Table of Contents
 The Waddells Mill Pond Site
 Additional Information on Poverty...
 The Proton Magnetometer: Its use...
 Ybor City: A Cuban Enclave...
 A Stratified Archaic Site in Lowndes...
 Alabama Pottery Types Montgomery...
 A Preliminary Excavation of th...
 Back Cover

Group Title: Florida anthropologist
Title: The Florida anthropologist
Full Citation
Permanent Link: http://ufdc.ufl.edu/UF00027829/00147
 Material Information
Title: The Florida anthropologist
Abbreviated Title: Fla. anthropol.
Physical Description: v. : ill. ; 24 cm.
Language: English
Creator: Florida Anthropological Society
Conference on Historic Site Archaeology
Publisher: Florida Anthropological Society.
Place of Publication: Gainesville
Frequency: quarterly[]
two no. a year[ former 1948-]
Subject: Indians of North America -- Antiquities -- Periodicals -- Florida   ( lcsh )
Antiquities -- Periodicals -- Florida   ( lcsh )
Genre: periodical   ( marcgt )
Summary: Contains papers of the Annual Conference on Historic Site Archeology.
Dates or Sequential Designation: v. 1- May 1948-
 Record Information
Bibliographic ID: UF00027829
Volume ID: VID00147
Source Institution: University of Florida
Holding Location: Department of Special Collections and Area Studies, George A. Smathers Libraries, University of Florida
Rights Management: All rights reserved by the source institution and holding location.
Resource Identifier: oclc - 01569447
lccn - 56028409
issn - 0015-3893

Table of Contents
    Table of Contents
        Table of Contents
    The Waddells Mill Pond Site
        Page 43
        Page 44
        Page 45
        Page 46
        Page 47
        Page 48
        Page 49
        Page 50
        Page 51
        Page 52
        Page 53
        Page 54
        Page 55
        Page 56
        Page 57
        Page 58
        Page 59
        Page 60
        Page 61
        Page 62
        Page 63
        Page 64
    Additional Information on Poverty Point Baked Clay Objects
        Page 65
        Page 66
        Page 67
        Page 68
        Page 69
        Page 70
        Page 71
        Page 72
        Page 73
        Page 74
        Page 75
        Page 76
    The Proton Magnetometer: Its use in Plotting the Distribution of the Ferrous Components of a Shipwreck Site as an Aid to Archeological Interpretation
        Page 77
        Page 78
        Page 79
        Page 80
        Page 81
        Page 82
        Page 83
        Page 84
    Ybor City: A Cuban Enclave in Tampa
        Page 85
        Page 86
        Page 87
        Page 88
        Page 89
        Page 90
    A Stratified Archaic Site in Lowndes County Alabama
        Page 91
        Page 92
        Page 93
        Page 94
        Page 95
        Page 96
        Page 97
    Alabama Pottery Types Montgomery Museum Collection Report
        Page 98
        Page 99
        Page 100
        Page 101
        Page 102
        Page 103
        Page 104
        Page 105
        Page 106
        Page 107
        Page 108
        Page 109
        Page 110
        Page 111
        Page 112
        Page 113
        Page 114
    A Preliminary Excavation of th eNarvaez Midden, St. Petersburg, Florida
        Page 115
        Page 116
        Page 117
        Page 118
        Page 119
        Page 120
        Page 121
        Page 122
        Page 123
        Page 124
    Back Cover
        Page 125
        Page 126
Full Text


2000 Florida Anthropological Society Inc.

The Florida Anthropological Society Inc. holds
source text of the Florida Anthropologist
considered the copyright holder for the text
these publications.

all rights to the
and shall be
and images of

The Florida Anthropological Society has made this publication
available to the University of Florida, for purposes of
digitization and Internet distribution.

The Florida Anthropological Society reserves all rights to this
publication. All uses, excluding those made under "fair use"
provisions of U.S. Code, Title 17, Section 107 are restricted.

Contact the Florida Anthropological Society for additional
information and permissions.

/ 1





a publication of the florida anthropological society





The Waddells Mill Pond Site
William M. Gardner
Poverty Point Baked Clay Objects
James F. Small
The Proton Magnetometer
Carl J. Clausen
Ybor City: A Cuban Enclave in Tampa
James W. Covington
A Stratified Archaic Site in Lowndes County Alabama
David W. Chase
A Preliminary Excavation of the Narvaez Midden
Frank Bushnell 1:

THE FLORIDA ANTHROPOLOGIST is published quarterly by Th
Florida Anthropological Society during March, June, Septem
ber, and December. Subscription is by membership in the So
city for individuals interested in the aims of the Society
Annual dues are $4.00 (Students $2.00). ENTERED AS SECON

Officers of the Society 1966
President: Roger T. Grange, University of South Florid
1st V. Pres.: J. Floyd Monk, 1960 SW 61st Court, Miam
2nd V. Pres.: James W. Covington, University of Tamp
Treasurer: Charles Arnade, University of South Florid
Sec.: Mrs. Evelyn Kessler, University of South Florid
Editor: Charles H. Fairbanks, University of Florid

Executive Committeemen 1966
William M. Goza, Box 246, Clearwate
James A. Ford, Florida State Museum, Gainesvill
Cliff E. Mattox, P. O. Box 521, Cocoa Beac
Charleston W. Tebeau, University of Miami, Coral Gable
Carl A. Benson 2310 Resthaven Drive, Orland

Resident Agent: Ripley Bullen, Florida State Museum, Gaines


William M. Gardner


The Waddells Mill Pond Site was a stockaded
cave site of the Fort Walton period. The ceramic
analysis indicates as many as 50 percent of the
vessels were decorated. Study of the modes of de-
coration indicate little change, but present quant-
ified data on the mechanisms available. Documen-
tary evidence suggests the site was the missioniz-
ed Chatot Indian village of San Carlos.


This paper describes the results of test excavations
conducted at the Waddells Mill Pond site, Ja 65, during 1959
and 1960, while the writer was a student at the University
of Florida under the late Dr. John M. Goggin.

The importance of the site lies in two areas, the uni-
queness of its environmental setting, and the possibility of
utilizing documentary evidence to suggest the causative fac-
tors behind the choice of this setting, and further to iden-
tify both the site and its occupants with a historically men-
tioned tribe and village.

It is also the largest known inland Fort Walton site
west of the Appalachicola River and the only site in north-
west Florida archeologically associated with a stockade.

Acknowledgements must be made to the owners of the land
on which Ja 65 is located, Marianna Plantation, Inc. Arran-
gement for excavations were made through Opekasit, Inc. of
Albany, Georgia, managers of the plantation. Ken Strickland
of Marianna deserves special mention for his voluntary help
during all excavations.

The Waddells Mill Pond site, Ja 65, is located in Jack-
son County, Florida, in the NW 1/4, SW 1/4, SW 1/4 of sec-
tion 33, range 11W, township 6N, some seven miles northwest
of Marianna.

The site is situated on top of a hill which is composed
of limestone of the Eocene Crystal River formation which un-
derlies most of the northern half of the county (Morse; 1955:
32). From the base of the hill flow a series of springs
which are the source of water for the pond and creek of the
same name. The construction of a dam some two miles down-
stream undoubtedly made the pond somewhat larger than it was
Florida Anthropologist, Vol. XIX, No. 2, June, 1966


''--,. '- :r="--- ::::

--- A




---- CAVE

n prehistoric times. Approximately six miles from its ori-
in, the creek empties into the Chipola River.

Unfortunately the site was never accurately mapped.
he sketch maps accompanying this paper, however, closely ap-
roximate existing conditions. On the maps, the site is di-
vided into sections for convenient reference. Section A is
he pond. Water depth is not more than five feet. Section
is a relatively level area that runs from the water's edge
o the sharp escarpment formed by section D. At the eastern
dge of section B is a lesser ridge, C, which is some eight
eet higher than section B and five feet below section D.

Beneath section C is a small cave, Cave 1, measuring
ome 20 by 30 feet with a sloping ceiling varying in height
from eight feet to less than two inches.

Section H is a level area bordered on three sides by
ridges. Cave 2, which lies beneath the ridge formed by J,
is approximately 45 feet east and west by 70 feet north and
south. The roof of the cave varies in height from 18 to 14
feet. Section J is either an old sinkhole or part of Cave 2
here the roof has fallen through. It is surrounded by
walls over 16 feet high.

Section D is the crest of the hill. The slope is so
gentle it is almost indistinguishable. Section E is a ridge
that forms the limit of the site. The ridge, or embankment,
is artificially constructed and varies from a few inches to
over three feet in height. Section F slopes rather sharply
down at one point meeting G, a sinkhole. The water level in
this sinkhole fluctuates with the water level of the pond.
Beyond this is a small stretch of uncultivated land and then
acres of pasture.

There are numerous crevices against the escarpments and
it is evident that much of the hill has been made into a ser-
ies of caves by water action.

Vegetation around the site includes pine, cedar, syca-
more, various types of oaks, and close to the water, cypress.


A total of 13 five-foot squares were excavated in the
five areas noted in Fig. 2. All material was screened. Ar-
bitrarily established six-inch levels were the units of ver-
tical excavation.

Five test squares were excavated in section K where
bedrock was reached at an average depth of 18 inches, vary-
ing from as little as 12 to as much as 24. The soil was
quite black and highly organic. About four inches above bed-

rock this black soil began merging into a dark red clay tha
became lighter as bedrock was reached. Cultural material
was present in considerable quantities, lessening slightly
as the soil became red.

Two squares were excavated in Cave 2, one of which wa
abandoned at the end of the first level. The other was car
ied to a depth of 36 inches. Evidence of occupation was
found as deep as 32 inches, however after 18 inches there
was a marked decrease in volume. The soil was a grayish
black deposit mixed with lime and ash to a depth of 24 inch-
es. The soil then began to change to a reddish clay which
became more red as the pit went deeper.

Three test squares were excavated in section B, with
each square varying in depth. The northernmost square, ad-
jacent to the escarpment, varied from 26 to 66 inches; the
next square from 26 to 48 inches; while the final square was
excavated to bedrock at a uniform depth of 24 inches.

Only one square was excavated in Cave 1. Instead of be
ing black, the soil was dark brown to bedrock at a depth of
30 inches.

Two five-foot squares were excavated in section D. Bed
rock was reached at a depth of 12 inches. The soil was a
reddish clay with differed only from the clay common to the
surrounding area in being a somewhat darker red.

Cultural material, dominated by ceramics and faunal re-
mains, was abundant in all areas of the site except section
D. The soil in this area was much more hardly packed as op-
posed to the areas below, the size of the individual sherds
was smaller, and artifacts were less abundant. This would
suggest that more people lived on the upper part of the site,
while certain areas of the lower part were used primarily
for the deposition of trash.



With the exception of a few sherds, the ceramic inven-
tory belongs to the types of the Fort Walton period as defin
ed by Willey (1949:452-470). Only five types are represent-
ed, Lake Jackson Plain, Fort Walton Incised, Point Washing-
ton Incised, Pensacola Plain and Pensacola Incised, with the
first two comprising nearly 90 percent of the total. Such
information affords little understanding of the nature of
the ceramics. Accordingly the analysis to be presented be-
low was undertaken. The purpose of this was not only to ga-
in a better understanding of the occupation at Ja 65 but to
present the data in a way that might facilitate comparisons
in future work.

This point becomes more clear in the following compari-
son. Combining the body and rim sherds into the basic cate-
ories of incised and/or punctated, and non-incised-punctat-
od, or plain, gives the following distribution (only the
First four levels are presented).


-6" 802 19.7% 3,279 80.3% 4,081
-12" 610 14.5% 3,597 85.5% 4,207
2-18" 443 18.4% 1,971 81.6% 2,414
8-24" 148 17.7% 629 82.3% 777

The obvious inference drawn from Table 1 is that the
lain category dominates the decorated category over 4:1.
y type, the plain category would, of course, be Lake Jack-
on Plain.

If, however, only rim sherds are put into these two cat-
gories, the result is as shown in Table 2.


0-6" 278 45.0% 340 55.0% 618
6-12" 259 50.9% 250 49.1% 509
12-18" 190 40.6% 278 59.4% 468
18-24" 68 47.9% 74 52.1% 142

The difference between a 4:1 ratio and a more nearly
1:1 ratio is rather startling. This is especially signific-
ant when reports in the literature are used for comparative
purposes. For instance, Sears (1960:77-78) has commented on
the relatively higher frequency of decorated Fort Walton
types in the northern part of the state as opposed to the
Tampa Bay area. This sort of comment is only meaningful if
the statistical distribution can be altered. It is not
enough to lump body and rim sherds together and hope the per-
centage of both from various sites are nearly equal. This
may very well not be the case. The decorative field of Fort
Walton pottery does not always, or even usually, cover the
entire vessel and breakage of a vessel is liable to result
in a number of plain sherds that may either be thrown into
residual plain or Lake Jackson Plain. Further the total num-
ber of rims is much more representative of the total vessel

The pottery was generally well fired and quite hard
with well smoothed surfaces. Certain of the material, an av-
erage of 10 percent through all levels, had a much rougher
surface and grit particles extruded through both exterior

and interior surfaces. A third ware, less than one percent
of the total at any level, constituted still another varia-
tion. The paste of this ware had a high mica content and a
somewhat soapy feel.

These differences may result from different clay sour-
ces, variations in methods of tempering, trade vessels, or
an earlier occupation (particularly the micaceous paste).
Whatever the case the vast majority of material forms a sin-
gle ware and is indicative of people with excellent control
of ceramic making.

Three types of material were used for tempering, grit,
limestone and shell. The distribution is presented in Table
3 (rims only).


