over most of the area of the mound but were
more frequent in the eastern half while burial offerings were much
more apt to be found towards the west. Depth measurements are
only occasionally given in the field notes for specific objects.

Probably the most interesting specimen is a beautifully carved

Figure 17.-Stone pendants, plumet type, from Jones mound. Florida
Geological Survey; Florida Park Service photographs.
5

53

FLORIDA GEOLOGICAL SURVEY

1II'

18.-Shell pendants

and Cassis

ipi1

I

/
(iw

//
II

)1I

) hl

liI

0 I t
INCHES
Lip, Burial 147, Jones mound.

and polished pendant of imported stone made in the shape of the
head of a deer (Fig. 15, a). It was screened from loose dirt in the
northeastern quarter of Section 189 at a depth of 48 inches. Burial
146 (see above), in the same quarter section at a depth of 30 inches,
was well supplied with pendants. Possibly the deer head pendant
originally belonged to this burial.
In Section 92, not associated directly with a burial but in such
a location it might have been an offering for a child, Burial 20, was
a cache of six Busycon dippers and three strings of shell beads
at a depth of 30 inches. In Section 133, at a depth of 48 inches,
was what appeared to be a cache of about 30 Busycon dippers as-
vessels. The cache was generously sprinkled with red ochre.
sociated with parts of one plain and three check-stamped clay

Figure

54

u(

REPORT OF INVESTIGATIONS No. 8 55

In Section 94, at a depth of 30 inches, was a neat pile or cache
of 18 pieces of chert. Some had been roughly worked into leaf-
shaped blanks. Other specimens found at this site include 170
arrow points, 28 scrapers, 15 drills, two knives, 31 chipped blanks,
eight hammerstones, a smooth stone ball, an abrader, 19 occurrences
of chert chips, 32 of decorated and 11 of plain pottery, and some
fragments of Busycon shells as well as a rare bead or pendant not
associated with a burial. To this list may be added 17 arrow points,
two scrapers, five drills, two blanks, two hammerstones, and plain
and decorated sherds from exploratory trench B (Fig. 12) and six
arrow points and some sherds from tests C and D. As many of the
stone tools were broken, quantities given are probably greater than
the actual number of tools represented.

A small collection of pottery from the Jones mound is at the
Florida State Museum in Gainesville (Cat. Nos. 76639-76657, Ace.
No. 3422, transferred Feb. 24, 1939, by the Florida Geological
Survey; examples are illustrated in Figure 20). This collection
has been classified as follows:

Restored
Safety Harbor Period Sherds Vessels
Pinellas Incised ----...........--.... --.. -- ----..................... 4
Safety Harbor Incised ....---..--------.--,--. 8 1
Pinellas Plain with notched rims ..----------....... 2
Safety Harbor or Weeden Island Period
Sarasota Incised .. ---..--------------------------- 1
Miscellaneous Incised ....-----..--------------- --... -- 1
St. Johns Check Stamped -----------..-----.------ 3 1
St. Johns Plain .-..................---- ..--- ..----...------- 8
Weeden Island Period
Papys Bayou Punctated ....---........--- ...--.----------- 23 1
Weeden Island Punctated on Pasco paste -.. 1
Totals 51 3

Parts of the necks of two waterbottle-shaped vessels are in-
cluded in the above list under Safety Harbor Incised. One exhibits
an incised, outstretched hand upon a punctation filled background,
fingers to the left and thumb towards the bottom (Fig. 20, a).
Pottery from the Jones mound clearly refers to both the Weeden
Island and Safety Harbor periods. Ceramic stratigraphy is sug-
gested by Simpson who wrote in an unpublished preliminary report
that pottery from the older or lower portion of the mound was
Weeden Island in type while the majority from the upper portion
appeared to be "degenerate Weeden Island" (Safety Harbor In-

FLORIDA GEOLOGICAL SURVEY

A B C D

E F G H

INCHES
Figure 19.-Miscellaneous artifacts from Jones mound. (a) miniature
punctated vessel; (b-d) perforated fossil shark teeth; (e-h) shell beads.

cised) -with check-stamped and pitted plain (Pasco Plain) sherds
predominating. It would appear the mound was started in late
Weeden Island I times and was still being used and additions made
to it during the Safety Harbor period.

No information is available in the field notes regarding types
or vertical distribution of arrow points except that some are identi-
fied as narrow triangular arrow points. However, Simpson, in
a preliminary published report (Anonymous, 1939, p. 59), wrote:

"In the eastern half of the mound proper [the upper part] and
decreasing towards the cerfietery level [lower zones] and towards
the western part of the mound, were found many small triangular
'Bird Points'." This distribution agrees with that given for the
pottery and indicates the use and continued construction of the
Jones mound into the Safety Harbor period. Such points were also
found in tests made in the eastern ends of the horseshoe-shaped
embankment (Fig. 12), this portion having apparently been built
during the Safety Harbor period.

56-

REPORT OF INVESTIGATIONS NO. 8

A few objects of metal and of glass were also found in the
Jones mound. These include a broken green glass bead near the
surface in Section 22, a piece of sheet copper in Section 61, a small
thin concave copper object, possibly an ornament, in Section 93, a
piece of thin copper in Section 130, a broken trade pipe in Section
146, a copper bead in Section 160, and small pieces of copper in
Section 204. The small ornament-like copper object from Section
93 was analyzed on February 11, 1938, by Mr. B. J. Owen, then
.Assistant State Chemist, and found to contain a relatively large
amount of nickel thus indicating European origin. Probably, some
of the other copper pieces are also of European or post-Columbian
origin.

The vertical locations of these objects are not available except
for the green glass bead. Due to the rather general horizontal dis-
tribution, it seems likely these specimens indicate that the Jones
mound was used and added to as late as the historic portion of
the Safety Harbor period.

SUMMARY

Excavations at the Jones mound revealed it to be a burial mound
partly surrounded by a horseshoe-shaped embankment. Pottery
found there pertained to both the Weeden Island and Safety Harbor
periods but sherds of the latter period were limited to upper zones.
Similarly, narrow triangular arrow points of the Safety Harbor
type were found in the eastern ends of the embankment and in
the upper zones of the eastern portion of the mound.

The Jones mound was built during Weeden Island times,
probably with a prepared subsurface base, and substantial additions
were made to it during the Safety Harbor period. The horseshoe-
shaped embankment was probably constructed, or at least ex-
tended easterly, during the latter period. Some additions to the
mound were probably made during the historic portion of the
Safety Harbor period.

Burial data; including a group bundle burial, a cremation, a
few burials with burned bones at relatively shallow depths (24
inches or less) and flexed interments continuing downward to a
depth of six feet; show the same sequence of burial habits encount-
ered earlier at other burial mounds. However, to agree with data
from the other mounds there should have been many more bundle

57

it: I

I, II' 't I '

-

;~c~~,
I'~'' rr

~PrC : 1
..
-;I

~ i""i:: I~ii

i
, i-.
)?.f( L. I`I ~
r` ~:r:s

~

) a

i* \;-:':

a

B

r.i...-

; i

F': I

,

i

I

I

i

I

r

I
I

:

;~-`?j/.r
i.

REPORT OF INVESTIGATIONS No. 8

burials in the upper zones. Even if isolated skulls and indeterminate
burials are considered representative of a bundle type of interment,
this discrepancy would seem to remain.

Various possibilities may explain this difficulty. One would be
that the Jones mound predominantly represents a Weeden Island
burial mound which was abandoned and then added to and reused
during the Safety Harbor period. Such a theory would seem to re-
quire the mound to have been used, during the Safety Harbor
period, as a domiciliary and not a burial mound. Concentration of
triangular arrow points in the eastern portion of the mound, where
burials are rare and not close to the surface (Figure 13), would
support this theory. It does not, however, satisfactorily explain
the two floors with associated fireplaces at shallow depths in the
southwestern quadrant.