0-6" 493 79.7% 118 19.1% 7 1.2% 618
6-12" 444 87.2% 54 10.6% 11 2.2% 509
12-18" 416 88.6% 49 10.5% 3 0.9% 468
18-24" 129 90.8% 8 5.6% 5 3.6% 142
24-30" 87 97.6% 1 1.2% 1 1.2% 89
30-36" 29 100% 0 0 0 0 29
36-42" 44 95.6% 0 0 2 4.4'% 46
42-48" 36 97.3% 0 0 1 2.7% 37
48-54" 20 100% 0 0 0 0 20
54-60" 7 100% 0 0 0 0 7
60-66" 2 100% 0 0 0 0 2
1,707 23 1,967

A rather significant percentage of the sherds are lime-
stone tempered, which, considering the geology of the site,
was probably nothing more than a substitute for other forms
of grit. What is interesting is the increase in percentage
of limestone tempering from bottom to top, indicating an in-
creasing reliance on such a substitute.

With four exceptions, two in the general area, lime-
stone tempered Fort Walton pottery has not been reported
elsewhere. One of the nearby sites was excavated by Bullen
(1949:1-9) who notes that many of the sherds excavated in
the parking area at the Florida Caverns State Park contained
limestone inclusions.

The other site, to be reported on in the near future,
is the Smith site. This site is located some three miles
southeast of Ja 65 and in a remarkably similar environmental
context. Occupation was inside an ancient sinkhole some 90
feet in diameter and surrounded by walls approximately 20
feet high. Analysis of material recovered from four five-

loot squares gave the distribution of tempering in Table 4.

-6" 159 54.4% 113 38.5% 21 7.1% 293
-12" 179 65.3% 77 28.4% 17 6.3% 270

Bullen gives no percentages but it is obvious that all
three sites are closely related. The significantly higher
percentage of limestone tempered sherds at the Smith site
would indicate it represents an occupation postdating the
initial occupation of Ja 65 and probably represents a move-
ent of peoples from the latter to the former site. The si-
ilarity of the site environs also suggests a close relation-

Bullen (1958) also excavated two sites, Jl and J3, in
he Jim Woodruff reservoir area along the Chattahoochee Ri-
er some 30 miles east of Ja 65, and recovered nine lime-
tone tempered sherds. Some sort of contact is indicated,
but how much is questionable. A possible answer is offered
in the section of conclusions in this paper.

Eleven modes of decoration were isolated for study. It
as hoped that a more detailed analysis of these attributes
would yield information relating to changes through time. At
the outset, it should be stated that, except in a general
way, the analysis did not yield this information. There
are several possible reasons for this. Three likely ones
are the wrong attributes were analyzed, there was little ch-
ange through time, or finally, not very much time was involv-
ed. The second of these reasons seems highly unlikely since
other evidence suggests there is change through time in Fort
Walton ceramics (this will be discussed more fully in a pa-
per on which the writer is working at present). The first
is very likely to be part of the answer since such sensitive
indicators as vessel shape and decorative field were not
thoroughly studied. The final stated reason is also quite
probable as the depth of the midden and amount of material
may very well be a factor of dense population rather than a
long occupation.

Even so, there are indications of certain trends that
do bear presentation, but more important is the quantitative
listing of the various mechanisms used by the occupants of
Ja 65 in decorating their pottery. The importance of this
lies in the possibility of future comparison with spatially
distant sites. As will be discussed later, the end occupa-
tion of this site is fairly accurately placed, as is the as-

sociation with a particular historical group. Assuming that
the geographical extend of Fort Walton involves different
tribal groupings (for instance Pensacola vs. Appalachee),
then one should expect differences that will not be evident
through the use of types. At the present time, there are
not enough detailed descriptions available for comparison,
but it is hoped that this analysis will be a beginning.

Eleven decorative modes and one mode for non-decoration
were isolated for study. They are: (on rims only)
1). Rim sherds with no form of decoration. (Plate 1, A-B)
2). Rims containing a row, or rows, of incised lines parall
el to the rim and parallel to each other. The number
varies from as few as one to as many as seven, with the
frequency of occurence decreasing with the number of in
cised lines. (Plates 1, C-D, 2, A-B)
3). Rims with curvilinear incised lines. (Plate 1, E-F)
4). Rims with both incision and punctation. (Plate 1, G-H)
5). Rims with lugs. Two types are definable, ovoid lugs
which approximate half circles pinched from the vessel
wall, and rectangular lugs which have the center depres
sed giving them a somewhat anthropomorphic or zoomor-
phic appearance added to the vessel wall. The two
types occur with equal frequency in the lower levels
but in the upper two levels the ovoid lug decreases
markedly in frequence. (Plate 2, A-B)
6). Rims with nodes. (Plate 2, C-D)
7). Rims with handles. (Plate 2, E-F)
8). Rims with pinching. (Plate 2, G)
9). Rims with circular indentations apparently made with
the tip of a finger, sometimes with a fingernail inci-
sion. (Plate 2, H)
10). Rims with lip notching or scalloping. (Plate 2, I)
11). Rims with strips of clay added to the vessel wall and
projecting out from the vessel on a horizontal plane.
In two cases the strip is notched. (Plate 2, J).
12). Rims with closely spaced notches on the exterior edge
of the lip. (Plates 1, C, F, H and 2, A, E)
The distribution is given in Table 5. The number of
occurrences do not add up to the total rims, since some
occur in combination.
The following tendencies are observable from Table 5.
1). There is an appreciable increase from bottom to top of
the use of parallel line incision and a corresponding
decrease in the absence of decoration and the use of in
cision and punctation.
2). The use of closely spaced notches or ticking on the rim
liD increases slightly from bottom to top.
3). All of the other modes are used sparingly at any given
stage of occupation but are present in all except the
lowest levels.



NODE 1 2 3 4 5 6 7 8 9 10 11 12 TOTAL
0"-6" 191 194 14 70 18 18 3 29 17 4 2 235 618
30.9% 317. 2.1% 11.3% 2.9% 2.9% .47 4.7% 2.7% .6% .3% 38.4%

6"-12" 147 167 13 79 29 26 5 17 3 3 2 176 509
28.8% 32.8% 2.5% 15.5% 5.7% 5.1% .9% 3.3% .6% .6% .4% 34.5%

12"-18" 184 106 16 68 24 15 6 12 2 1 1 133 468
39.3% 22.6% 3.4% 14.5% 5.1% 3.2% 1% 2.5% .4% .2% .2% 28.4%

18"-24" 45 33 2 33 6 2 1 6 0 1 0 49 142
31.6% 23.2% 1.4% 23.9% 4.2% 1.4%. .7% 4.2% 0% .7% 0% 34.5%

24"-30" 35 16 3 10 10 9 0 0 1 0 0 25 89
39.3% 18% 3.4% 11.2% 11.2% 10.1% 0% 0% 1.1% 0% 0% 28.1%

30"-36" 12 5 2 6 1 0 0 0 0 0 0 9 29
41.4% 17.2% 6.9% 20.6% 3.47 0% 0% 0% 0% 0% 0% 31.0%

36"-42" 23 5 0 2 3 2 2 0 0 0 0 8 46
50.0% 10.8% 0% 4.3% 6.5% 4.3% 4.3% 0% 0% 0% 0% 17.4%

42"-48" 17 5 1 3 5 6 0 0 0 0 0 7 37
45.9% 13.5% 2.7% 8.1% 13.5% 16.2% 0% 0% 0% 0% 0% 18.9%

TABLE 5 Continued

NODE 1 2 3 4 5 6 7 8 9 10 11 12 TOTAL
48"-54" 13 3 0 0 4 0 0 0 0 0 0 5 20
65.0% 15.0% 0% 0% 20% 0% 0% 0% 0% 0% 0% 25%

54"-60" 2 3 0 2 1 0 0 0 0 0 0 1 7
28.5% 42.8% 0% 28.5% 14.2% 0% 0% 0% 0% 0% 0% 14.2%

60"-66" 0 1 0 0 1 0 0 0 0 0 0 0 2
0% 50% 0% 0% 50% 0% 0%7 0% 0% 0% 0% 0%

699 538 51 273 102 78 17 64 23 9 5 650 1,967
34.0% 27.3% 2.5% 13.4% 5.1% 3.9% .8% 3.2% 1.1% .4% .2% 33%


A total of 27 non-Fort Walton series
ered. Type names are given, where known,
depth at which they were found.
Swift Creek Complicated Stamped
(early variety)
Carabelle Incised
(Plate 3, C)
Carabelle Punctated
West Fla. Cord Marked(?)
Wakulla Check Stamped ]
Jefferson Complicated Stamped
(Plate 3, B)
Aucilla Incised
Cord Marked
Engraved (Caddo-like)

Engraved (shell tempered)

Polished engraved
Incised (Plate 3, A)

sherds were recov-
in addition to the

1 18-24"
1 24-30"
1 12-18"


1 0-6"
2 24-30"
1 6-12"
2 12-18"
1 12-18"
1 42-48"
1 36-42"
1 18-24"

The Caddo-like sherds have engraved cross-hatched lines
Dn the surface. The polished engraved sherd is highly burn-
ished and has a long folded rim. The shell tempered engraved
sherd from 12-18" has two circles intersected by parallel
lines, while the other has six parallel lines. The incised
sherd is from a shallow bowl with slightly incurving walls.
rhe rim, at one point, projects rather markedly on a horizon-
tal plane. On this projection is a deeply incised "tad pole-
like" element bordered by two smaller elements of similar de-
sign. A row of five lines is incised on the exterior of the
bowl. The paste is highly micaceous.


Three pottery discs were excavated. All came from the
lower levels, two from 12-18", one from 18-24". (Plate 3,D)

Two sherds from the 18-24" level with grooves worn in
the edge evidently had a secondary use as hones.

Six pieces of fired clay contained basket impressions.
N11 appear to be plain plaiting.

Three possible pipe fragments were recovered. One each
from 6-12", 12-18", and 18-24".

Three human effigies were excavated which had been at-
tached to vessels at one time. Only one, from the 0-6" le-
vel, shows any degree of representation. The others, from
5-12" and 24-30" are quite crude and may possibly be re-
jects. (Plate 3, F)

A bird effigy was found at the 0-6" level. It is near-
ly done. Beneath the bird's craw are incised loops. The
rim of the vessel to which it was attached is still intact
in part. The paste contains limestone. (Plate 3, E)

Five of the mushroom-shaped objects which have been
found at other Fort Walton sites were among the artifacts ex-
cavated. These objects have been called pottery trowels,
bottle stoppers, and ear plugs. The presence of an incised-
punctated design on one and a node-like projection on the
face of another fairly well precludes their use as pottery
trowels. These objects could represent transference of a
utilitarian tool, such as a pottery trowel, to something
that became purely ornamental. (Plate 3, G,H,I)


The shell inventory consists of a columella chisel from
the 48-54" level; a complete celt from 6-12" (Plate 4, M),
and part of one from the same level; part of an awl from 6-
12"; half of a bivalve with a hole drilled in the upper por-
tion, presumably for use as a neckpiece, from 12-18" (Plate
4, L); and a knobbed shell earpin from the 6-12" level
(Plate 4, N).


A variety of worked and unworked stone was found at Ja
65. Included in the inventory are projectile points, a uni-
face scraper (Plate 4, G), a fragment of a polished stone
celt, four fragments of unclassifiable worked stone, two
fragments of grinding stones, a limestone discoidal and a
number of battered river pebbles.

The celt fragment, a highly polished and smoothed black
igneous rock, came from the 6-12" level; the scraper from 12-
18"; and the limestone discoidal from 0-6".

Six whole, or nearly whole points were recovered, in
addition to the midsections of three more and the tip of ano-

One of the six complete points from the 18-24" level is
a beveled corner notched point with a concave base similar
to the so-called spinner points. It is more heavily patinat-
ed than the remainder of the chert material. (Plate 4,A).
The other point from this level has a slightly flaring stem
(Plate 4, C).

A basally notched point came from the 12-18" level
(Plate 4, B). Two stemmed points were found in the 6-12" le-
vel (Plate 4, D-E) and another in the 0-6" level (Plate 4,F).


Abundant remains of animals were found throughout all
levels. A thorough analysis of 1,881 of these were made by
Wing (n.d.).

The individual animals found in the midden represent
what is probably an almost complete census of the types of
game animals that were hunted in the area at the time.

Among these are deer, several varieties of turtle, oppo-
sum, rabbit, raccoon, gray fox, black bear, turkey, alligat-
or, gar fish, squirrel, beaver, wood rat, dog, skunk, puma,
crow, frog and sea turtle.

The most abundant remains are of deer and the several
varieties of turtle, The sea turtle and beaver are not natu-
ral habitants of the area and were brought in from other
areas. The beaver was probably from the Chattahoochee-Flint-
Appalachicola River area, and the sea turtle from the Gulf
of Mexico.

Two animal remains deserving special mention are pig
and cow or bison. The pig remains come from the surface
area of Cave 2 and are almost certainly not connected with
the Indian occupation. In the same area are bits of modern
glass ware.

The cow or bison bone is another matter. It came from
the 12-18" level, which is fairly deep within the midden.

Swanton (1946, pp. 324-327) offers some discussion on
bison in the southeast. In 1675 Bishop Calderon included
bison among the animals hunted by the Timucua. In 1739 Ogle-
thorpe found a large herd in central Georgia. Don Carlos de
Siguenza y Gongora mentions bison several times during his
exploration of Pensacola Bay in 1693. Torres de Ayala, dur-
ing an overland expedition from Tallahassee to Pensacola en-
countered numerous bison tracks along the Blackwater River.
This was also in 1693.

Very little was found in the way of vegetal remains.
The material recovered was six carbonized corncobs and seve-
ral fragments of corncobs. Their recovery is probably for-
tuitous. They were found 24 inches down in Section B. This
material has not been observed by qualified botanists and
further identification is not possible at this time.


A single, primary, semi-flexed burial was found in Sec-
tion H. The skeleton was examined by Dr. William C. Massey,

who tentatively identified it as an adolescent female.
There was some indication of occipital flattening. Inter-
ment was in the midden, on the right side with the skull ori
ented towards the west. There was no definite association
artifacts with the burial.


The low, sometimes barely discernible horseshoe-shaped
ridge, G. was first noticed by James A. Ford, who visited
the site during the April, 1960, excavations.