Another possibility is that Indians who used the Jones mound
for their interments were more conservative in their burial habits
as far as the actual body was concerned, than others in the region.
They may have adopted the carnal houses, triangular points, and
new pottery styles of the Safety Harbor period but continued to
bind their dead. Possibly the population at this site became very
small after the beginning of Safety Harbor times with most of the
people moving elsewhere.

The two floors with associated fireplaces in the southwestern
part of the Jones mound may have represented carnal houses. That
they probably were not habitation floors is suggested by the fact
that domestic artifacts, found in the fill of the mound, were less
frequent near these features than elsewhere.

Examination of the excavation plan (Fig. 13) indicates a pe-
culiar horizontal distribution of burials. The greatest concentra-
tion is north of the first floor. This concentration continues, with
a little less intensity, to the east and curves around the second floor
up to its fireplace. This arrangement might be expected if the
floors were present during times of inhumation.

Exceptions to the above are Burial 56, shown for the area of

Figure 20.-Pottery from Jones Mound: a, Neck of water bottle, showing
human hand, Safety Harbor Incised; b, Pinellas Incised, sub-type B, with
handle; c, Safety Harbor Incised; d, Weeden Island Punctated; e, St. Johns
Check Stamped; f, Safety Harbor Incised. Sherds about N'-size; e, 8 inches
across; f, 4 inches high. Florida State Museum photographs.

59

FLORIDA GEOLOGICAL SURVEY

the second floor, and an indeterminate burial, below and predating
its fireplace. Burial 56 was intrusive into this floor and must post-
date its construction.

Both floors at the Jones mound differ from those found at the
Snavely mounds, which were taken to represent habitation house
floors. The floors at the Snavely mound were three inches thick,
described as discolored ash and charcoal area, about 25 to 35 feet
across, without adjacent fireplaces and located on the flat top of
the mound, and in an area of heavy artifact concentration. Those
at the Jones mound, on the other hand, were six inches thick, com-
posed of black, heavy, and somewhat "sticky" or "oily" dirt, about
15 feet across, associated with adjacent fireplaces, located at the
sloping sides of the mound, and in an area of relatively low artifact
concentration.

These differences plus the horizontal distribution of burials
suggest, but do not prove, that the floors at the Jones mound may
represent carnal houses where bodies may have been kept and
semi-preserved by smoke prior to inhumation. Groups of flexed
burials support this possibility. One cremation and three burials
with partially burned bones are also suggestive.

Mention has been made of the extreme flexure of the skeletons
found at the Jones mound. To achieve it, bodies must have been
tightly bound, probably before rigor mortis occurred. If such
tightly bound bodies were not left too long in a carnal house before
interment, they would produce flexed and not bundle burials.

There are several similarities between the Jones and Thomas
burial mounds. Both show use from Weeden Island into the historic
p:)rt of the Safety Harbor period, both have embankments or earth-
works, and at both, duckbill and many plumbob-type pendants were
found. In the upper zones at the Thomas mound, many burials
classed as vertical bundle burials were found, but these were, in
many cases, articulated sitting burials in which the skull had fallen
into the pelvic region. Simpson says "these interments were tightly
flexed and undoubtedly bound before inhumation." If these burials
had been laid horizontally or those at the Jones mound placed ver-
tically, it is doubtful if any difference would have been discerned.

At the Thomas mound duckbill pendants, while present, were
not found in situ. At the Jones mound eleven naturalistically

60

REPORT OF INVESTIGATIONS NO. 8

carved stone pendants were uncovered representing a deer, eight
clucks, a paraquet, and an osprey. Many of these were found at
the necks and chests of flexed skeletons in the western part of the
mound between Sections 175 and 189 and between depths of 30 and
48 inches, associated with the plumbob type of stone and shell
pendants. In Section 217 the paraquet head pendant was found
with Burial 163 at a depth of 18 inches while a sherd of Weeden
Island Punctated was associated with Burial 165 listed for a
depth of 14 to 20 inches in the same section. For this reason and
because narrow triangular arrow points, representative of the
Safety Harbor period, concentrated in the eastern portion of the
mound while pendants came from western sections, it seems fairly
certain these specialized pendants are Weeden Island, probably
very late Weeden Island, in date.

Both the Jones and Thomas mounds seem to have been related
to special groups or relatively rich villages to judge from burial
goods. Perhaps these sites were the homes of regional chiefs or
priests. Perhaps a caste system is indicated by the presence or ab-
sence of rich burial goods.

While the Jones mound was used for burial purposes from about
1100 A.D. to about 1600 A.D., all of the special bird head pendants
were made of the same material with the same workmanship. It
would seem they must represent a relatively short time span in the
total history of the construction and use of the mound. They were
probably made by one person or a small group of persons, possibly
at the Jones site, although the material, being a volcanic rock,
must have been imported from the mountainous area to the north
of Florida.

PICNIC MOUND

The Picnic mound, sometimes called the Thatcher mound, was
located about one-half mile southwest of the town of Picnic on
the south bank of Hurrah Creek which flows into the Alafia River
(Fig. 1).

In the field notes, Simpson emphasizes the amount of disturbance
which had occurred prior to excavation, particularly in an upper
portion or secondary mound. For this reason it is difficult to ac-
curately determine the aboriginal appearance. An excavation plan
and three profiles will be found in Figure 21.

62 FLORIDA GEOLOGICAL SURVEY

Simpson describes a secondary mound as 60 to 70 feet in di-

ameter and four feet in height, completely encircled by a shallow

depression or trench, and superimposed over a broad, low primary

mound (Anonymous (J. Clarence Simpson), 1939, pp. 60-61). To

judge from the surface profiles (Fig. 21), the lower primary mound

extended at least 120 feet north and south (A to B) andsupported

a ridge (a) along its southern border. This ridge, as well as the

lateral extension of the primary mound to the north, gave the ap-

pearance of a ditch or borrow pit, as mentioned by Simpson, to

I I

015 014 BIg I14

atll
III

141

010 01 IIs *I IIe

I
3, 4, 5

B

100 rFlT

6 10 II II IS f41I I314 1 1 1

.0
A [

4 +
I OseB1
+

ALARIA OF
L ,
WHITE S A NO

.L L

$-9
AA

AL

1 +1 LILL9

IIl 13 4 114 Il6 I g ll?1 it l li IS o 1I1 01 1 1i 1I4 195 16 2 7 1M90 9 L

PLAN

K -
---!SJA "n

PROFILES

* B)UNDLE BURIALS

A.PoLIXIDO URIIALS .*INFANT BURIAL
ISOLATEATI SKULLS I INDITIRMIHATE IUAIAL
I.FrMUR ANO IAR SPOOLI AT oIPTH OF flit

Figure 21.-Picnic mound, excavation plan and profiles.

1o
so

,to

140

IGO

t40

a0o

100

I*,@

'I

r-

w i i I

.L

REPORT OF INVESTIGATIONS NO. 8

the area immediately surrounding the upper part or secondary
mound. The primary mound reached a height of about three feet
above the general ground level while the ridge extended two feet
and the secondary mound about four feet higher.

It is not certain whether or not the Picnic mound had a pre-
pared subsurface base. "Remains of cypress stumps, roots, and
knees, were abundant at the water level on the northern side of
the mound, in some instances projecting over a foot into the mound
soil itself" (Anonymous, 1989, p. 61). The bases of these stumps
were at water level in yellow sand. Burial 48 in Section 209 at a
depth of 60 inches had a quantity of burned wood near the pelvis.
Burial 52 in Section 207 at a depth of 72 inches was "at contact with
original yellow sand" and "on original surface of base." Fragments
of the femur of this interment were beneath charcoal. There is
less evidence of a prepared sub-mound base at the Picnic mound
than at some of the mounds previously discussed.

Both primary and secondary portions of the Picnic mound
were built of dark, heavy, loamy sand with no demarcation lines
to indicate pauses in construction. There was, however, a zone
of white sand, three inches thick and about 25 feet in diameter,
located near the center of the mound (Fig. 21). The western edge
of this white sand was 30 inches and the eastern edge 30 to 34
inches below the surface while elevation above the base of the
mound is given as 24 inches for the western and 24 to 30 inches
for the eastern edge. As shown in the middle profile (Fig. 21), the
layer of white sand was slightly dome-shaped.

As has been mentioned, the secondary mound had been nearly
entirely dug over- prior to excavation. Originally it contained
burials and many post-Columbian objects. Screening of this dis-
turbed dirt produced fragmentary skeletal material, objects of
European origin, including many glass beads of various types, and
nearly 100 narrow triangular arrow points. Lower zones produced
77 undisturbed burials whose vertical distribution by types is
given in Table 6. Apparently, these interments came from the
lower portion of the secondary mound as well as from the under-
lying primary mound.

In spite of the aforesaid disturbance the vertical distribution
of burial types is similar at the Picnic mound to that at other
mounds in Hillsborough County. Bundle burials were uncovered at

63

64 FLORIDA GEOLOGICAL SURVEY

TABLE 6.
VERTICAL DISTRIBUTION OF BURIALS AT PICNIC MOUND

TypeDepths in feet
T ype ... .
1-2 2-3 3-4 4-5 5-6 6-7
Cremation .... -.--.. ...- 1
In BTsycon dipper .. 1
Vertical bundle ..--. ..... 2..----..._._. 2
Horizontal bundle .......... 7 8 4 1
Isolated skulls ..... .. .. ........... 2 1 10 4 1
Flexed _. .... ..... .. 5 8 6 8 6
Indeterminate ------... -- ..- 11 12
'Child, 2-4 years
'Femur, ear spools

relatively shallow average depths as compared with flexed burials.

Except that field designations of age and sex show many more
adult males than adult females, a normal population is indicated
as follows: one baby, three children, two youths, 24 male adults,
four female adults, 10 adults of unknown sex, 10 old male adults,
four old female adults, and 19 unspecified as to age or sex.

In eleven instances, bodies were accompanied by burial goods,
articles of personal adornment, or specimens accidentally included
in grave fill. Busycon shells or dippers have been mentioned earlier
for other burial mounds but close association between such utensils
and skeletal material was more prominent at Picnic mound than
elsewhere in Hillsborough County.

Burial 27, a group burial at a depth of 40 inches, was supplied
with a ceremonially broken pottery vessel, eight inches in diameter.
Ochre had been spread over the skulls and the vessel. Burial 43, a
flexed interment at a depth of 60 inches, appears to have been sup-
plied with a small pot and a Busycon dipper. A flexed burial,
number 46, at a depth of 46 inches, had sherds of a small decorated
pot in the vicinity of the ribs. The fill of Burial 39, a flexed burial
at a depth of 36 inches, contained the neck of a Safety Harbor In-
cised water bottle.

Burial 25, at a depth of 24 inches, consisted of parts of the
skeletons of a very young adult and an infant with which four
Buisycon shells were associated. Apparently, this was a bundle
burial, and Busycon shells had been placed over each skull and
over some of the other bones. A flexed burial at a depth of 60

REPORT OF INVESTIGATIONS NO. 8

inches, Burial 57, was supplied with a Busycon dipper placed just
behind the skull.

A shell bead and the bones of an infant found in a Busycon
dipper, at a depth of 26 inches, comprised Burial 40. Forty ad-
ditional shell beads are listed in the field notes after the notes on
this burial but association is not stated. Three large shell beads
and the bones of a very young child formed Burial 50, found at a
depth of 50 inches. A large shell bead was also found at lhe neck
of the flexed interment of an old male, Burial 48, at a depth of 36
inches.

The two most interesting interments were Burials 36 and 52.
The latter consisted of fragments of a femur and two copper-
covered cypress ear spools. These specimens are not illustrated but
each consisted of an oval-shaped piece of wood, two and one-eighth
by two and one-half inches, with a convex, copper-covered outside.
The opposite side was flat except for a "lug," about one-half inch
in diameter and one-quarter of an inch thick, located adjacent
to one of the longer sides. This was the deepest burial, found at a
depth of 72 inches lying upon undisturbed yellow sand. Another
ear spool (Fig. 22, c), made of stone and covered on its flat outer
side with copper, is recorded for this site but its original location
is not given in the field notes.

Burial 36, uncovered at a depth of 30 inches is mentioned as
disturbed but appears from the burial sketch to represent a bundle
burial. Objects associated with this burial included a bead of
fossilized manatee bone, one-half inch in diameter; two large blue
.glass beads; remnants of iron and of copper; and two carved bone
ornaments stained with copper salts (Fig. 22, e-f).

Pottery was distributed at random throughout both the primary
and secondary mounds. Unfortunately, sherds from the secondary
mound were not segregated from the others and trends in ceramic
designs can not be determined at the Picnic mound. Most of the
decorated and some of the undecorated pottery is stored at the
Florida State Museum in Gainesville (Cat. Nos. 76658-76735, Acc.
No. 3422, transferred February 24, 1939, by the Florida Geological
Survey) and examples are illustrated in Figures 22 and 23. This
collection has been classified as follows:

65

E F

LJJ

I N OHE S

i ..,.t
i::
'"'
~f-t- ~ I
i -
1 r t.
,. `-::*?`
; i:j;'ii~ ""
;:

REPORT OF INVESTIGATIONS No. 8 67

Restored
Safety Harbor Period Sherds Vessels
Lake Jackson Plain .----------.--------- --------....... 3
Pinellas Incised ------------------...............------------ 17
Pinellas Plain, notched rim --......................... 2
Plain, Pinellas paste ....---.....----. -------------... 7
Safety Harbor Incised ..-...--....-- ------------.... 441 3
Ft. Walton Incised .......---.-.-.-.-----------..... 19
Plain, Ft. Walton paste ..........--...................... 2
Safety Harbor or Weeden Island Period
St. Johns Check Stamped ...-------------..-.......28
St. Johns Plain --.---.------ -----............--........ 4
Little Manatee Zoned Stamped -..._------------ 1
Miscellaneous Stamped -----.....-.....-.............._- 1
Miscellaneous Incised ....--- ..-.....................32
Belle Glade Plain ----......-............................... 42
Smooth Plain ----.........------- -..-.................. 4
Residual Plain ...-------------- ---... ----.----... 15
Weeden Island Period
W eeden Island Incised ...................................... 2
Weeden Island Punctated -.--.---. ----------... 1
Papys Bayou Punctated .......-------..---------- 4
Dunns Creek Red --.----------..------ --------------. 1
Totals 199 4
'Includes ten with interior red paint.
2Includes two with interior red paint.

Some of this pottery is of the Weeden Island period but much of
it is referred to Safety Harbor times. Included among the restored
vessels are a frog effigy container (Fig. 22, j) and a water bottle
decorated with incised hands upon a punctation filled background
(Fig. 22, i). The hand is repeated three times on the body and
twice on the neck. Parts of decorated necks of three other water
bottles are also in the collection (Fig. 23, c, g, 1).

The pencil drawings of the pottery from this site, that accom-
panied the notebooks, include two small representations of human
faces with pierced ears. These specimens are not illustrated in
this report but they are one and one-half to two inches in size and
are made of clay. Both were rim decorations of clay vessels.

Stone tools found at the Picnic mound include ten scrapers,
nine abraders, two spear points, several stemmed points, four

Figure 22.-Miscellaneous artifacts from Picnic Mound: a, Silver disc;
b, decorated silver pendant; c. ear spool of stone (steatite?), front copper
covered; d-f, carved bone ornaments, copper-stained; g-h, stone pendants;
i-j, Safety Harbor Incised vessels. i, 10 inches high; j, 3/2 inches high. a-h,
Florida Geological Survey, Tallahassee; i-j, Florida State Museum, Gainesville.
Florida Park Service photographs.