Four attempts to locate postholes were later made. The
first three two-foot wide cuts through the ridge failed to
turn up any. However considerable evidence was found sug-
gesting the artificial nature of the ridge. The dark red
clay composition of the ridge gave way to the lighter red
clay natural to the area slightly below the level of the
ground on either side of the ridge. The top soil north of
the ridge was scraped away. The light red clay was found at
a depth of about an inch. On the south side, the clay stay-
ed dark red until a depth of six inches, or two inches above
where it appeared in the test conducted in section D. The
darkness of the clay on the south side resulted from occupa-
tion and the clay from which the embankment was built appar-
ently came from this side. The lighter colored clay on the
north side was due to a lack of occupation. The inch of
dark clay on the north side was due to humus development and
probably erosional deposition from the south side. Cultural
material was virtually absent on the north side.

The fourth two-foot wide trench, later expanded to four
feet, revealed five postholes. The ridge at this point was
1.5 ft. high and five feet wide. Three postholes appeared
at a depth of .3 ft. disappearing at 1.3 ft. The other two
were found at .45 ft. and disappeared at .95 ft. Distance
between postholes was only .15 ft.


Willey (1949, p. 469) and Smith (1956, p. 123) both
feel that Fort Walton culture was extant at the time of th
Spanish entry into Florida. Numerous Fort Walton period
sites have been found in association with historic trade
material. Smith ibidd) discusses seven of these and Lazarus
(1961, 1961A, 1965) has reported evidence of European con-
tact around the Fort Walton-Pensacola area.

The tribes encountered in the area by the Spanish in-
clude the Appalachee, Appalachicola, Sawokli, Chatot, Pensa-
cola, Tawasa, Pawotki and Yuchi. The Chatot are assigned to
the region most closely corresponding to Ja 65. Swanton (19-
22:134) places them near the mid-course of the Chipola River.

Two missions were apparently established among the Cha-
Lut (ibid:135), San Nicolas and San Carlos. The first refer-
ence to San Nicolas places it nine leagues west of the Appa-
lachicola River, while three leagues beyond this was San
Carlos (Boyd, 1958:258). Another reference (ibid:259) puts
the distances as ten and four leagues respectively.

During the first overland expedition to Pensacola, Aya-
la "pitched camp in a cave, a very pleasant spot called San
Nicolas, where there was a Choctaw (sic) village" (ibid:260).
This is fairly specific documentary evidence associating the
Chatot with the specific ecological setting of Ja 65, the
Smith Site, and the Florida Caverns State Park.

Boyd suggests that Rock Arch cave, which is on Park pro-
perty, is the only cave of sufficient size in the area to
have housed a sizeable group. He feels this is the location
of San Nicolas and San Carlos lay further west. The State
park is located very close to a natural bridge over the Chip-
ola and thus would be ideally situated. The population of
San Nicolas was estimated to have 100 persons. The sites
Bullen found seem to be too small to have had a population
of this size, but undiscovered sites may be in the area.

Waddells Mill Pond lies between six and eight miles
slightly northwest of the state park, which certainly falls
within the three or four leagues mentioned by the Spanish.
The density of occupation at Ja 65 and in the surrounding
fields would fit well with the 400 people estimated by the
Spanish for the population of San Carlos.

It seems reasonably certain that Ja 65 is indeed the
village of San Carlos. Although no historical material was
found, documentary evidence as to location and ecological
setting support this statement.

The two missions among the Chatot were established in
1674 (Boyd:259) and abandoned the same year because of hosti-
lities on the part of the Chisca (Yuchi) who lived to the
west. Shortly after that, between 1675 and 1677, a number
of Chatot moved to the west bank of the Appalachicola River.
Boyd (ibid:258) correlates the relocation of San Carlos with
Bullen's sites 1 and 3 in the Jim Woodruff reservoir. As
mentioned previously, Bullen found nine limestone tempered
sherds. This is indicative of contact with Ja 65, but does
not seem to suggest a relationship strong enough to equate
with relocation.

Regardless, though, this date establishes an end point
for the occupation of San Carlos and by extension, Ja 65.
This date extends Fort Walton period occupation well into
historic times. It is impossible to determine from the evi-
dence available when the occupation began. The depths of
the midden, an average of at least eighteen inches, and down

to 32 in the cave, would suggest an occupation covering the
entire range. These depths though are deceiving, and are
probably indicative of a high density of population rather
than an extended temporal span.

The beginning date of Fort Walton throughout the north
west Florida region has received varied treatment. A number
of writers have suggested the period had barely begun before
historic times. This seems untenable in light of the recent
dating of 1250-1500 A.D. accorded Moundville by McKenzie (19-
65:56), and the Moundville related material from the Gulf
coast particularly at Jolly Bay and Bear Point. Moore (1901
459-465) and Holmes (1898, Plates LIV-LVIII) illustrate mate-
rial from these sites which fits very well into Phase II of
Munson's (1965) five-phase seriation of Cult material. This
would certainly be no later than 1350 A.D. A study of Moore
and Holmes' works (in a paper to be completed soon by the
present writer) indicates there is considerable change in
vessel form and design field in Fort Walton ceramics suggest-
ing a somewhat longer temporal span than is usually attribut-
ed to the period. Other evidence suggests the initial begin-
nings may go even earlier.

The stockade at the site is indicative of somewhat un-
easy times. This may have stemmed from the hostility on the
part of the Chisca mentioned earlier, or it may reflect the
nature of the occupants. According to Swanton (1922:134),
the Chatot were apparently the only tribe in the area main-
taining an existence apart from the Creek Confederacy.
Their warlike nature is attested by a letter from the Gover-
nor of Florida following a peace treaty between the "Chaca-
tos, Applachocolos and Amacanos" in which he says "It is an
extraordinary thing, because the aforementioned Chacatos ne-
ver had peace with anybody" (ibid:135). This lack of peace
with neighboring tribes plus the constant inroads of the
Chisca is probably also the reason they chose to live in and
around caves, which serve as natural defenses.

It is doubted these peoples lived within the stockade
at all times. Living conditions within these confines would
have been quite hard. In addition, extensive Fort Walton
sites are scattered over the fields of the surrounding area.
What is more likely is a withdrawal from the field sites to
the stockaded caves in time of stress, at least for most of
the population. Probably some lived on the hill at all
times. Another indication of stress is the burial. Inter-
ment in a midden is a departure from the usual Fort Walton
burial pattern, and it probably represents a precipitous in-

In summary, it has been established with a high degree
of reliability that the Waddells Mill Pond site was the Cha-
tot village of San Carlos mentioned in Spanish documents.

rhe presence of a stockade and occupation of
shelters more than likely resulted from the
with neighboring tribes. The end date of the
Between 1675 and 1677. A beginning date of
tentatively suggested because of the absence
related material. The few engraved sherds
somewhat earlier.

caves and rock
stress of war
occupation was
after 1500 is
of Moundville
may place this

The peoples subsistence was based on farming, hunting
and shellfish collecting. The dominance of faunal remains
over vegetal remains may be either a fact of preservation or
heavy emphasis on hunting as opposed to agriculture.

The population estimate of 400 persons given by the
Spanish seems reasonable in light of the density of occupat-
ion indicated by the archeology.


Boyd, Mark F.
1958 Historic sites in and around the Jim Woodruff
reservoir area, Florida-Georgia. River Basin
Survey Papers, No. 13, Bureau of American Eth-
nology, Bull. 169, pp. 197-314.

Bullen, Ripley P.
1949 Indian Sites at Florida Caverns State Park,
Florida Anthropologist, Vol. II, pp. 1-9.

1958 Six sites near the Chattahoochee River
Jim Woodruff Reservoir area, Florida.
Basin Survey Papers, No. 14, Bureau of
can Ethnology, Bull. 169, pp. 317-78.

in the

Holmes, William H.
1898 Aboriginal pottery of the eastern United
States. Bureau of American Ethnology, Annual
Report, No. -2.

Lazarus, William C.
1961 Ten middens on the Navy Live Oak Reservation,
Santa Rosa County, Florida. Florida Anthropo-
logist, Vol. XIV, pp. 49-64.

1961A The Fort Walton Culture west of the Appalachi-
cola. Paper presented to the 161T Southeast-
ern Archeological Conference.

1965 Coin dating in the Fort Walton period. Flo-
rida Anthropologist, Vol. XVIII, pp. 221-22T7

McKenzie, Douglas H.
1965 Pottery types of the Moundville Phase. Proce-
ings of the 20th Southeastern Archeological
Conference, Bull. 2, pp. 55-64.

Moore, Clarence B.
1901 Certain aboriginal remains of the northwest
Florida coast, Part I. Journal of the Academy
of Natural Sciences, Philadelphia, Vol. II.

Morse, Wayne E.
1955 Geology of Jackson County, Florida. Geologi-
cal Publication No. 37, Florida Geological Sur-
vey, Tallahassee.

Munson, Patrick J.
1965 Stylistic seriation of selected design ele-
ments of the Southern Cult. Paper presented al
the 64th Annual meeting of the American Anthro-
pological Association, Denver.

Sears, William H.
1960 In comments on the Johns Pass mound by Ozzie
Ostrander, Florida Anthropologist, Vol. XIII,
pp. 77-79.

Smith, Hale G.
1956 The European and the Indian, Florida Anthropo-
logical Publications, No. 4.

Swanton, John R.
1922 Early History of the Creek Indians and their
neighbors. Bureau of American Ethnology, Bull

1946 The Indians of the southeastern United States.
Bureau of American Ethnology, Bull. 129.

Willey, Gordon R.
1949 Archeology of the Florida Gulf coast. Smith-
sonian Miscellaneous Collections, Vol. 113.

Wing, Elizabeth
n.d. Faunal remains from a Jackson County, Florida,
Indian site. Manuscript.

Webb, William S. and David L. DeJarnette
1942 An archeological survey of Pickwick Basin in
the adjacent portions of the states of Alabama
Mississippi, and Tennessee, Bureau of American
Ethnology, Bull. 129.



A-B, Plain rim sherds (Lake Jackson Plain). E-F. Rim sherds with curvilinear incised lines (Point Washington Incised).
C-D. Rim sherds with parallel incised lines (Fort Walton Incised). G-H. Rim sherds with both incision and punctuation (Fort Walton Incised).





I I tIi 1 t 1
C liidi tI i i, R g(ahe

Cr Utal C I r 1111

* o I 1 '1 C



c D

B 0


0 N2 H 2








James F. Small


A report on two previously unreported sites
on which Poverty Point Type Baked Clay Objects
have been found and some additional information
and comparisons on this type of artifact from
other known sites.

The report includes illustrations of Baked
Clay Objects from five widely separated sites.

This paper is not primarily concerned "with the sites
from which Baked Clay Objects have previously been reported;
however, some comment on these sites is essential. This is
particularly true of the two major sites of Poverty Point
and Jaketown and the Choctawhatchee Bay Site in Northwest
Florida. These sites will be discussed very briefly and only
in respect to the subjects pertaining to this report.


Poverty Point was first reported by Clarence B. Moore
in 1913. It was Moore who first called attention to the un-
usual Baked Clay Objects found in Louisiana during his field
work in 1913. These Baked Clay Objects have since been re-
ferred to as Poverty Point Type Baked Clay Objects simply be-
cause they were found in such abundance there. Actually,
Moore reports finding them on several sites in Louisiana and
illustrates them from the Schwing Place in Iberville Parish
and the Hopeka Plantation, Catahoula Parish in addition to
Poverty Point. Obviously they are found on quite a number
of sites adjacent to the Gulf Coast and inland, primarily in
the Mississippi and Ohio River Valleys. Moore was unable to
reach any conclusion as to what these artifacts were used
for although he hypothesized that they were some form of gam-
bling device.

The first really enlightening report as far as Poverty
Point and the Baked Clay Objects is concerned, was authored
by James A. Ford and Clarence H. Webb in 1956.

This report points out the very interesting fact that
while there is a great deal of similarity between the Baked
Clay Objects found on Poverty Point Site and those found on
the Jaketown Site, there are also unique differences. These
differences do not involve paste and size as much as they do
Florida Anthropologist, Vol. XIX, No. 2, June, 1966

classification, i.e. some types found at Poverty Point are
not found at all at Jaketown and vice versa. Actually, this
fact seems to fit all of the sites on which Baked Clay Ob-
jects were manufactured,local variants being the rule rather
than the exception although certain types have a very wide
distribution. Ford and Webb found Baked Clay Objects in and
around several cooking pits and to quote them, "This dis-
pelled any doubts as to their usage." They also say that we
can stop being scientifically diffident and refer to them as
artificial cooking stones. I would like to explain to any-
one who is not familiar with the term, the probable methods
in which an artificial cooking stone was used. Before the
invention of cooking utensils of stone and clay which could
withstand both heat and liquid, the Aborigine needed some
method of boiling water in order to make soups and stews,
etc. It is surmised that this was accomplished by digging a
shallow hole in level ground and lining it with skin or some
material through which water would not readily pass. They
would then heat, on a nearby fire, cooking stones or Baked
Clay Objects which they would drop into the soup or stew un-
til it was cooked. The presence of the liquid would keep
the stones from burning a hole through whatever lining was
used. Another method in which they could have been used was
by digging a hole in the ground and building a fire in which
the cooking stones were heated.

The cooking stones or Baked Clay Objects were then re-
moved and the hole was then lined with wet grass or leaves
and a layer of meat placed upon it. Then another layer of
grass or leaves and then the hot cooking stone or Baked
Clay Objects placed on top. The whole pit was then covered
with loose earth and the meat cooked by steaming. This me-
thod is still used by some of the Aboriginal Tribes of Aus-

The Baked Clay Objects are found in areas where stone
is not abundant or where the type of stone present tends to
fracturewhen heated and could not be used for this purpose .
There is at least one exception to this; however, I will en-
deavor to point out that the Baked Clay Objects from this
site are intrusive and were probably brought into the area
from the south. The Baked Clay Objects are stone substi-
tutes and in all probability are pre-pottery in origin, al-
though they may have continued in use through at least the
early pottery making stages.