68 FLORIDA GEOLOGICAL SURVEY

r. '~2~

\"
i
.I
"..i: :I
r
rP1~

q 4A

.... **' *^;^ ^ ^ I
,Il

-, .*.* .. "^ ^
'7
411
r
..t

.,I. '. '', ,
.. ; *. *" *" *:' r. '
,,. .':,

.'./ ..
':~.. I;-.,;. ;"- ','.

: '' '"
~ ~ ~ : ,,,,.,

LA

LL

If

1 '' J' '

ow'
t~
I\Ir P ~ <'

r
.
.... ~
1.2

I

i
i
I

t
r
i

1

:
rr.
'r'
: i.
P~
5~~'r '' r
'
.,@~;'
''' '
;

,

~~ii~5 ;LZt~3
r;~ ~,sn.
?X1
Lc-!,,,-.: ,'Iu' !
...
E
91 .,;.,., i,
u~iisXi
~lt'-~udclr .:r'- ~-~
-I,% "5~YC~";UI' .
.

1F`, ~~'~

REPORT OF INVESTIGATIONS NO. 8

stone pendants, three pieces of galena, two pieces of mica, and over
eighty narrow triangular arrow points. Points were restricted to
higher elevations and, hence, appeared relatively late in the con-
struction of the mound. Simpson (Anonymous, 1939, p. 61) con-
firmed this when he reported that nearly 100 small triangular "Bird
Points" were screened from the disturbed upper secondary mound.
The most unique pendant is illustrated (Fig. 22, h). Others were
of simple plumbob types (Fig. 22, g). The mica came from Sec-
tions 169 and 189 at depths of 50 and 40 inches respectively or be-
low the deposit of white sand. This location should indicate a
relatively early or Weeden Island date of deposition.

Shell artifacts include a large hammer or pick as well as the
beads and Busycon dippers mentioned earlier. One drilled and five
undrilled shark's teeth were also found. Other artifacts of bone
are represented by a bone awl, a pear-shaped ornament, and the
two carved and copper-stained bone objects (Fig. 22, e-f) found
with Burial 36. A similar carved bone specimen, exhibiting part
of a shaft, was found in Section 207 (Fig. 22, d).

A large number of glass beads and post-Columbian metal ob-
jects were found, particularly in the upper or secondary mound,
both during the W.P.A. excavation and previously. Some of these
have been mentioned. An iron celtiform axe came from disturbed
dirt at a depth of 26 inches.

Silver objects included a small claw- or fang-like object, a per-
forated disc (Fig. 22, a), a decorated pendant (Fig. 22, b), an
undescribed pendant, and what appears to have been the cover
of a tobacco pipe. The last object was probably lost by someone
while digging in the mound during the last 100 years.

The small, silver, claw-like object, one and five-eights inches in
overall length, is of considerable interest. It was made in middle
America and, undoubtedly, came from the wreck of a Spanish
vessel. As such wrecks usually occurred in south Florida, com-
munication in some form between that area and Hillsborough
County seems indicated.

Figure 23.-Miscellaneous sherds from Picnic Mound: a-b, Ft. Walton
Incised; c-d, g, 1, Safety Harbor Incised; e-f, Pinellas Incised; h, Pinellas Plain
with notched lip; i, Lake Jackson Plain, sub-type C; j, rim sherd with notched
lip and punctated body; k, Weeden Island Punctated. g, 1, necks of water bottles
(1, double scale). Florida State Museum photographs.
6

69

FLORIDA GEOLOGICAL SURVEY

Horizontally, specimens of metal concentrated southeast of the
center of the mound in a triangular area outlined by Sections 144,
225, and 231. It seems evident they and the glass beads came from
relatively shallow depths and indicate use of the Picnic mound for
burial purposes during the historic part of the Safety Harbor
period.

Sherds, eleven arrow points, a chipped blank, an abrader, a
small hammerstone, and two pieces of iron were found in Trenches
B and C, dug in the area of the ridge at the southern end of the
mound (Fig. 21). The iron came from a depth of 40 inches in
Section C-7 which would suggest the ridge to have been added in
pos t-Columbian times or during the historic part of the Safety
Harbor period.

SUMMARY

The Picnic mound may have been originally built, in part, to
cover the deepest burial, Burial 52, which had copper-covered ear
spools. When completed it was probably a low, dome-shaped affair
and probably also contained Burials 57 (at a depth of 60 inches)
and 64 (found below the deposit of white sand) and perhaps some
ot the other deep flexed interments. This construction occurred
in Weeden Island times and may have commenced as early as 1000
A.D.

Subsequently, a cap of white sand, three inches thick and 25
feet in diameter, was deposited on the dome-shaped central part
(f the mound. Significance of this cap is not evident but knowledge
of its location and the avoidance of its area for inhumation is sug-
gested by the horizontal distribution of other burials as shown on
the excavation plan (Fig. 21).

As time passed, the secondary mound was added and also the
low ridge to the south. These additions may have been made gradu-
ally over a long time or as major projects over a short time. Toward
the completion of the mound bundle burials began to displace flexed
interments as the burial norm. Much of the upper portion was
made during the Safety Harbor times to judge from the pottery
and triangular arrow points while final additions occurred during
the historic portion of that period.

While burials are not shown on the excavation plan (Fig. 21)

70

REPORT OF INVESTIGATIONS NO. 8

for the part of the mound built over the deposit of sand, this
does not imply knowledge of and avoidance of this area throughout
the useful life of the mound. It will be remembered that most of
the upper portion of the secondary mound had been disturbed prior
to excavation and that burials had been found. If their locations
were known and plotted, it is likely there would have been some
over the layer of the sand.

Presumedly, such burials would have been either bundle inter-
ments or cremations, as these types have been found at other
mounds with historic goods. In support of this assumption it may
be pointed out that the only burial with historic goods found dur-
ing excavation of the Picnic mound, Burial 36, was apparently of
the bundle type.

There are obvious similarities between the Thomas, Jones, and
Picnic mounds in succession of burial forms, types of pottery, and
presence of some form of earthworks. As the ceramic stratigraphy
in these mounds has not been well-preserved, it may be worth
noting that the one with the highest percentage of Safety Harbor
pottery types, the Picnic mound, is also the one with the greatest
amount of historic material.

SELLNER SHELL MIDDENS

Extending for about a half mile along the south bank of the
mouth of the Little Manatee River, opposite the Thomas mound
and about two miles north of Sun City, are, or were, large deposits
of shells. Such shell heaps were lived on by Indians while being
formed by accumulating refuse discarded by the inhabitants.

Excavations were made in three places; the eastern part of the
main shell heap, owned by Robert Sellner; the next lot to the west,
part of the same deposit but owned by a Mr. Smith; and smaller
shell deposits about one-quarter mile to the east, owned by Henry
Sellner. It is not clear from the field notes how much was exca-
vated'at Robert Sellner's but it has been estimated as an area about
40 by 70 feet. At Mr. Smith's plot, 1875 square feet were dug but
the digging was substantially less on Henry Sellner's lot. Work at
Robert Sellner's and Smith's will be considered together as the
same strata were found in both cases.

The main deposit varied from seven to nine feet in thickness,

71

FLORIDA GEOLOGICAL SURVEY

depending on surface elevation and slope of the underlying ground
and consisted of shells, discarded from innumerable meals, with
which were mixed bones of food animals, charcoal, ashes, pottery,
and artifacts made of shell, bone, and stone. The charcoal, ashes,
food bones, sherds, and artifacts were concentrated in "midden
layers" and were separated by zones of relatively clean shells.

Four layers of concentrated midden material were uncovered at
depths of one and one-half, three and one-half, seven and one-half,
and nine feet, respectively. The highest midden layer was six to
eight inches in thickness. The lowest covered an area some 60 feet
across while the highest extended 130 feet. Intermediate midden
layers had widths between these extremes indicating growth of
th, shell heap laterally as well as vertically with time.

At various places these midden strata thickened and dipped
downward to form fireplaces filled, primarily, with charcoal, ashes,
and burned shells. Three such fireplaces, two to two and one-half
feet in diameter, were found in the highest midden layer at Robert
Sellner's and four, varying from four to six and one-half feet in
d(i:meter, in various strata at Smith's. The largest, at the base of
Ihe heap at Smith's, contained ashes and had a vertical dimension
of two feet.

Oyster shells predominated in the heap but other shells, native
of the region, were also present. At a depth of four feet at Robert
Sellner's, oysters, coquina, scallop, and clam shells were noted.
Other types of shells are listed later as used in the manufacture of
tools. For the deepest midden strata, oyster shells, fish bones, sting
ray spines, and fragments of turtle carapaces are mentioned.

The base of the shell heap at both Robert Sellner's and Smith's
rested on gray beach sand but in addition two mangrove stumps
were found at the bottom near the center of the deposit at Smith's.

One burial, the flexed interment of an adult male, was found
near the surface on the north side of the shell heap at Smith's. It
wa. unaccompanied by any burial objects.

Specimens recovered from the mound included a copper penny,
a copper punch, modern iron and brass, lead weights, a long glass
head, and bones of a pig and of a horse or of a cow. Most of these
items were found near the surface and may be considered recent

72

REPORT OF INVESTIGATIONS No. 8

with the possible exception of the horse and pig bones. These bones
were split, as if for marrow extraction, and found apparently, as
deep as one to two feet. However, they appear to have been limited
to two locations near the northeastern border of the heap so that
they are possibly intrusive.

Field notes indicate vertical distribution of certain aboriginal
artifacts, in some cases by depth measurements, in others by in-
clusion in the upper three feet or in the lower part of the shell
heap. Thus some of the specimens may be divided into an upper
or relatively late and a lower or relatively early period. Unfor-
tunately, none of the objects themselves are available.

Check-stamped and other decorated sherds were limited to up-
per zones. St. Johns Plain, present in both zones, appears to have
been more common in the upper zone. Other plain sherds were
about equally divided between the two zones while those from
lower levels at Smith's are referred to as "low grade pottery."

The most common artifacts-small shell hammers, bone awls,
detached columellae, abraders of sandstone, Venus shell anvils,
columella gouges or chisels, and Oliva shell beads, in descending
order of frequency-were typical of both zones although many
more of the hammers came from the upper zone and more of the
columella chisels from the lower zone. A Busycon dipper and a
bone artifact with barbed sides were found in the upper zone.
Interestingly, a lump of red ochre, a ground deer jaw, two bear teeth
pendants (drilled?), a long shell bead, a fragment of a celt, and
a double knobbed bone artifact came from the lower zone. Other
specimens include two arrow points, worked sting ray spines, a
bead of manatee bone, an antler punch, a polished stone, one-half
inch in diameter and one and one-half inches in length, and a
columella pendant, which was uncovered at a depth of 60 inches.

Evidences of food animals, other than those of shell fish, in-
cluded bones of deer, bear, opossum, raccoon, rabbit, bird, turtle,
alligator, and fish, including rays and sharks. Among mammals,
deer bones were the most, and bear bones the least, common. Fish
bones are mentioned as very common, turtle as common, and al-
ligator as fairly common.

At Henry Sellner's, one-quarter mile to the east, six small shell
midden deposits were located. Each presumedly represented in-

73

FLORIDA GEOLOGICAL SURVEY

divi(lual house sites. One was circular, nine feet in diameter, but
most were oval, varying from nine to fifteen feet in length and
from six to eight feet in width. Close to four of these deposits,
on the side opposite the river, were shallow depressions.

Exploratory trenches were dug in two of the shell deposits and
between one of them and one of the shallow depressions. Plain
sherds, detached columellae, etc., were found. These exploratory
trenches were not very productive and little work was done. It
is likely such small deposits of shell and refuse represent the
earliest stage of large shell ridges such as were excavated at Robert
Sellner's and Smith's.

SUMMARY

Excavation at the Sellner shell heap was done by removing the
upper 36 inches over a large area before underlying deposits were
examined. When this work was done, fifteen years ago, pottery
types and their significance as well as chronological implications
of various types of stone, bone, and shell tools were practically
unknown. This work represents the first extensive attempt to
secure chronological data from a Florida shell heap. It is most
unfortunate all the data have not survived for analysis.

Evidence of at least two archaeological periods was secured
at the Sellner shell midden. It is reasonable to believe that the
upper levels, containing check-stamped and other decorated pot-
tery, were deposited during Weeden Island II and, possibly, also
during later times. Deeper deposits with predominantly plain
pottery are referred to the Weeden Island I or Perico Island per-
iods. Undoubtedly, this shell heap started to accumulate during
what we refer to as the Perico Island period, about 100 B.C. to
700 A.D., but probably very late in that period, perhaps about
500 A.D. If it were still inhabited at the close of the Weeden Is-
land II period, for which we have indicated a date of about 1400
A.D., it had a useful life of about 900 years! Such use, however,
may have been intermittent.

Indians lived on the growing Sellner shell heap where they
brought the shell fish which supplied the bulk of their sustenance.
To this home they also brought deer and other products of hunting,
and probably nuts and roots, to supplement their diet. Fishing
and turtle collecting must have been major operations to judge

REPORT OF INVESTIGATIONS No. 8

from the large number of bones of these species reported. Here
they also made their pottery containers and other tools, clothes,
and ornaments. Probably they were protected by some form of
house.

A small group of Indians undoubtedly lived for generations
at the Sellner shell midden. After death they may have been in-
terred in a nearby but now unknown burial mound or, quite likely,
ferried across the Little Manatee River to repose in what we refer
to as the Thomas burial mound.

BUCK ISLAND

Buck Island, a relatively high, sandy island completely sur-
rounded by dense cypress swamp, is located on the south side of
Cypress Creek, a mile west of its confluence with the Hillsborough
River and about nine miles northeast of Tampa. At the time of
excavation the island belonged to the estate of the then late Percy
A. Rockefeller.

Due to the surrounding swamp, it was necessary to construct
an access causeway which consisted of 880 feet of bridge and 526
feet of earth fill. The island had an area of about 20 acres, on
111/2 of which evidence of Indian occupation was found. Explora-
tion consisted of a main excavation, 105 by 160 feet, a smaller ex-
cavation, 20 by 55 feet, and fifteen exploratory trenches.

The island was formed of sand, the upper three to seven feet
of which is described as white in color. At these various depths
representing the base of white sand, limonitic concentrations and,
in places, yellowish, crusty, limonite-cemented sand was found.
Junction between these zones was extremely irregular. The lower
zone is probably a post-occupation phenomenon resulting from de-
position of iron salts by seeping surface water.

Apparently, the surface of the island consisted of various
sandy 'ridges, upon which and in which evidence of Indian occu-
pation was found, and a burial area or mound. The latter is de-
scribed as low, irregular, and previously disturbed to a large ex-
tent. Apparently, the burial area was formed by digging a large
hole and throwing sand outward for burials around the margin.

Twenty-eight burials, some multiple, were found near the center

75

FLORIDA GEOLOGICAL SURVEY

IIIl^K^^^^H

'N",

D

IWI

p. 0O

L~J
'Now',

* i ~'4*.

all.;

Bi

t
i :.i.
.t;l~:
~'

b:' r
j

i
rir
~7., ,.~1-~11
..... .~. i.
iiCE .
-,t.-. ;
;tii- .Ir:
d ;*tr~~":, ~it,' ,
~~.~; ` C
i
i
c''
~. f i'

Figure 24.-Miscellaneous artifacts from Buck Island: a, Gold disc with
central gold button; b, duck head vessel adrona, Weeden Island Zoned Red;

76

,4 :; ~ .~

1~

"~'', t
'I

I

: 1

~~

REPORT OF INVESTIGATIONS NO. 8

of the main excavation. Of these twenty-one were concentrated
in an area 30 feet across, five were 15 to 25 feet further to the
north, and two were 30 and 50 feet further to the west. While
some disturbance had occurred, all burials seem to have been
secondary interments of the bundle type although a few may have
been those of isolated skulls.