Actually, Ford, Phillips, and Haag in the Jaketown re-
port call attention to the fact that Baked Clay Objects have
been found on sites which date from the Mississippi time per-
iod as late as historical times. Mississippi period sites
have produced a very small number of Baked Clay Objects and
they could have been picked up on an older site and redepos-
ited during the Mississippi period. There is no conclusive

evidence that they were manufactured during the Mississippi
period, and to be perfectly fair, we cannot prove at this
time that they were not made during the latter period. To
support the fact that they might have possibly been made dur-
ing this latter period, those Baked Clay Objects found on
Mississippi sites are generally smaller in size than those
found on the Archaic sites. These smaller balls may have had
a different use than the earlier varieties, however, they are
in all probability still stone substitutes of some sort.

In making the above statement the writer is intentional-
ly ignoring the similar "Clay Balls" found in the western
part of the United States and is speaking only of those
Baked Clay Objects found in the southeastern section of the


The second major site from which Baked Clay Objects are
reported is the Jaketown Site in west central Mississippi.
This report was written in 1955 by James A. Ford, Philip
Phillips, and William A. Haag. This report established the
classification and nomenclature used in describing the Baked
Clay Objects in this paper.


The third site to be discussed is the Elliott's Point
Site in Okaloosa County, Florida. This site was first re-
ported by Col. William C. Lazarus in the Florida Anthropolo-
gist in 1958. Apparently the Elliott's Point Site is just
one of many sites in the Choctawhatchee Bay area on which
Baked Clay Objects are found. The Baked Clay Objects from
north Florida, illustrated with this article, are from a
site northwest of the Elliott's Point Site on the shore of
Choctawhatchee Bay. These Baked Clay Objects were found on
the beach and actually in the water of the bay shore. This
site is rather unusual in that in addition to the more com-
mon types of Baked Clay Objects found in this area, both in-
cised and decorated forms are present. One fragmentary spec-
imen is unique in its type of decoration which appears to
have been made with a blunt tool of some sort.

The Baked Clay Objects from Elliott's Point and adjacent
areas are considerably harder than those found on other
sites. Some of them are almost like concrete in texture.
Additional Elliott's Point complex sites are reported by Dr.
Charles H. Fairbanks in the Florida Anthropologist of Sept-
ember of 1958. Dr. Fairbanks reports on three sites from
the south side of Choctawhatchee Bay. Apparently, Baked
Clay Objects are present on a number of sites in the area.


The Kelly Site in Clark County, Indiana lies in an ex-
cellent location for an Aboriginal Village Site. It is sit-
uated on a high bank of the Ohio River adjacent to the falls
of the Ohio, which lie between Clark County, Indiana and Jef-
ferson County, Kentucky. The Baked Clay Objects from the
Kelly Site have been previously reported to some extently
Ford, Phillips, and Haag in the Jaketown Site Report. How-
ever, the information available to them was incomplete at
the time the Jaketown Report was written. The site consists
primarily of a shell midden of Archaic origin; however,
there is a small woodland site adjoining it. The artifacts
recovered from the shell midden are very similar to those
found in the shell mound at Indian Knoll in the Green River
section of Kentucky. Since the building of the dam across
the Ohio River, just upstream, the site has gradually eroded
away from the washing of the spring floods each year. In
the early 1950's the site was bulldozed and hauled away to
be used in the construction of the flood wall which sur-
rounds New Albany, Indiana. In the spring of 1950 the Hale
brothers, who lived on the adjoining farm, found a cache of
Baked Clay Objects exposed in the slumped bank of the river
as the flood waters receded. The cache was found in the ar-
chaic portion of the site, however, they could have been in-

The writer arrived on the scene the afternoon of the
same day in which the cache was found and was escorted tothe
spot by one of the Hale brothers. The pit from which the
Baked Clay Objects has been removed was still filled with
fragments of Baked Clay Objects and sandy red clay where
many had been dissolved by water action. The Hale brothers
salvaged approximately two hundred specimens which represent-
ed about two thirds of the original cache. They were kind
enough to.present me with a sample of several different clas-
sifications of Baked Clay Objects which very closely resembe
those found at Poverty Point. I had hoped, after acquiring
samples of Baked Clay Objects from Poverty Point and Jaketown
that I might possibly establish the origin of the Kelly Site
Baked Clay Objects; however, the Kelly Site cache does not
identify itself with either Poverty Point or Jaketown. The
cache contained a number of examples of biconical punched
which are not found at Poverty Point and it also contained a
number of melon shaped of which only one example has been re-
ported from Jaketown. The biconical pmctate design referred
to here, also differs from the Jaketown biconical punched in
that those from the Kelly Site were punched with a small,
blunt implement of some kind while those from Jaketown show
finger tip indentations. I would like to emphasize that
Baked Clay Objects were never found on this site other than
in the single cache plus the fact that the paste used in

their manufacture was not of local origin. This leads one
to a fairly definite conclusion that they were intrusive on
this site. The paste very closely resembles that of the Pov-
erty Point Baked Clay Objects; however, as pointed out be-
fore, there were types present in the cache that are not
found at Poverty Point.


The Tick Island Midden (Vo. 24) in Volusia County on
the St. Johns River has recently produced a number of Baked
Clay Objects. To the best of my knowledge they have never
before been reported from this site or from this far south.
These objects were found by watching the washer at Mr. Wes-
ter Branton's shell yard in DeLeon Springs, Florida. This
operation was described in a previous report (see An Unusual
Incised Vessel, Florida Anthropologist, December 1964). The
paste used in the Tick Island Baked Clay Objects is complete-
ly different than that used in other Baked Clay Objects, how-
ever, this is not unusual as they seem to have used whatever
type clay was locally abundant in the manufacturing of these
objects. There is no evidence of a tempering material of
any kind being used. Some of the forms coming from the Tick
Island site, while not identical, do resemble the classifica-
tions used to describe the Poverty Point and Jaketown Baked
Clay Objects. The Baked Clay Objects from Tick Island are
much lighter in weight than those from the other sites.
This, of course, is primarily due to the type of paste used
in their manufacture. While there is no possible way to pin
the Baked Clay Objects from Tick Island down to a certain
time period, due to the manner in which they were found, the
majority of the sherds coming through the washer with the ob-
jects were Orange fiber-tempered. This would tend to place
them in a fairly early horizon at Tick Island. There is no
evidence of a micro flint industry at either Tick Island or
the Kelly Site, while this is a part of the cultural material
from the other three sites referred to in this report. The
accompanying photographs will help to illustrate the various
classifications from the five different sites discussed in
this paper. The photographs are to scale and show the com-
parative sizes of the Baked Clay Objects from the different

There are many questions yet to be answered regarding
Baked Clay Objects. Why would they spend so much time and
labor decorating so utilitarian an object as a cooking stone?
A handful of baked clay would have done.the same job. Why
do we have a basic form of decorating appearing in such
large numbers on certain sites? Did the different designs
have a special significance? Was it possible that they used
biconical to cook fish, cylindrical to cook meat and spher-
oidal to cook vegetables? Is it possible that a specific

design was intended to propitiate the spirit of the departed
animal on which it was used?

All of these questions are, of course, hypothetical and
unanswerable at this time. Future investigation and researdi
may give us some of the answers cnthis very interesting type
of artifact.


The writing of this paper would not have been possible
without the help and cooperation of the following:

Mr. Carl Alexander of Epps, Louisiana who kindly sent me
the samples of Poverty Point Baked Clay Objects.

Mr. Elston Fagan of Ft. Walton Beach, Florida who
loaned me the Baked Clay Objects from the Choctawhatchee Bay

Dr. Charles H. Fairbanks of the University of Florida,
Gainesville, Florida who loaned me the Baked Clay Objects
from Jaketown.

Mr. Carl Benson of Orlando, Florida who took the accom-
panying photographs.

Dr. James A. Ford of the Florida State Museum who ass-
isted in several ways.

To these gentlemen I would like to express my most sin-
cere appreciation.


Fairbanks, Charles H.
1959 Additional Elliott's Point Complex Sites, The
Florida Anthropologist, Vol. XII, #4.

Ford, James

Ford, James

A., Phillips, Philip and Haag, William G.
The Jaketown Site in west central Mississippi.
Vol. #45, Part 1, Anthropological Papers of the
American Museum of Natural History.

A. and Webb, Clarence H.
Poverty Point, a late Archaic site in Louisiana
Vol. #46, Part 1, Anthropological papers of the
American Museum of Natural History.

Lazarus, William C.
1958 A Poverty Point Complex in Florida.
ida Anthropologist, Vol. XI, #1.

The Flor-

Moore, Clarence B.
1913 Some Aboriginal Sites in Louisiana and Arkansas
Journal of the Academy of Natural Sciences of
Philadelphia, Second series Vol. XVI.









































Carl J. Clausen


During the summer months of 1965, a commer-
cial salvage company working under a state-issued
salvage lease utilized a privately developed and
technically advanced proton-type magnetometer to
systematically survey and chart the distribution
of ferrous components of an historical shipwreck
off the Florida East Coast. This paper will re-
port on this new marine application of the magneto-
meter as a potentially useful interpretive tool
for the archeologist interested in underwater his-
torical sites.


The application of the magnetometer is not new in the
search of sunken ships by persons interested in the recovery
of valuables (Potter 1960:52-54), or by those with a genuine
but untrained interest in the recovery of historical objects
from underseas wrecks (Link 1959:71-75). Until recently,
however, utilization of this instrument, a device for measur-
ing total magnetic field intensity in a given area, has been
more or less limited to attempts simply to locate wrecked
vessels. There apparently have been no successful attempts
to further utilize the instrument in plotting the distribu-
tion of the ferrous elements of an individual wreck as an in-
terpretive aid. This restricted usage has arisen not
because of any lack of ingenuity or experimentation on the
part of the salvagers, who readily developed quite adequate
systems for searching large areas (Harnett 1962), but prima-
rily because of certain limiting factors in most currently
available units, designed principally for various geophysi-
cal applications where read-out frequency is less critical,
have response rates ranging from once every two seconds to
about twice a second. This rate is not sufficiently rapid
to be used in the efficiency-dictated survey method; i.e..
manually navigating a motor vessel trailing the sensina ele-
ment through a predetermined pattern, to produce useful

The mechanics of the system described in this report,
essentially the same as that used by Black and Johnston in
their survey of the Angel Site in Indiana (1962:199-205),
Florida Anthropologist, Vol. XIX, No. 2, June, 1966

were worked out with the author's urging by two principals
of Treasure Salvors Incorporated, affiliated with the Real
Eight Comoany which holds the State lease off the East Coast
of Florida. The instrument used was a privately constructed
proton-type magnetometer of advanced design specially devel-
oped for use in prospecting for wrecks. Adequate descrip-
tion of the theory and mechanics of the various types of mag-
netometer is available elsewhere and need not be discussed
here. This unit's principle advantage over the commercially
available instruments referred to previously lies in its
ability to make measurements at rates of either four times
per second with an estimated 1 Gamma sensitivity, or with
lower resolution at eight times per second.


The survey was carried out from a small fiberglass in-
board-outboard vessel which was equipped with an engine revo-
lution counter and a compensated marine compass, and was
stocked with a quantity of buoys made with empty bleach bot-
tles for floats and concrete blocks for anchors. An accu-
rate stop watch was carried.

The subject wreck lies in shallow water close to shore
some 4,200 yards south of the Sebastian Inlet on the Florida
East Coast below Cape Kennedy. Archeological and documen-
tary evidence supports this site as the location of the
wreck of the flagship or Capitana of General Don Juan de
Ubilla. commander of the Flota portion of the Spanish fleet
wrecked in this area in 1715 (Fernandez Duro 1900: Vol. 6,
121-27). Ashore, opposite the wreck on a narrow island, is
the site BR-139, the 1715-16 camp of the survivors and salva-
gers of the fleet (Smith 1956:88-94).

The salvagers had worked at this site for several sea-
sons and made significant recoveries;l but due to the strong
surge, limited visibility, shifting sands and scattered con-
dition to the wreckage, there was little factual knowledge
concerning the extent or orientation of the wreck.

The survey required the systematic exposure of the
wreck site to the magnetometer. This was accomplished by
towing the sensing element of the instrument behind the sur-
vey vessel at a constant speed back and forth over the wreck
site and surrounding area on a series of equidistant par-
allel courses. The spacing of the courses was dictated by
the desired overlap in readings, the depth of the water and
the sensitivity of the instrument. Runs too closely spaced
were inefficient, as the intense magnetic disturbances or
anomalies created by massive iron objects such as cannons or
anchors cancelled out the smaller anomalies, while wide spac-
ing naturally resulted in incomplete data. For the purpose

of this survey, a spacing was chosen which insured, it was
felt, that large anomalies would appear only on adjacent
courses. This compromise appeared to assure that a represen-
tative sample of the smaller ferrous components would be in-
cluded in the survey. To visually guide the helmsman in the
surveying process, a grid system of buoys was installed over
the area.2 The crid. oriented on a base buoy surveyed in
from ashore, was laid in the following manner: The X and Y
axes were established by the survey vessel closely passing
the base buoy while proceeding at a set speed (tachometer)
on a magnetic heading and setting a predetermined number of
buoys at timed intervals (stopwatch). This process was re-
peated on a course at 90 degrees to the first. The balance
of the grid was set using the same technique; timed runs par-
allel to the first axis at intervals of the second.

Data on magnetic distortions encountered by the instru-
ment in surveying the area were permanently recorded on a
continually moving graph. The operator of the magnetometer
noted on the moving tape the coordinates and direction of
the particular course through the grid and the point at
which the sensing element entered and left the pattern and
crossed grid lines perpendicular to the direction of travel.
In this manner a record was produced of the anomalies, their
intensity and location, from which it was a simple matter to
construct a magnetic chart for the area in which the major
ferrous components of the wrecked ship, i.e., the cannons,3
anchors, main structural fasteninas and concentrations of
smaller fittings and ship's gear were located.

Prior to the destruction of the vessel, the various fer-
rous items mentioned above enjoyed an intimate physical rela-
tionship with the balance of the material comprising the ves-
sel and its cargo. It is assumed, with certain qualifica-
tions, for the purpose of this paper, that this relationship
still exists and that the distribution of these ferrous com-
ponents is generally reflective of the distribution of the
remaining elements of the wreck.5

The principal qualification is, of course, that these
remaining elements, in many instances of considerably less
specific gravity than iron, would be affected to a greater
degree by the scattering action of waves and currents. Some
of these materials would therefore be encountered scattered
outside of the ferrous nucleus of the wreck in a sort of
"halo" effect. Naturally the circumstances of a vessel's
demise have a great effect on exactly how close this rela-
tionship would be. In the case of the vessel surveyed,
wrecked on a weather shore during a hurricane, it would be
expected that the correlation would be at its lowest. Un-
fortunately, at present the technique is new, and insuffi-
cient data available, to go much beyond observing that there
are strong indications that the majority of the surviving

elements of a wreck will be associated with this ferrous dis-

By associating to scale the data from the magnetic sur-
vey with a map drawn from both black and white and color
aerial photos6 of the underwater terrain and features ashore,
a graphic portrayal of the ferrous element of this wreck's
distribution in relationship to its physical surroundings
emerges. (See accompanying figure.)


An examination of the results of the survey as plotted
clearly indicates that the remains of the ferrous components
of the vessel, its gear and cargo, lie dispersed in a cres-
cent-shaped area of more than eight acres. This scattering
which appears in other 1715 wrecks (Clausen 1965:1-5, 27),
can be attributed to the circumstances of the loss of the
fleet. The testimony of the survivors of the disaster re-
lates that most of the vessels were driven aground during
the height of a severe hurricane during the night of Wednes-
day, July 31, and the early morning hours of August 1, 1715.
Further statements place the wind which drove them ashore as
coming form "just north of the east" with an intensity which
caused the sea at first light to appear as "shooting arrows"
to those huddled ashore (Menendez. Captain Sebastian and
Barriga, Captain Don Fernando Ignazio. Testimony in A.G.I.

Under the above conditions. most of the vessels prob-
ably suffered massive structural damage on or shortly follow-
ing constant with the submerged ribbons of lime and coquina
rock paralleling the coast. The intense action of the sea
apparently dashed the vessels to pieces, scattering wreckage
over acres of offshore bottoms.

The documents previously cited also contain statements
which seem to indicate that the Capitana, probably a frigate
40-50 guns, may have sectioned horizontally, possibly along
the turn of the bilae, upon impact with the offshore rocks
and that the superstructure broke up closer to shore.

The nature of the demise of the Capitana alluded to in
these documents may account for the General "crescent shape"
of the wreck's distribution. With the east winds working
seas against the coast, which in this area tends generally
northwest, a powerful northward current would be generated
along the shore. The disintegrating superstructure lightened
by the loss of the ballast and possibly most of the heavy,
registered cargo of silver, would be forced to the west by
the action of the seas and wind; and, as it approached the

shore, would come increasingly under the effect of the power-
ful northward current paralleling the beach. Under the re-
sultant of these two forces, the wreck would be expected to
follow a more or less elliptical path as it broke up.

Evidence amassed from different areas of the wreck dur-
ing several seasons of salvage supports the above explana-
tion. For example, during the 1965 salvage season most of
the work was concentrated on the eastern. seaward arm of the
wreck, which significantly was unknown to the salvagers
prior to the magnetometer survey. In this area were found
dozens of 50 pound discs of copper and more than a ton of
both clumps and loose silver coins, including one nearly in-
tact example of one of the 1,300 wooden packing cases of 250-
plus pounds of silver registered to the ship. This heavy
cargo normally would be stowed low in a ship, possibly along
the ballast. The fact that most of the heavy forged rudder
hanging gear--pintle and gudgeon assemblies--were recovered
near the copper discs and silver coins further supports this
area as the approximate position of the vessel's bottom.

Nearer shore and along the arm of the crescent parallel-
ing the beach where they would be expected to be, after rid-
ing the gun and upper decks of the disintegrating superstruc-
ture, were most of the 40 or so cannons carried by the war-
ship. In this shoreward area the salvage companies have in
past years also recovered intact examples of Kang H'si
period porcelain sweetmeat bowls and teacups, whole Guadala-
jara ware vessels and examples of intricate gold jewelry,
materials which would generally be found stored higher in a
vessel's cargo.


The systematic survey of an underwater historical
period shipwreck site usina an advanced proton-type magneto-
meter has permitted us to plot the distribution of the major
ferrous components of the wrecked vessel in situ.

The assumption is made, based on the close relationship
these components once enjoyed with the balance of the vessel
and its cargo, that their distribution is representative to
a degree of the distribution of the remaining elements of
the wrecked ship. Preliminary recoveries made on this wreck
tend to support this assumption. Additional evidence based
on further recoveries of material from this and other sur-
veyed wrecks is needed to qualify this relationship.

Utilizing the magnetic survey data in conjunction with
maps drawn from aerial photos, the main distribution of a
wreck, even one deeply buried in sand, can be plotted. This
affords the archeologist or other researcher important and
often otherwise unavailable preliminary information concern-
ing fruitful areas for recovery operations.


1 V






This distributional data alone or with additional infor-
mation supplied from historical references and actual recov-
eries from the wreck can lend important insights into the
probable circumstances of the vessel's demise, aiding the ar-
cheologist in a more complete interpretation of the site.

With further refinement the technique holds promise of
developing into a useful interpretive tool for the archeolo-
gist interested in underwater historical period wreck sites.


1. A site report covering in detail the materials recovered
from this important 1715 wreck is under preparation and
should be ready for publication in the near future.

2. The accuracy of this type of survey could be improved by
utilizing a short range electronic navigating system
coupled with an onboard computer and programmed guidance
mechanism. However, the cost of such a system would be

3. Surprisingly, considering the many sources of error in
the system, when intense anomalies on the chart were
checked against several known locations of, for example,
cannons on the wreck, it was found in some cases that the
chart was accurate to within an estimated eight feet.

4. It is of interest to note that according to official sour-
ces the Bounty, the wooden sailing vessel built from the
keel up at Lunenburg, Nova Scotia, by Metro-Goldwyn-Mayer
for the motion picture Mutiny on the Bounty, required
nearly 29 tons of bolts, spikes and bar iron for forging
in its construction.

5. For a detailed description of what materials tend to sur-
vive in the hostile environment afforded by the sea, with
an explanation of why, see Mendel Peterson's article on
"The Conditions of Materials Found in Salt Water" in
Diving into the Past, 1964.

6. The aerial photographs were taken with a 35 mm. single
lens reflex camera from an altitude of approximately 3000
feet. The filter system included a polorizer, a chrome
haze filter and various color filters. Both black and
white and color positive films were used, the latter be-
cause it was learned that certain nuances of color, rele-
gated to various shades of grey in black and white photos,
were important in interpreting bottom terrain. The color
positive feature offers an additional advantage in that
they may be projected, providing inexpensive though tem-
porary enlargement to any desired scale, a factor val-
uable for study as well as in producing charts.


Black, Glenn A., and Johnston, Richard B.
1962 "A Test of Magnetometry as an Aid to Archeology,";
American Antiquity, Vol. 28, No. 2.

Clausen, Carl J.
1965 "A 1715 Spanish Treasure Ship." Contributions of
The Florida State Museum, Social Sciences, No.12

Fernandez Duro, Cesareo
1900 Armanda Espanola, Vol. 6. Madrid.

Harnett, Charles B.
1962 Operation Lodestone -- A Marine Archeological
Exploration of Cape Canaveral and the Florida
Keys Utilizing the Varian M-49, a Proton Magneto-
meter, to determine the Location of Ancient Ship-
wrecks. Paper printed for limited distribution
by the author.

Link, Marion Clayton
1959 Sea Diver. New York



Mutiny on the Bounty. A phamphlet prepared by MGM
on the production of the motion picture of the
same name.
Mendel L.
The condition of materials found in salt water
In Diving into the Past. Proceedings of a con-
ference on Underwater Archeology Sponsored by
the Minnesota Historical Society, St. Paul,
April 26-27, 1963. Ed. Holmquist, Wheeler. The
Minnesota Historical Society and the Council of
Underwater Archeology.

Potter, John S., Jr.
1960 The Treasure Diver's Guide. New York.

Smith, Hale

"The European and the Indian," Florida Anthropo-
logical Society Publications No. 4, Ga7inesville.

Unpublished sources.

AGI 58-1-30, Documents 49 and 491, January 23,
1716. North Carolina Department of Archives and
History, Raleigh, North Carolina.


James W. Covington

During the period from 1870 to 1910 thousands of emmigr-
ants from eastern and southern Europe poured into the great
industrial cities of the Eastern and Middle Western United
States. In order to protect themselves in a seemingly hos-
tile land these newcomers moved into ghettos known as "Lit-
tle Italys", "Little Warsaws", and "Little Budapests", situ-
ated in the crowded tenament neighborhoods from which the
previous inhabitants had left in search of better homes
which denoted an improved social status.

After the early residents who had been of English,
Irish or German background had departed, the settlements as-
sumed an Old World atmosphere. For example, in a Polish
neighborhood in St. Louis, a weekly newspaper completely
printed in Polish was published for many years and several
Polish sports clubs for young men flourished. It was typi-
cal of such neighborhoods to contain social clubs, newspap-
ers, restaurants and churches in which only the national lan-
guage and cultural traits were used. It was as if part of
Warsaw or any other eastern European city had been transpor-
ted several thousand miles to another continent but for many
of the inhabitants it had been merely the exchanging of one
slum area for another in a slightly different form ( Wish,
1952, 238-270).

In competition for jobs the newcomers were severely re-
stricted. Few of them had any command of English and conse-
quently they were forced to take unskilled positions in com-
mon labor which required little use of the English language.
Of course, such positions did not pay very much. Yet, condi-
tions in the "Promised Land" were much better than wages and
life in Europe and letters written to overseas friends told
about better food, clothing and a whole new way of life.

Ybor City, a part of Tampa, Florida, was settled in
1886 by a group of Cuban cigar workers from Havana, Key West
and New York City. Ybor City had much in common with the im-
migrants'settlements found elsewhere in the United States,
but there were some important differences. First and perhaps
most important of these differences was the fact that Ybor
City was solely established for these immigrants and did not
represent a tenement or slum area at that time. Second, it
must be noted that most of the Cubans had not come to the
United States in search of greater opportunity but had fled
here to avoid the Spanish soldiers who had committed many
brutal acts in Cuba during and after the Ten Years War (1868-
1878). Even Vincente Ybor, a native of Spain, had moved his
cigar business to Key West in 1869 when the war assumed seri-
ous proportions.
Florida Anthropologist, Vol. XIX, No. 2, June, 1966

In 1885, Vincente Martinez Ybor decided to search for a
spot where he would locate a branch unit of his Key West Ci-
gar Factory operation and selected Tampa as a suitable site.
With the financial assistance of the Tampa Board of Trade
more than a thousand acres of land were purchased and plans
were laid for the establishment of Ybor City. The site sele-
cted for the town was mostly covered with pine trees, Dalmet-
toes and oak trees and had served as a cattle range (Tampa
Tribune February 23, 1894). There were several orange gro-
ves, marshes, small lakes and isolated farmhouses scattered
throughout the area. The primary job at this time was to
cut down the trees, clear the land, plot the town and mark
off the streets, factory and home sites. At six in the morn-
ing of October 8, 1885 a force of woodcutters moved into the
pine forest and began clearing the Florida wilderness (Tam-
pa Journal February 10, 1890).

In planning his development Ybor believed that he could
make more of a profit by inviting other cigar factories to
locate free of land costs in the area and selling cottages
to the workers. In order to separate his cigar business
from his land operations, Ybor established the Ybor City
Land and Improvement Company and land in blocks of 450 by
650 feet was offered to any cigar manufacturer who would lo-
cate in Tampa. Of course, Ybor made a profit by erecting
frame buildings to house the cigar factories at an average
price of six thousand dollars (Tampa Tribune May 25, 1894).
Frame cottages for the cigar factory workers were sold on a
time installment plan at a price that averaged about nine
hundred dollars. Monthly or weekly payments ranged from
thirty-five to seventy-five dollars a month (Tampa Weekly
Journal May 12, 1887

Before purchasing homes, many male workers resided in
the numerous boarding houses and small hotels until their
families were able to join them or until they had sufficient
funds to make the down payment for the house and purchase
suitable furnishings. None of the frame houses at that time
had running water, electric lights or inside toilets. Side-
walks were paths worn in the sand.

Workers for the cigar factories came from Key West, Ha-
vana and several Northern cities. One person arriving from
Havana in 1886 recalled that due to the shallow water his
ship was not able to dock in Tampa and he used a smaller
landing boat to travel from the ship to the dock in downtown
Tampa (Del Rio, 1950, P. 6). An elderly worker recalled
that many of his friends had come to Tampa tightly packed to-
gether in the hold of the ship (Tampa Tribune, September 8,
1965). When a fire destroyed Ybor's factory in Key West dur-
ing March of 1886, he decided to shift his entire operations
to Tampa and most of his Key West workers came by ship to
Tampa and retained their positions (Del Rio, 1950, P. 9).
Florida Anthropologist, Vol. XIX, No. 2, June, 1966

Although Ybor had been first on the scene, it was a com-
petitor's factory that was able to manufacture the first ci-
gars produced in Ybor City. When Cubans refused to work
alongside a Spaniard at Ybor's plant, a stalemate developed
and Sanchez and Haya were able to commence operations during
the Spring of 1886 and became known as Factory Number One
(Grismer, 1950, P. 183). Finally the Spaniard was dismissed
and Ybor's factory became the second one to make cigars. In
1886 R. Monne's factory of New York was lured to Tampa by an
offer of $15,000 by Ybor. Mr. Monne stressed the use of do-
mestic tobacco from New York, Connecticut and Wisconsin but
the other manufacturers used tobacco imported from Cuba
(Tampa Weekly Journal, September 17, 1888). When Monne arri-
ved, he proclaimed that he needed a thousand workers and off-
ered to hire entire families including men, women and child-
ren. Due to the widespread use of child labor in the vari-
ous cigar factories, Hillsborough County was the first coun-
ty in the state to adopt a local option compulsory school
attendance law in 1918.