Skeletal material was badly decomposed and damaged by tree
roots so that no information is available regarding age or sex.
Artifacts associated with burials, placed there either intentionally
or accidentally, consisted of three stone beads, one inch in length,
a conch shell, sherds, and two pieces of gold.

Burials were recorded between depths of one and three feet.
Burial 19, at a depth of 24 inches, comprised six skulls and scat-
tered bones. Under one of the skulls was a decorated gold disc. In
an adjacent section, also at a depth of 24 inches, was a much smaller
undecorated gold disc. These discs have been reassembled to form
one ornament as shown in the illustration (Fig. 24, a).

Plain and decorated pottery, narrow triangular (Safety Har-
bor type) and stemmed arrow points, and a stone celt were also
found in the burial area. The celt was five and three-quarters inches
long, two and three-quarters inches wide at the edge and one and
one-quarter inches wide at the poll end.

Data from the site as a whole suggest the presence of a pre-
ceramic occupation. Near the burial area, pottery and other arti-
facts were found to lie in a zone, eight to twelve inches thick, about
a foot below the surface. In another location, sherds occurred
between the surface and a depth of 18 inches. Chips and worked
chert, on the other hand, were recorded for depths of five feet and
mentioned several times as occurring at the base of white sand and
in the upper part of the limonitic zone. Individual spear points were
noted at depths of 24 and 38 inches and a cache of eight of them
at a depth of 36 inches. Those in the cache averaged five and one-
half inches in length and one and three-quarters inches in width.
A charcoal deposit was found at 27 inches and a deer jaw at 36

c, Pinellas Incised; d, small vessel, Safety Harbor Incised; e, Pinellas Plain
with notched rim; f, Papys Bayou Punctated; g-i, Weeden Island Punctated;
j-l, Pinellas Incised vessels, restored. d, 32 inches high, j-l, 10-11 inches in
diameter, a, Florida Geological Survey; balance Florida State Museum
photographs.

77

FLORIDA GEOLOGICAL SURVEY

inches. If sherds were not found below a depth, of 18 inches, the
sum of these data should indicate the presence of a preceramic
zone.

Except as mentioned above, specimens cannot be placed verti-
cally. The field catalogue includes, in addition to items mentioned
earlier, 43 spear points, 350 whole and fragmentary arrow points
as well as 20 specifically mentioned as narrow triangular in shape,
30 scrapers, six hammerstones, a sandstone abrader, four shell
columella, and over 125 fragments of Busycon shells. This ma-
terial and also the pottery came from excavations throughout the
site, not just from the burial area.

Both plain and decorated pottery is mentioned in the field notes
but identification was not attempted. Some of this collection is
now at the Florida State Museum in Gainesville (Cat. Nos. 76736-
76787, Acc. No. 3422, transferred February 24, 1939, by the Flori-
da Geological Survey; examples are illustrated in Figure 24). A
classification of this pottery follows:

Restored
Safety Harbor Period Sherds Vessels
Lake Jackson Plain .............................. ...... 6
Pinellas Incised ------------------------1 3
Pinella Incised ... .. ......... .. ........ ....... 1 3
Pinellas Plain, notched rim ..........-..... .......... 15
Plain, Pinellas paste ........ ................ ..... .. 42
Safety Harbor Incised ............. .................. 7 2
Engraved (?) (see below) ...... .... --....1....... 1
Safety Harbor or Weeden Island Period
Englewood Incised ................ ..... ... ........... 6
Sarasota Incised .---....----........----- ..-- ..-.-- 1
St. Johns Check Stamped .......-..... .. ........... 3
St. Johns Plain ... ........ .......... .. .... 18
Weeden Island Period
Weeden Island Zoned Red, adorno .-----..-......
Weeden Island Punctated --..........--------.......-- --
W eeden Island Plain ..... ...........-..... .........--- 2
Papys Bayou Punctated ..... ..---..-----............ 25
Similar but on Pasco paste -------.............- 2
Tampa Complicated Stamped ........................ 5
St. Petersburg Incised ........................- 3
Plain, sand-tempered ........ .........-....-- .. 6 1
Unique
Boat-shaped with loop handle ......................
Alternate oblique hatching and punctations in
plain area in band below rim, body Wakulla
Check Stamped ....-.................-................ 3

78

1471

Totals

REPORT OF INVESTIGATIONS NO. 8

Under restored vessels, one was listed as engraved. The few
sherds, from which this vessel was formed, exhibit fine, apparently
post-fired incisions. The pattern cannot be determined with cer-
tainty but seems to represent part of a "feathered being."

The Weeden Island Zoned Red adorno clearly represents a duck
(Fig. 24, b). The collection includes pottery types typical of both
Weeden Island and Safety Harbor periods. Unfortunately sherds
cannot be segregated except typologically.

SUMMARY

Excavations at Buck Island disclosed an extensive village area
and a small burial area or mound from which twenty-eight in-
terments, apparently all of the secondary or bundle type, were
removed.

As a gold disc was found under the skull of one of these burials,
and all burials were of the same type, it may be suggested these
interments were made during the Safety Harbor period although
some may date from late Weeden Island II times. Pottery from
the rest of the site refers to both Weeden Island and Safety Harbor
times.

A preceramic occupation has been suggested based on the
field data. If one were present, there must have been an haitus at
the site from preceramic times up to the Weeden Island period,
as ceramic material to fill such a gap was not encountered.

CONCLUSIONS

Available data from eleven excavations conducted in Hills-
borough County in 1935-8 under W.P.A. auspices have been pre-
sented. Sites include a shell midden village, two domicilliary
mounds, and eight burial mounds. With the probable exception of
the lowest levels at Buck Island, all of these remains were those
of Indians who manufactured pottery.

Information from the shell midden village located at Sellner's,
showed Indians had lived there, either continuously or intermit-
tently, for an extremely long time, undoubtedly several hundred
years, while the accumulation of refuse attained a depth of seven
to nine feet. During this time changes occurred in industrial prod-

79

FLORIDA GEOLOGICAL SURVEY

ucts. The most important of these, in terms of local chronology,
was the relatively late appearance of check-stamped and decorated
pottery.

The data are not sufficiently specific to permit reconstruction
of aboriginal life at different time periods. Nevertheless, the large
inventory makes the reconstruction of Indian life on shell middens
possible if questions of chronology are forgotten. This is of con-
siderable interest as living on shell heaps was typical of the Tampa
Bay area for a thousand years or more. Evidence from other large
shell middens of the region agrees with that from Sellner's to
show that such villages had an extremely long life which spanned
two or more archaeological periods.

One of the unresolved problems of Florida archaeology may be
included under the term "domiciliary mounds." These low but
fairly wide "mounds" have been sporadically investigated by
Clarence B. Moore and others with very meager results. The
W.P.A. excavations included two sites, Spender and Snavely, which
represented this type of mound.

Work at the Spender mound did little more than show it to
have been built by man. Removal of a shell cap, about two feet
thick, five years prior to excavation, is the probable reason for
the scarcity of artifacts, or features suggesting houses. Both
Snavely mounds contained ash and charcoal areas, near their sur-
faces, which were suggestive of house floors although no post holes,
for walls, were noted. The horizontal distribution of sherds and
chipped stone at Snavely Mound A supported the theory it was
built as the foundation for a home.

That such mounds were built during the Safety Harbor period,
is indicated by small triangular arrow points, typical of that
period, found at Snavely's. The pottery and a bundle burial would
not argue otherwise. It is rather likely that domiciliary mounds
were not built during earlier periods but this assumption is based
entirely on negative evidence.

Where available, stratigraphic data from the eight burial
mounds agree with the accepted chronology of the Tampa Bay
area. These data, combined with that from Sellner's shell midden,
confirm that the Weeden Island period followed one during which
pottery decoration was rare and was, in turn, followed by the

80

REPORT OF INVESTIGATIONS NO. 8

Safety Harbor period. At the Thomas mound, for example, deco-
rated pottery came from zones higher than those which produced
primarily undecorated sherds while types of the Safety Harbor
period were limited to shallow depths. Similarly, at the Jones
mound, Safety Harbor ceramics overlaid those of the Weeden Island
period. Triangular arrow points, a Safety Harbor type, were rela-
tively high in the Picnic -and Jones mounds. The presence of
European goods at superficial depths in five of these mounds in-
dicates their use into the historic portion of the Safety Harbor
period.

Data also support the theory that some Weeden Island mounds
had been built upon prepared subsurface bases. Special pottery
caches near mound peripheries do not seem to be a Weeden Island
period trait of Hillsborough County. Carved and polished stone
pendants (of duckbill and other special shapes) may be correlated
with the Weeden Island period but those of the plumbob types
were also used in Safety Harbor times.

The late introduction of carnal houses was suggested by data
from the Jones mound. At least what was taken to represent such
structures, was found near the surface of this mound while similar
structural features were not found at intermediate depths in this
or any of the other mounds.

Use of burial mounds over long periods of time is indicated by
cultural stratification. These data include pottery superposition,
relatively shallow depths of triangular arrow points versus stemmed
points, and superposition of burial types. The latter will be dis-
cussed more fully later.

Mound structure also supports such a theory. For Thomas a
primary and a secondary mound have been suggested. At Picnic a
white sand divided two zones of midden sand and at Jones a humus
layer separated sand zones, piled upon a prepared base. At
Lykes the prepared base was covered by midden shell over which
sand had been deposited while at Cockroach white sand separated
a midden zone from clean shells.

The sum total of these data indicates the large number of in-
dividuals found buried in. these and similar mounds is probably
a function of time and not that of a large local population.

FLORIDA GEOLOGICAL SURVEY

Superposition of types of burials at the eight mounds presents
a picture of changing habits in disposal of the dead. These types
include skeletons prone or extended on the back, flexed on the
side, isolated skulls, bundles of more or less disarticulated bones,
and cremations, listed in their apparent chronological order.

Prone interments were found at only two sites, Cockroach Key
and the Thomas mound. At the former site one or two such burials
were found in the mound proper near the bottom of the burial
zone. At Thomas seven prone burials were uncovered during the
second visit, all in the lowest foot of skeletal producing deposit.
At the famous Weeden Island burial mound on the west side of
Tampa Bay (Fig. 1), prone interments were in lower zones into
which flexed burials had intruded (Willey, 1949, p. 108).
These data agree in demonstrating prone interments to be
the oldest type at the mounds under consideration. In none of
these cases do they appear to be burials of especially important
people. In terms of culture periods of the region, they may be
early Weeden Island I in date or, more likely, they represent a
holdover trait from the earlier Perico Island period.
Flexed burials were found in the same levels as prone inter-
ments in the two mounds mentioned above but were also present
at substantially shallower depths. They were also uncovered in the
Picnic, Jones, and Lykes mound. In each of these mounds bundle
interments were also found and all occurred at an average depth
shallower than the flexed burials, although there were substantial
overlaps vertically. This suggests a gradual change in burial modes
with bundle burials replacing flexed inhumations. At the Weeden
Island mound, again, bundle burials overlay flexed ones (Willey,
1949, p. 108).
Of the five mounds, Thomas, Cockroach, Picnic, Jones and
Lykes, with bundle burials higher than flexed burials, Cockroach
Key produced check-stamped pottery-presumedly a Weeden Is-
land II type-in higher zones while Thomas, Picnic, and Jones con-
tained Safety Harbor as well as Weeden Island pottery types. At
Thomas the Safety Harbor pottery was, apparently, limited to
higher zones. At Picnic, narrow triangular arrow points, a Safety
Harbor type, were relatively high. At Jones, both Safety Harbor
sherds and triangular arrow points were relatively high. For the
Lykes mound, there are no data regarding either pottery or arrow
points.

82

REPORT OF INVESTIGATIONS NO. 8

Data from Thomas, Picnic, and Jones suggest a correlation be-
tween flexed burials and Weeden Island pottery and between bun-
dle burials, Safety Harbor sherds, and triangular arrow points.
However, the lack of such arrow points or of definitively Safety
Harbor pottery at Cockroach Key argues differently, as does the
presence of some Weeden Island pottery at Buck Island and at
Cagnini with bundle burials but no flexed interments.

The inference is that the change from flexed to bundle burial
occurred entirely within the Weeden Island period. That this
change might represent a difference between an early and a late
Weeden Island is likely but that it might be correlated with Weeden
Island I and II as now defined is less certain. Presumedly, this
change occurred gradually and some people, like those of the Jones
site, were more conservative than others.

Because of the extreme deterioration of skeletal material fre-
quently mentioned in the field notes, it is possible that some burials
of isolated skulls should have been classified as other forms of
interments. In some cases there is no question but that isolated
skulls were buried. Their vertical distribution seems to span both
the time when flexed burials prevailed and when bundle burials
were the mode.

That bundle burials plus some cremations were the vogue in
Safety Harbor times seems beyond question. Not only is this in-
dicated by data presented here but the burial mound at the Safety
Harbor site as well as Parrish Mound I produced only bundle in-
terments (Willey, 1949, pp. 136 and 143). Both mounds contained
Safety Harbor pottery types exclusively plus historic goods.
The few cremations, while not particularly shallow, came from
relatively high zones at the Thomas, Picnic, Jones, Branch, and
Cagnini mounds. It may be suggested this was a new burial type
being introduced about the time of the abandonment of these
mounds. At Parrish Mound II, 39 out of 41 interments were crema-
tions and the pottery and arrow points were of Safety Harbor types
found associated with European goods (Willey, 1949, pp. 147-152).
Some of the cultural changes mentioned in this report, such
as the introduction of triangular arrow points and the modification
of Weeden Island ceramics to those of the Safety Harbor period,
represent results of influences entering Florida from the north-
west. We consider such influences to have originated with Middle

83

FLORIDA GEOLOGICAL SURVEY

Mississippi cultures of the Mississippi River drainage. Their
dynamic effect in the Tampa Bay area and Hillsborough County
was to change the culture of the inhabitants from what we call
Weeden Island to the Safety Harbor culture found by the first
Spaniards.

Data from W.P.A. excavations in Hillsborough County support
chronologies otherwise established and give us our first good in-
formation on the burial habits of the inhabitants during Weeden
Island and early Safety Harbor times. While the data are tantaliz-
ingly inconclusive, probably none of it would have been secured if
this work had not been done. Certainly, much less of it would
have been preserved and this report would not have been possible,
if it had not been for the personal interest of J. Clarence Simpson.

BIBLIOGRAPHY

Anonymous (J. Clarence Simpson)
1939 Notes on Two Interesting Mounds Excavated in Hillsborough
County: Third Biennial Report, Florida State Board of Conser-
vation, Archaeological Survey, Biennium ending June 30, 1938,
Tallahassee, pp. 56-62.
Bullen, Ripley P.
1949 The Woodward Site: The Florida Anthropologist, Vol. II, Nos.
3-4, Gainesville, pp. 49-64.
1951 The Terra Ceia Site, Manatee County, Florida: Florida Anthro-
pological Society, Publication No. 3, Gainesville.
Goggin, John M.
1950 Florida Archeology-1950: The Florida Anthropologist, Vol. III,
Nos. 1-2, Gainesville, pp. 9-20.
Griffin, John W.
1950 Test Excavations at the Lake Jackson Site: American Antiquity,
Vol. 16, No. 2, Menasha, pp. 99-112.
Griffin, John W. and Bullen, Ripley P.
1950 The Safety Harbor Site, Pinellas County, Florida: Florida An-
thropological Society, Publication No. 2, Gainesville.
Moore, Clarence B.
1900 Certain Antiquities of the Florida West.Coast: Journal of the
Academy of Natural Sciences of Philadelphia, Second Series, Vol.
XI, Pt. 3, pp. 350-394.
1902 Certain Aboriginal Remains of the Northwest Florida Coast:
Journal of the Academy of Natural Sciences of Philadelphia,
Second Series, Vol. XII, Pt. 2, pp. 126-358.
Simpson, J. Clarence
1937 Report on Activities in Hillborough County, Florida State Board
of Conservation-Archaeological Survey, Second Biennial Report,
Biennium ending June 30, 1936, pp. 109-116.
1948 Folsom-Like Points from Florida; The Florida Anthropologist,
Vol. I, Nos. 1-2, Gainesville, pp. 11-14.
Willey, Gordon R.
1948 Culture Sequence for the Manatee Region of West Florida:
American Antiquity, Vol. 13, No. 4, Menasha, pp. 209-218.
1949 Archeology of the Florida Gulf Coast: Smithsonian Miscellaneous
Collections, Vol. 113, Washington.