Within a short time after its establishment Ybor City
resembled a portion of Havana transported to Tampa. One
newspaper writer referred to the place as "our Havana" and
told about his walk along the streets listening to the sound
of piano and guitar music and sons suna in Spanish (Tampa
Journal, February 14, 1889). Restaurants in the Ybor City
neighborhood were given Spanish names but served both Cuban
and Anglo-Saxon meals. Hotels bore such names as Hotel de
la Habana and Hotel Pasaje (a famous Havana hotel).

One unique Cuban influence was the reader in the cigar
factory. In the morning he mounted a small pulpit and read
aloud to the workers the latest news from a recent Havana
newspaper and, during the afternoons, Spanish novels and bal-
lads were featured (Tampa Journal, February 10, 1890).
When he desired a rest, the reader smoked the best cigars
and drank a little rum or sipped some black Cuban coffee
from a demitasse. The reader was not on the payroll of the
factory but was supported by contributions of twenty-five
cents per worker per week. It was estimated that one news-
paper editor working as a reader received the princely pay
of $125.00 a week. The average cigar worker received be-
tween fifteen and twenty dollars a week but a few packers
were able to earn as much as fifty dollars. In a fashion fi-
miliar to that of the reader, the workers were able to enjoy
an average of seven cigars a day the record smoked for one
day was twenty cigars. The young boys were encouraged to
smoke a six inch cigar known to the workers as a dead horse.

Social and fraternal clubs were soon organized. In AD-
ril, 1890, the El Liceo Cubano was formally opened as a re-
creation center designed principally for the male Cubans.

Most of the dedication day's speeches were in Spanish but
one speaker, Colonel Macfarlane spoke in English (Tampa
Journal, April 17, 1890). The frame building which had for-
merly housed Ybor's original factory served very well as a
social center. The upper floor was converted into a theatre
and many outstanding plays from Cuba and Spain were present-
ed there. (Del Rio, 1950, P 11). In fact, to many observ-
ers the productions were superior to the ones produced in
Anglo-Saxon Tampa. The Knights of Light, a Cuban secret soc-
iety founded in Philadelphia in 1868 had several lodges in
Tampa. Other fraternal organizations organized in Ybor City
included the Odd Fellows and Centro Espanol. In 1902, when
the Centro Asturiano Hospital was founded, a delegation from
the Central Club in Havana came to Tampa and remained fcr
many months to insure that the principles of the Cuban organ-
ization would not be ignored. Privileges of this organiza-
tion included medical benefits a group service which was
very unusual in the United States at this time.

Gradually Anglo-Saxon names were hired as teachers. At
tirst the classes were not too successful for parents wanted
their children to work in the cigar factories. At least one
girl claimed that she did not attend school for fear she
would encounter some of the numerous snakes or alligators en
route. This was probably true. In order to preserve inter-
est in speaking Castillian Spanish several private schools
were organized and at least one conducted classes from seven
to nine each evening but these institutions were not very su-
ccessful. Gradually the Spanish spoken in Ybor City declin-
ed in quality and by 1966 a Cuban refugee has a somewhat dif-
ficult time understanding the dialect known as "Ybor City
Spanish" which is mixed with Italian, English and words of
independent or obscure origin.

Most of the residents of Ybor City had been exposed to
strong Roman Catholic influences in Cuba and this denomina-
tion became established in Tampa. The first Mass was conduc-
ted in a private home and the sermon was delivered in Eng-
lish by Father Quinland but all succeeding sermons were con-
ducted in Spanish by Father De Carricre (Del Rio, 1950, P.
48). The Baptist and Congregational groups used Anglo-Saxon
missionaries at first but soon developed Latin followers and
ministers. The W.C.T.U. held its meeting at the Baptist Ch-
urch and attempted to make Ybor City and Tampa dry but fail-
ed by a decisive vote of 232 to 63 in Ybor City (Tampa Week-
ly Journal, October 5, 1887).

At first the Cubans restricted themselves to the cigar
industry and Anglo-Saxons assumed control of other types of
businesses. One of the first saloons was run by a German na-
med Mugge. The Clark Brothers established a dairy and the
Ybor City Ice Plant was operated by a non-Latin owner. Offi-
cers of the Bank of Ybor City founded in 1905 included such

non-Latin personalities as John Trice, President, Adam Katz,
Vice-President and George Simpson, Cashier. Mr. Ybor hired
George Chamberlain to serve as Secretary and Treasurer of
his land company and completely handle the financial affairs
of the firm. Even a few Anglo-Saxons worked in the cigar

Gradually the Cubans adjusted to their new environment
and prepared for a long term tenancy in the enclave. A week-
ly newspaper La Rivista de la Florida, published in Spanish
commenced operations in 1888. The editor was Pamon Rivero y
Rivero. He was a Cuban journalist who had visited Key West
and Tampa and wrote favorable stories about Tampa in order
to lure the workers from Key West (Tampa Weekly Journal,
August 17, 1888). One person named .Garcia opened a dry
goods store and another Garcia operated.a saloon. Francisco
Velasco advertised his imported groceries as being the cheap-
est in Ybor City. Within several years Ybor City was annex-
ed as the Fourth Ward of Tampa and the Latins demonstrated
their block voting in the 1894 election for a Fourth ward
councilman: Haya 189, Ramirez 207, Pena 127, Filograma 118
and Meyer 10 (Tampa Tribune, March 8, 1894).

To the average worker in the cigar factory it was as if
he had not left Havana. He worked in a factory where Eng-
lish was not used at all and many persons did not bother to
learn the language. Cuban foods including yucca, malanga,
black beans and white rice were brought to Tampa by the week-
ly tobacco ships and were sold in the Ybor City grocery sto-
res. At night the men visited the various social clubs
where they acted in a fashion typical of Cuban men. Roads
near the beaches were called "Little Malecons" (a drive
along Havana Harbor) and charcoal instead of wood was used
for fuel purposes in the homes. Of course, the Fourth of
July was observed but larger parades were held commemorating
the several dates held dear by Cuban freedom fighters.

The Cubans demonstrated their love of music on all pos-
sible occasions. Picnics were held at De Soto Park, Port
Tampa and Ballast Point and one feature of the picnics was
the presentation of plays and musical numbers. Cuban food
was consumed at the picnics and one favorite of the crowds
was sidera, a Spanish drink made of fermented apple juice.

Thus we have seen how the Cuban enclave was started
in Ybor City. Other elements including Spanish and Italian
joined the Cubans but Ybor City remained essentially Cuban.
It became famous for its Cuban sandwiches and bollitos -
foods no longer known to the people in Cuba. Many persons
would live and die in Ybor City without having any reason to
learn English at all.


Cash, William T.
1938 The Story of Florida, Vol. 2, The American His-
torical Soclety, Inc., New York

Covington, James w.
1955 The Story of Southwestern Florida, Lewis Histo-
rical Publishing Company, New York.

Del Rio, Emilio
1950 Yo'Fui Uno de los Funadores de Ybor City, Pri-
vately printed, Tampa, Florida.

Grismer, Karl H.
1950 Tampa, St. Petersburg Times Publishing Company,
St. Petersburg, Florida.

Handlin, Oscar
1932 The Uprooted, Little Brown and Company, Boston.

Kennedy, John F.
1964 A Nation of Immigrants, Harper and Row, New
Torkz- -

Wish, Harvey
1952 Thought in Modern America, Logmans, Green and
Company, New York.


Tampa Journal

Tampa Tribune


David W. Chase


Excavation of an ancient river terrace now
bordering the Tensaw Creek in Lowndes County, re-
vealed a succession of superposed preceramic camps.
Evidence uncovered involved almost a continuous
occupation of the site from Dalton-Big Sandy times
through Early Woodland. The significance of this
particular project involved a confirmation of the
Cambron-Hulse projectile point typology for Ala-
bama (Cambron and Hulse, 1964). The project also
points up the need for closer working relation-
ships with the geologist in determining very early
human settlement patterns as dictated by the geo-
morphology of the terrain involved.

In July, 1965, the exploration of the Tensaw Creek Site
began with very promising results. Random tests made since
the spring of the same year implied the presence of a con-
tinuum of occupational evidence to a depth of three feet and
involving a number of cultural manifestations insofar as
artifact differences revealed.

The Tensaw Creek Site (1 Lo 9) is located on an old
river terrace or ancient meander bend which is now partially
filled with swampy vegetation and through which flows the
Tensaw Creek. The present position of the Alabama River is
roughly two miles north of the site. (Fig. 1) Owing to
its position among stream cut terraces which characterize
the upper levels of the meander plain, the site enjoyed a
certain protection from winter winds as it lies in the lower
slope of a sandy ridge located immediately to the north-west.
The soils on the site consist mostly of loose dark red or
brown sands in the upper thirty inches. Below this, sands
become much lighter and grade into gravel-sands and finally,
into pea-to walnut-sized gravels which may have characteri-
zed the river terrace at the time of its initial occupancy.


The excavation procedure involved the reduction of the
site surface into five foot square units which are referred
to as 'sections'. These were staked out for control pur-
poses along a pre-established base line which in turn was
mapped in from a datum point. Each section was ultimately
excavated in six inch arbitrary levels materials from each
level being kept separate by section to insure valid inter-
pretation of the site stratigraphy. (Fig. 2)
Florida Anthropologist, Vol. XIX, No. 2, June, 1966

By the end of September, 1965, a total of thirty sec-
tions had been explored. Artifact counts were moderately im-
pressive considering the nature of the site and the amount
of time which was spent on its excavation. Manpower was
drawn from amateur volunteers, mostly members of the Montgo-
mery Archeological Society who could only work on week-ends.

A total of 2,478 sherds of pottery and 565 stone
objects, including 264 projectile points were recovered dur-
ing the entire project.

There were a great variety of projectile points, most
of these having been typed by Cambron and Hulse in their
recent publication on this subject (Cambron and Hulse, 1964).
Only two types found pertained to a Late Woodland prove-
nience (Noluchucky and Knight Island). The balance were as-
signable to Transitional, Archaic (early and late), Late
Archaic and Early Woodland. Since the definition of Middle
Archaic in the southeast is not always too clear on a typo-
logical basis except on widely differentiating time esti-
mates, I merely employed only two terms in referring to Ar-
chaic, one being 'Archaic' and the other 'Late Archaic'.


Eight projectile points were found which could be de-
finitely assigned to a pre-Archaic or 'Transitional'horizon.
These included three Dalton types (two Colberts, one Green-
briar), one Hardaway, a Beaver Lake and a Paint Rock Valley.
All of these finds occurred in either level 4 or level 5.
No artifacts were found below level 5 in any section.


This category, as it has been mentioned, would embrace
Early to Middle Archaic known projectile point types. Most
frequently found of the main types recognized were: Big San-
dy I, Morrow Mountain and Benton Stemmed. The vertical dis-
tribution of these shows the predominance of Morrow Mountain
point types coming from level 3 (sixteen specimens from this
level); the same can be said of the Benton Stemmed group--
most having been recovered from level 3 (out of nineteen
found, ten came from level 3). Among the Big Sandy I types,
a point that is considered an Early Archaic time marker in
the southeast, twenty nine were found, twenty-two of these
being in levels 4 and 5.

Of other Archaic types, five Kirk, and one Stanley were
identified. These together with types indicated by Cambron
and Hulse as being either Early or Middle Archaic were dis-
tributed as follows: Level 1: six; Level 2: twelve; Level 3:
thirty-eight; Level 4: eighteen; and Level 5: fourteen.


Of those types considered to be in this time level, Sa-
vannah River (27 found), Elora (26 found) and Gary (9 found)
were the most numerous in that order and best represented
the Late Archaic usage of the site (Fig. V). In all, 78 pro-
jectiles were recovered which could be appropriately as-
signed to a Late Archaic provenience. In the over-all anal-
ysis, the vertical distribution of these at Lo 9 indicated
the highest concentration in Levels 2 and 3. It would appear
possible that certain of these had ceramic associations,
possibly with the fiber tempered producing zone. In the
distribution of the Savannah River type points, a trend to-
ward the main concentration in Level 3 is seen. Type distri-
bution of the Savannah River specimens in Levlls 1 and 2 sug-
gests a continued usage of the point into Earliest Woodland
times. Not all Elora configurated points conformed precise-
ly to the Cambron and Hulse illustration or description but
rather implied variant forms. The latter were regarded thus-
ly and for typology purposes in this paper, were called
Elora Points. The vertical distribution of the Elora type
point indicated a concentration mainly in Levels 2 and 3 .
The Gary Point, although originally typed in Texas (Suhm and
Krieger, 1954) seems to be fairly common in Alabama collec-
tions and its occurrence on the Tensaw Creek site was no
surprise. Only 9 specimens were found, however, 8 of these
being evenly distributed between Levels 2 and 3 with 1 being
found in Level 1.


The presence of a high concentration of pottery in
Levels 1 and 2 implied a terminal occupation of the site in
Woodland times. Found together with the ceramic specimens,
often in direct association, were a number of projectile
types which Cambron and Hulse consider either Early Woodland
or else transitional from Late Archaic into Early Woodland.
Predominant among these were Swan Lake (44 fouhd); Mud Creek
(12 found) which group also included a number of stemmed
forms which could better be described as 'Sugar Creek' in
terms of the illustrated example which appeared in the jour-
nal of Alabama Archeology, Vol. VIII (DeJarnette, Kurjack
and Cambron, 1962), p. 68, and Candy Creek of which 8 clas-
sic examples were recovered. A total of 91 projectile
points were found which could be possibly assignable to
either Late Archaic-Early Woodland Transition Period or
Early Woodland. Of these, approximately 69% were found in
Level 1, 25% in Level 2 and the balance distributed between
Levels 3 and 4.