84

REPORT OF INVESTIGATIONS NO. 8

Under restored vessels, one was listed as engraved. The few
sherds, from which this vessel was formed, exhibit fine, apparently
post-fired incisions. The pattern cannot be determined with cer-
tainty but seems to represent part of a "feathered being."

The Weeden Island Zoned Red adorno clearly represents a duck
(Fig. 24, b). The collection includes pottery types typical of both
Weeden Island and Safety Harbor periods. Unfortunately sherds
cannot be segregated except typologically.

SUMMARY

Excavations at Buck Island disclosed an extensive village area
and a small burial area or mound from which twenty-eight in-
terments, apparently all of the secondary or bundle type, were
removed.

As a gold disc was found under the skull of one of these burials,
and all burials were of the same type, it may be suggested these
interments were made during the Safety Harbor period although
some may date from late Weeden Island II times. Pottery from
the rest of the site refers to both Weeden Island and Safety Harbor
times.

A preceramic occupation has been suggested based on the
field data. If one were present, there must have been an haitus at
the site from preceramic times up to the Weeden Island period,
as ceramic material to fill such a gap was not encountered.

CONCLUSIONS

Available data from eleven excavations conducted in Hills-
borough County in 1935-8 under W.P.A. auspices have been pre-
sented. Sites include a shell midden village, two domicilliary
mounds, and eight burial mounds. With the probable exception of
the lowest levels at Buck Island, all of these remains were those
of Indians who manufactured pottery.

Information from the shell midden village located at Sellner's,
showed Indians had lived there, either continuously or intermit-
tently, for an extremely long time, undoubtedly several hundred
years, while the accumulation of refuse attained a depth of seven
to nine feet. During this time changes occurred in industrial prod-

79

FLRD GEOLOSk ( IC SUfRiW

COPYRIGHT NOTICE
[year of publication as printed] Florida Geological Survey [source text]

The Florida Geological Survey holds all rights to the source text of
this electronic resource on behalf of the State of Florida. The
Florida Geological Survey shall be considered the copyright holder
for the text of this publication.

Under the Statutes of the State of Florida (FS 257.05; 257.105, and
377.075), the Florida Geologic Survey (Tallahassee, FL), publisher of
the Florida Geologic Survey, as a division of state government,
makes its documents public (i.e., published) and extends to the
state's official agencies and libraries, including the University of
Florida's Smathers Libraries, rights of reproduction.

The Florida Geological Survey has made its publications available to
the University of Florida, on behalf of the State University System of
Florida, for the purpose of digitization and Internet distribution.

The Florida Geological Survey reserves all rights to its publications.
All uses, excluding those made under "fair use" provisions of U.S.
copyright legislation (U.S. Code, Title 17, Section 107), are
restricted. Contact the Florida Geological Survey for additional
information and permissions.

FLORIDA GEOLOGICAL SURVEY

Mississippi cultures of the Mississippi River drainage. Their
dynamic effect in the Tampa Bay area and Hillsborough County
was to change the culture of the inhabitants from what we call
Weeden Island to the Safety Harbor culture found by the first
Spaniards.

Data from W.P.A. excavations in Hillsborough County support
chronologies otherwise established and give us our first good in-
formation on the burial habits of the inhabitants during Weeden
Island and early Safety Harbor times. While the data are tantaliz-
ingly inconclusive, probably none of it would have been secured if
this work had not been done. Certainly, much less of it would
have been preserved and this report would not have been possible,
if it had not been for the personal interest of J. Clarence Simpson.

BIBLIOGRAPHY

Anonymous (J. Clarence Simpson)
1939 Notes on Two Interesting Mounds Excavated in Hillsborough
County: Third Biennial Report, Florida State Board of Conser-
vation, Archaeological Survey, Biennium ending June 30, 1938,
Tallahassee, pp. 56-62.
Bullen, Ripley P.
1949 The Woodward Site: The Florida Anthropologist, Vol. II, Nos.
3-4, Gainesville, pp. 49-64.
1951 The Terra Ceia Site, Manatee County, Florida: Florida Anthro-
pological Society, Publication No. 3, Gainesville.
Goggin, John M.
1950 Florida Archeology-1950: The Florida Anthropologist, Vol. III,
Nos. 1-2, Gainesville, pp. 9-20.
Griffin, John W.
1950 Test Excavations at the Lake Jackson Site: American Antiquity,
Vol. 16, No. 2, Menasha, pp. 99-112.
Griffin, John W. and Bullen, Ripley P.
1950 The Safety Harbor Site, Pinellas County, Florida: Florida An-
thropological Society, Publication No. 2, Gainesville.
Moore, Clarence B.
1900 Certain Antiquities of the Florida West.Coast: Journal of the
Academy of Natural Sciences of Philadelphia, Second Series, Vol.
XI, Pt. 3, pp. 350-394.
1902 Certain Aboriginal Remains of the Northwest Florida Coast:
Journal of the Academy of Natural Sciences of Philadelphia,
Second Series, Vol. XII, Pt. 2, pp. 126-358.
Simpson, J. Clarence
1937 Report on Activities in Hillborough County, Florida State Board
of Conservation-Archaeological Survey, Second Biennial Report,
Biennium ending June 30, 1936, pp. 109-116.
1948 Folsom-Like Points from Florida; The Florida Anthropologist,
Vol. I, Nos. 1-2, Gainesville, pp. 11-14.
Willey, Gordon R.
1948 Culture Sequence for the Manatee Region of West Florida:
American Antiquity, Vol. 13, No. 4, Menasha, pp. 209-218.
1949 Archeology of the Florida Gulf Coast: Smithsonian Miscellaneous
Collections, Vol. 113, Washington.

84

	Front Matter
	Title Page
	Florida State Board of Conserv...
	Transmittal letter
	Table of Contents
	List of Illustrations
	J. Clarence Simpson (1910-1952...
	Introduction
	Excavations
	Conclusions
	Bibliography

MILLISECOND	CLASS.METHOD	MESSAGE
0	sobekcm_page_globals.constructor
0	sobekcm_page_globals.constructor	Application State validated or built
0	sobekcm_database.verify_item_lookup_object
0	sobekcm_page_globals.constructor	Navigation Object created from URI query string
0	sobekcm_database.verify_item_lookup_object
0	sobekcm_page_globals.display_item	Retrieving item or group information
0	sobekcm_page_globals.get_entire_collection_hierarchy	Retrieving hierarchy information
0	sobekcm_assistant.get_entire_collection_hierarchy
0	cached_data_manager.retrieve_item_aggregation
0	cached_data_manager.retrieve_item_aggregation	Found item aggregation on local cache
0	item_aggregation_builder.get_item_aggregation	Found 'all' item aggregation in cache
0	system.web.ui.page.page_load (ufdc.page_load)
0	sobekcm_page_globals.constructor.on_page_load
0	html_echo_mainwriter.add_style_references	Adding style references to HTML
0	html_echo_mainwriter.add_text_to_page	Reading the text from the file and echoing back to the output stream
2	html_echo_mainwriter.add_text_to_page	Finished reading and writing the file