A total of 36 categories of non-projectile point arti-

facts of stone were recovered. These include knives, scrap-
ers, pebble hammers, worked blanks or objects in process,
choppers (usually of schist or sandstone), drills and ste-
atite bowl sherds. Certain of these appeared to be diagnos-
tic of earlier or later occupational levels in the site. Of
the latter, special mention must be made of two tool types.
One of these was the end or 'thumbnail' scraper and side
scraper group, and the other involved very small retouched
flake tools. Fifteen end scrapers were found, all but one
being in the lowest three levels. Twelve were recovered
from levels 4 and 5 implying a very early provenience for
this type of tool. All of these were of the plane-convex
type. Among the side scrapers, two types were involved-
elongated and 'spokeshave' type. Only 7 scrapers of the
'spokeshave' variety were found, 4 of these were in Levels
4 and 5. Steatite sherds were 10 in all plus one possible
pipe bowl fragment. Six came from Level 2 and 4 each from
Levels 1 and 3. Steatite is obviously late on the site and
probably used well into pottery times to supplement the ves-
sel supply.

Of the small flake tools, two categories were construed:
'small flake knife' and 'retouched flake'- The former ob-
ject was nothing more than a percussion flake whose longest
edges or edge had been finely retouched to produce an effi-
cient cutting blade. Of these 6 were found, 2 in Level 3, 3
in Level 4 and 1 in Level 5. The retouched flake was simply
a miniature scrapers, scraper-knife or very fine cutting or
scraping tool used possibly for garment making. Of these
26 in all were recovered and distributed as follows: 3 in
Level 2. 6 in Level 3, 8 in Level 4, and 9 in Level.5. The
occurrence on the site of both of these tool types implied
Early to Middle Archaic usage.

In the ground stone group, one fragment of a greenstone
celt was found in Level 1, a small fragment of what appeared
to be galena, also from Level 1, and four specimens of
ground sandstone--possibly a gorget--came from Level 3. Six
specimens of.roughly battered or semi-ground (one face only)
pebbles were found. These may have been grinding stones or
manos. All of these were found in the lowest three levels.
Two hollowed stones, probably metates, came from Levels 4
and 5 respectviely.

Pottery (see Table I)

The bulk of ceramic finds was confined to Levels 1 and
2. Of the few sherds found in lower levels, most seem to
have trickled down animal burrows or had fallen into burned
out root holes.

In general, ceramics can be reduced to four main groups.
Two of these are known through publications and two are as

yet undescribed in archeological literature.

Fiber Tempered

It is generally accepted that fiber tempered pottery is
the very earliest type in all parts of the southeast. The
type recovered on the Tensaw Site was a plain, soft ware
with a brown to yellow surface and black core (seen in fresh-
ly broken specimens). Surface vermiculations were common.
Rims were straight with rounded lips. The type appears to
be mostly like a Ceorgia variant known as Stallings Plain.
(Griffin, James B., 1943) In all, 87 sherds were found, 70
of these being in level 2, the rest in Level 1, 3. and 4. It
is clearly the first pottery on the site in terms of verti-
cal position. One pit (feature 15) was assianable to the
fiber temper level. This was a deep feature whose lower con-
tents included a large number of charred hickory nuts and
hackberry seeds. A dating sample was taken from this fea-

Check Stamped

A total of twenty-five check stamped sherds were found.
These were distributed between Levels 1 and 2 with 10 in Le-
vel 1 and 15 in Level 2. Two tetrapodal bases were involved
indicating an early position in the pottery series. The re-
lated known type is possibly either Cartersville Check Stamp-
ed on the Chattahooche River or Wright Check Stamped in the
Guntersville Basin (Heimlich, 1952).

Other Stamped forms (Excepting Tensaw Stamped)

These involved Simple Stamped (3 sherds), Fabric Impres-
sed, probably a Dunlap Fabric Impressed variant (6 sherds),
and 1 sherd of complicated stamped. There were too few of
these to warrant comment except that they came from the up-
per two levels.

Calloway Plain (See pottery description of type)

This ware is not described in the literature but has ap-
peared repeatedly on Central Alabama sites included in the
Montgomery Museum survey. In stratified sites, it occurs
consistently in Early Woodland levels. It may be slightly
later than the general run of tetrapodal check stamps but
earlier than Weeden Island and Late Swift Creek. It in-
volves a straight sided, rounded base somewhat thin ware
with either a mica-grit or mica-sand temper. The color is
usually orange-brown or yellow brown. Some are dark grey.
At the Tensaw sites. 77% of all Calloway sherds occurredin
level 1 and all but 3 sherds of it in level 2. A type des-
cription of this ware is included with this paper.

Tensaw Plain (See pottery description of type)

This is also an unpublished type. It involves a light
brown to orange colored sand tempered thin ware. Body seems
to be more or less globular or slightly elongated with some
rim constriction and version to the rim. Lips are rounded
or squared. Some are dowel or twig notched in the manner of
Early Swift Creek rims. A few lips are slightly folded.
Surface is smooth on the interior and exterior with occasion-
al tool marks showing. Temper in nearly all specimens seems
to be fine sand. A very few were impregnated with finely
graded quartz grit grains.

Tensaw Stamped

This ware is essentially Tensaw Plain with the added em-
bellishment of a form of rocker stamping, apparently done
with the edae of a larae flake or shell. The area of decora-
tion appears to be in the shoulder or upper 1/3d of the ves-
sel. The closest known type may be Santa Posa Stamped (Wil-
ley 1949). Both of the Tensaw ceramic types were found main-
ly in Level 1. In grading the pottery by temper. 1,407
plain sand tempered sherds were found in Level 1, 502 in Le-
vel 2 and 59 in Level 3. The bulk of these may have been of
the Tensaw Plain type. Only 21 specimens of Tensaw Stamped
were found: 7 in Level 1 and 14 in Level 2.

Features (See Feature Descriptions)

A total of 17 features were found. Of these 2 were
ruled out as being burned out tree roots (F-2 and F-4). Six
were a form of cracked pebble pile--possibly a form of
hearth. The deepest of these opened 14" from surface. Most
seemed to have been a late manifestation on the site contain-
ing sherds no later than check stamped tetrapodal typesor
fiber tempered ware. One feature contained about 1/3 of a
steatite bowl with lug. Other features were dark stained
pits whose outline could be easily traced. Diameters in-
volved both ovate and circular mouths, most were of the
rounded bottom type. The deepest was 26" from opening (pit
15) but the pebble hearth types did not exceed 10" in thick-
ness. The largest of the latter measured 38" in diameter.


It is felt that the Tensaw Creek Site is unique in that
it gives us a rare glimpse of the cultural continuum from ear-
liest lithic times through earliest ceramic times in Central
Alabama. This multi-component station was probably succes-
sively used as a hunting camp for as long as 8 to 9 thousand
years. Its geographical location on the banks of the an-
cient Alabama River presents a picture of campsite selection
by some of Alabama's first human inhabitants.

It is unfortunate that bone was not preserved on the
site. It would be safe to assume that if conditions lent to
the preservation of bone refuse objects and tools, the arti-
facts count would have been substantially larger.

As a result of this work, we now have a better concept
of lithic and ceramic associations among the hunting-gather-
ing peoples of this area. Plans for resumption of work on
this important site are being made for the summer season of
1966 by the Montgomery Museum and the Montgomery Archeologi-
cal Society.


Cambron. James W. and David C. Hulse
1964 Handbook of Alabama Archeology, Part I Point
Types, Archeological Research AssQciation of Ala-
bama Inc.


David L.. Edward B. Kuriack and James W. Cambron
Stanfield Worlev Bluff Shelter Excavations.
Journal of Alabama Archeology, vol. 8, Nos. 1
and 2.

Griffin, James B.
1943 An Analysis and Interpretation of the Ceramic
Remains From Two Sites Near Buford, S.C. Anthro-
pological Papers No. 22, BAE Bulletin 133, Wash-
inaton. D. C.'


Suhm. Dee

Marion Dunlevy
Guntersville Basin Pottery. Geological Survey c
Alabama, Museum Paper No. 32.

Ann and Alex D. Kriecer
An Introductory Handbook of Texas Archeoloay.
Bulletin of the Texas Archeological Society, Vol.
XXV. Texas Archeological Society. Austin.

Willey, Gordon R.
1949 Archeology of the Florida Gulf Coast, Smithson-
ian Miscellaneous Collections, Vol. 113.


Type Name:
Date: September 1, 1965

PASTE: Method of Manufacture: Coiling
Temper: Sand, frequently micaceous
Texture: Smooth exterior and interior, rarely course
Hardness: About 2.5
Color: Buff to orange, occasionally brown
Core usually black to gray

Scraped or smoothed with pebble. Tooling marks
seen often near rim.


FORM: Rim: Rim slightly everted in some specimens, straight
and almost pinched in others. Small rim foldsin
many specimens
Lip: Rounded or squared (flat) Some rims may be not-
ched in the manner of Early Swift Creek Rims.
Others have pinched nodes spaced about !" apart
beneath lip.
Body: Body seems to be globular possibly semi-co--
noidal with some rim version. Necks are
slightly constricted.
Base: Base is either rounded or semi-conoidal. Small
tetrapods on some specimens but absent in most.

GEOGRAPHICAL RANGE: Unknown. Tensaw Creek site (1 Lo 9)
only site type has been seen
except for a few surface finds
along the Long Swamp Creek
further west of the Tensaw.

CHRONOLOGICAL POSITION: Early Woodland. Possible relation-
ship with Santa Rosa Swift Creek se-
ries. More remotely with Calloway.

POSSIBLE RELATIONSHIPS: Early Swift Creek in Georgia San-
ta Rosa Swift Creek on the Florida
Gulf Coast.



Type Name:

Date: September 1, 1965


PASTE: Method of Manufacture: (All descriptive data other
than decoration is the same as for Tensaw Creek Plain)



Series of curvilinear zig-zag lines produced by
rocking the edge of a sharp flat wood or stone
chip back and forth. These lines of stamped de-
signs seem to be horizontally aligned or paral-
lel with the rim.

FORM: Rim:



POSSIBLE RELATIONSHIPS: Santa Rosa Stamped may be closest re-
lated type geographically.



Type Name: Date: May 16,1964


PASTE: Method of Manufacture: Coiling
Temper: Fine to medium grit with liberal mica inclu-
Texture: Somewhat course and lumpy
Hardness: 3
Color: Usually orange or buff in the typical sherd.
Grey is known.

SURFACE FINISH: Tool smoothed but tooling marks rarely show


FORM: Rim: Straight to lip
Lip: Neatly squared, sometimes slightly extruded and
everted. Extrusion may be inward or outward.
Lip grooves are known.
Body: Not certain, probably elongated beaker shape.
Base: Rounded. Tetrapods apparently not known in
the type.

GEOGRAPHICAL RANGE: Unknown. Several sites found in Mont-
gomery area from Wetumpka to Lowndes

CHRONOLOGICAL POSITION: Known to be associated with Cobb's
Swamp Complex types. Several sing-
le component sites known. Small
hemispherical sand mounds may be as-
sociated with complex.


1 LO 9

TABLE I. Pottery Distribution by Type

Type or Description Level 1 Level 2 Level 3 Level 4 Level 5

Fiber Tempered Plain 6 70 8 1 2

Check Stamped Type 10 15

Calloway Plain 17 18 3

Tensaw Plain & Stamped* 271 549

Grit Tempered Plain 124 52 3

Sand Tempered Plain 1,407 502 59 7

Simple Stamped 1 2

Fabric Impressed (Dunlap?) 6

*Determined on basis of rim form, body configuration or decoration as noted in type description. Con-
ceivably the category 'Sand Tempered Plain' without doubt included a high percentage of Tensaw Plain


1 LO 9

TABLE II. Projectile Point Distribution by Type

Type Level 1 Level 2 Level 3 Level 4 Level 5

Camp Creek 2
Candy Creek 6 2
Baker's Creek 2
Bradley Spike 3
Copena Triangular 5
Mud Creek 7 3 1 1
Hamilton 1 1
Swan Lake 30 10 4
Savannah River 5 5 10 7
Elora 3 10 11 2
Appalachian Stemmed 2
Cotaco Creek 4 1
Gazy 1 4 4
Kays Stemmed 1 1 2
Benton Stemmed 1 6 10 2
Morrow Mountain (type) 1 5 16
Morrow Mt. Straight Base 1 1 2
Kirk Serrated 1 1
Kirk Corner Notched 1 1 1
Big Sandy I 1 4 12 10
Colbert Dalton 2
Greenbrier Dalton 1
Hardaway 1
Stanley 1
Paint Rock Valley 1
Beaver Lake 1

1 LO 9

TABLE III. Stone Tools, Distribution by Type

Type or Description Level 1 Level 2 Level 3 Level 4 Level 5

Ovate Knife 5 6 3 4
Retouched Flake Knife 2 3 1
Lanceolate spall knife 1 1
Stemmed knife 1 1
End Scraper "thumb-nail" 1 2 6 6
Side Scraper 4 1 6 8 5
Plano-convex end scraper 3 1 4 15 11
"Spokeshave" scraper 1 1 1 2 2
"Keeled" or humpback scraper 5
Retouched flake tool 3 6 9 12
Expanded base drill 3
Straight drill 1
Pebble hammerstone 7 22 22 12 10
Mano 3 2 1
Grinding Stone 1 1
"Nut" Stone 1 1
Steatite sherds 2 6 2
Steatite pipe (?) fragment 1
Chert chips & spalls 97 116 85 103 67
"Sugar" Quartz chips 59 123 122 63 32
Quartz (not counted)*
*Quartz or quartzite was the most common material used for stone artifacts. Chips and spalls numbered in
the thousands. It was estimated that approximately fifty quartz or quartzite ships were found for every
chert specimen. "Sugar" quartz is a sandy yellow to gray crystalline quartzite which may have been obtain-
ed from Monroe, Wilcox or Baldwin Counties. Cherts involved local black pebble chert, Fort Payne chert or
Ocala light colored cherts from southeastern Alabama. Only objects considered to be of the greatest arch-
eological significance were included in this table




1. None

Pebble Pile

Burned tree not recorded

Chips, 2 sand t. sherds



OPENS: Level 1
Oval: 35" x 24"
Diam. 8" deep

OPENS: 6"deep
Diam. 19y' x 17"
Depth: 10"




4 (Burned out tap root- no record

Fiber Tempered sherd (one)
Third of steatite bowl
Broken tip of pp

Tetrapod, 4 sherds
Sand t. pl., 1 pebble
hammer, 1 chert chip, 21
quartz chips, 1 galena frag.
1 smashed section of
sand temp plain vessel

Pebble pile

Pebble pile



18" x 24"

24" x 31"



(Large spread of refuse at first mistaken for a pit determined to be merely a concentration
of organic stain not necessairly attributable to Indian occupation) E-810

8. Two sherds, sand temp Plain,
20 sherds, fiber tempered,
6 pebble hammers, 2 chert chips
12 quartz chips, 1 steatite sherd

Pebble Pile

Diam: 29" x 33
Depth: 9"


FEATURES LO 9 (Continued)




(Like Feature 7, this was a very large dark area heavily charged with charcoal and organic
soil whose perimeters ran well beyond excavated areas. Estimated diameter was approximate-
ly 8' 10'. Opening for heavy concentration was a 7"-8" and depth over all was not more
than 7". From this was recovered 140 sherds of sand tempered plain pottery, 6 sherds of
grit tempered plain ware, 3 sherds of fiber tempered ware. One Swan Lake type projectile
also recovered. One fragment of a ground sandstone gorget plus a number of chert and
quartz chips completed the inventory.)

10. No artifacts

11. No artifacts

12. None

13. 2 quartz chips
1 Stemmed lanceolate knife
1 pp, Baker's Creek type

14. None
14. None

pebble pile

Pebble pile


Pebble pile




14" deep
18" x 12"

10" x 14"

Opened 8"
Diam.: 14"







FEATURES LO 9 (Continued)



3 chert chips
13 Quartz chips
2 fiber temp plain sherds
1 chert sidescraper
1 Ovate knife, quartz
Hickory nuts (C-14 sample taken)
Hackberry (?) seeds

15 Quartz chips

2 chert chips
24 quartz chips
1. pebble hammer


slope side
flat bottom




gravel fill

Opened 10"
Diam.: 22"
Depth: 26"




Opened: 9"
Diam.: 19"
Depth: 19"

Opens: 6"
Diam.: 38"
Depth: 30"



1 thru 3: Tensaw Plain rim types. 4-5 Tensaw Stamped 6:
Fiber Tempered Plain, 7-8 Check Stamped type.



Level 1, Left to Right:

Elongate drill, Ovate knife, Stemmed Concave
"Spokeshave" scraper, stemmed end scraper.

Level 2, Left to Right:

Side scraper knife combination, retouched
spall, thumb-nail type end scraper.

Level 3, Left to Right:

Expanded base drill, narrow elongated drill,
Ovate blade, probably knife, laterally trimmed
quartzite spall, chert side scraper.

Level 4, Left to Right:

Large knife, end scraper, end scraper, 'keeled'
end scraper, chert end scraper, small chert
flake delicately flaked along left edge

Level 5, Left to Right:

End scraper, quartz, End scraper, black chert,
Serrated or 'saw-edge' spall, use unknown,
large end scraper, knife.



Left to Right

Level 1: Copena Triangular variant*; Copena Tri-
angular variant, short type; Swan Lake;
Mountain Fork variant; Halifax.

Level 2: All Elora or Elora variants.

Level 3: Savannah River, McIntyre, Savannah Riv-
er Variant, Kays Variant.

Level 4: First four, Big Sandy I; Big Sandy Var-

Level 5: Dalton, Greenbrier, Dalton, Colbert,
Dalton, Greenbrier, Hardaway.

*Variant implies a stylistic departure from
classical published description. Such differ-
ences are slight but enough to remove specimen
from classical category. Writer holds view that
regional variants do exist among known types and
should be regarded as such.



ScA le
I miL&

Figure 1. Map Showing Location of 1 Lo 9

Co u NI

500E 505E 510E 515E 520E 525E 530E 535E
"A" "B" "C" "DF "E" "F" "G" "H"
'A I 1 IE I I G

820N -

S 815 --


805N -

m. Stake No 2

80ON- 8001 -520E



0N .--

I85 _; _____-p--

(1 Lo 9)
780 Contour interval: 1 Ft.

k Pebble pit or hearth
775N- Refuse pit

770N Scale

Mlp Stake No. 1
765N /h 765N-500E
765N ^ r /

DATUM POINT: 760.9'N 500.0'E
( (Nail in slashed pine tree)
760NF e

Figure 11







6" < (, < gD --
6 *

'7- c o ,,HEAVY GRAVELk a

I L09


LEVI 77%

LEV 2 6 t9%

< LEV3 i 4%















I 4II%

I Lo 9 oo00%




LEY 5 B $ Y *


Frank Bushnell

The large village site known locally as the Narvaez
Site or the Jungle Mound is located in the north-west quar-
ter of Section 13, Township 31S, Range 15E, St. Petersburg,
Pinellas County, Florida. This location would more commonly
or locally be given as being on Park Street, North and west
of the Admiral Farragut Academy, on the mainland shore of
Boca Ciega Bay opposite John's Pass. The site is listed in
the University of Florida, Archeological Site Survey files
as Pinellas 54.

The nature of the site is that of a midden, in fairly
good or original condition. It measures approximately 300
yards from north to south and 100 yards from east to west.
The highest elevation above the water is about 12 feet. The
midden does not extend fully to the bay on the west, recent
fill producing an extension of the land into the bay. Approx-
imately one fourth of the site is in a wooded condition
(northern quarter). The remainder is residential. Structur-
ally, the midden is characterized by a series of rises and
ridges of compact shell, rich with artifactual material.
The characteristic flora of the site is a climax forest of
Live Oak, (Quercus virginiana Mill). The midden shell is
predominantly Oyster (Crassostrea virginica Gmelin), Horse
Conch (Fasciolaria gigantea Kiener), Lightning Whelk (Busy-
con perversum Linne), Tulip Shel (Fasciolaria tulipa Linne),
and Scallop (Pecten irradians Lamarck).

During the spring of 1964 attention was drawn to this
midden due to many interesting and typical Safety Harbor
artifacts that were being "pot-holed" from the wooded sec-
tion of the site by neighborhood children (and others).
Many similarities were noted between this site and the
Safety Harbor Site at Maximo Point (Sears, 1958 and Bushnel,
1962). A decision was reached to extend a test trench or
pit into one of the black earth plateaus, which seemed to be
richer in material than the shell ridges. Permission was
granted by Mr. Harold Anderson, a local businessman and
long time resident, whose home is located directly on the
midden. It should be noted that the Andersons, having a
great appreciation of natural Florida, in constructing their
home, left their portion of the midden in the original wood-
ed condition a situation rarely seen today, unfortunately.


After considering various areas of the midden a plateau
in the general south-central area was chosen for testing.
Florida Anthropologist, Vol. XIX, No. 2, June, 1966

The primary reason for picking this location was due to the
presence of this rather definite plateau or steppe. It had
previously been experienced that these flat areas seem to re-
present habitational areas, the actual rides being rather
devoid of cultural material. This situation is commented on
by Sears in his article on the Turner River Site (Sears,

In the zone to be excavated, a square ten feet by ten
feet was marked off. This square was then excavated by care-
ful trowelling, six inches of midden being removed at a time.
All material of importance was then bagged and labeled by
level and was later analyzed in the laboratory.


0 6 Inches. At this depth, the midden soil was
very loose and tended to be granular due to the action of
roots and numerous earthworms. This soil was very rich in
artifactual material, with numerous small sherds of a good
quality Pinellas ware. In addition to these, there were two
Olive Jar sherds, one unfinished (undrilled) barrel-shaped
Columella bead and of great interest, one bird effigy (rim-
effigy) on Pinellas paste. From the body shape and the posi-
tion of the eyes which were bulged in effect it could
possibly have represented the Carolina Paroauet (the beak
was missing). Effigy adornos similar to this one have been
seen in collections from the Maximo Point site. All have
the eyes bulging out to quite an extent, but most seem to re-
present the Great Horned Owl.

Molluscan remains were sparse but fragments of deer,
turtle, and fish bones were very numerous.

At the very bottom of the five inch depth the earth be-
came quite compact and greasy in texture, in what might
possibly have been an old floor level. Sherds were very nu-
merous on both sides of this five inch depth.

Unlike the situation in most local Safety Harbor mid-
dens, flint work was not evident in this upper zone although
it seems to be abundant in other areas of the midden at the
same level.

6-12 Inches. The previously mentioned floor-like layer
persisted for two inches in this second level. No post-
holes were noted but a large fire pit was observed and seve-
ral artifacts were associated with the marginal areas of ash.
These artifacts were as follows:
1. One bone pin (deer bone) undecorated, with a
flat head.
2. One conch Columella plummet.
3. One limestone plummet.

4. One unfinished quartz pebble plummet.
5. Two piles of turtle bone.
6. Two worked sting-ray spine (projectiles).
7. One restorable "pot" of Pinellas Plain.

The fire-pit itself was in the shape of an inverted
cone, eight inches deep and sixteen inches in diameter. Ash-
es were well consumed, the earth having a smooth, clay-like

12-18 Inches. This zone appeared to have no evidence
of stratigraphic layering but seemed to be a rather loose
continuation of the previously mentioned floor. The arti-
facts here seemed to have been trodden in from above.

In this layer were three iron nails, four olive iar.
sherds, two unfinished conch Columella beads (barrel shaped
but undrilled), and one large, transversely broken Strombus
celt. The soil in this zone contained large amounts of
scattered charcoal.

* 18-24 Inches. Loose, black midden earth continued to
twenty inches where another hard, compact floor level was
reached. The upper levels of this zone were very rich in
deer, fish and turtle bone. Another fire-pit observed at
twenty inches, in the south-west quarter of the test pit.
Lying near this fire pit were:
1. One flint cone, elongated and sticking up in
the floor in a vertical position.
2. Two stone (flint) spherical hammers.
3. One sandstone grinding slab,
4. The upper half of a well made deer bone pin.
This also was sticking up as though it had been
intentionally shoved into the dirt.

In the north-east quarter of the test were:
1. A large pile of fragile surf clam shells (Mac-
tra fragilis Gmelin).
2. One large pile of non-reconstructable Pinellas
Plain sherds.
3. One bone pin fragment.
4. One large pile of turtle bone.
5. One projectile point (broken-stem end).
6. One sandstone hone.

S24-30 Inches. The fire-pit observed at a depth of twen-
ty inches extended to the thirty inch level. The most in-
teresting artifact in this level was the missing half of the
previously described bone pin (ref.-the 18-24 inch depths).
This missing half was found pointing straight down into the
soil, from twenty-four to twenty-six inches. Both pieces
fit together perfectly.

At twenty-eight inches was another large mass of Surf
Clam shells, one projectile point or drill of the typical
"Pinellas" shape or type and a battered Busycon shell hammer.

One interesting fragment of what seems to have been a
limestone disk came from the very bottom of the twenty-four
inch level.

30-36 Inches. The thirty inch level seemed to repre-
sent another "floor", as a large fire-pit was at this depth.
Animal bone was quite concentrated but no artifacts other
than a few sherds were found.

One large deposit of Clam shells (Venue campechiensis
Gmelin) was to the north side of this centrally located fire-

The unusually large amount of fresh water turtle bone
in the upper levels is quite interesting in that there was
within recent times an aquatic pond just to the east of the

36-42 Inches. Still another possible floor area with a
fire-pit occurred at 36 inches. Associated near this fire
area was a Busycon bowl with a Busycon hoe nested within the
central cavity. In general, thirty-six to forty-two inches
was a rather bare zone.

42-48 Inches. Nothing of great interest was recovered
from this zone. There were few bones, shells or sherds and
no fire-pits. All sherds seemed to be a rather poor qual--
ity of Pinellas Plain. The soil itself tended to be rather
compact but without stratification.

48-83 Inches. From the forty-eight inch depth to the
bottom of the midden very few details were noticed, with the
exception of a possible floor and fire-pit between forty-
eight and fifty-four inches. Characteristic Pinellas Plain
notched rims did not occur below the fifty-four inch level.
This pit had at its edge a large deer antler tine and a Busy-
con bowl. A compact layer occurred at sixty inches and the
midden refuse became more concentrated as far as organic
staining and shells are concerned. At the sixty-six to sev-
enty-two level, shells became quite compact. These consist-
ed primarily of oyster and scallop shells with a few Busycon
and Strombus alatus. There were very few sherds but those
that were identifiable compared well with the poorer quality
Pinellas Plain sherds that are typical at the Maximo Point

At seventy-two inches, Pockets of shell extended down
into a mixture of midden refuse and white sand, this sand
resting on a hard pan-like sterile layer at the eighty-three
inch depth. (Fig. 1).

/aew/PaN SaNi aV/d M iAfode SoiL Askes SheLL
AM/,/de So;L

Fig. 1. Floor Levels On South Wall Of Test


0-6 6-12 12-18 18-24 24-30 30-36 36-42 42-48 48-54 54-60 60-66 66-72
Pinellas Plain
Body Sherds 200 220 626 220 292 100 130 30 80 34 48 22
Rim Sherds
Round 18 38 16 6 6 2 16 4 8 4 8 4
Rim Sherds
Flat 16 4 26 4 16 4 8 2 2
Rim Sherds
Flat Notched 2 10 40 18 16 12 8 2 2
Pinellas Paste
Rim Bird Effigy 1
Jefferson Ware 2
Sand Temp. Plain 2
Unique Pinellas
Incised 2 2
Pinellas Strap
Handle 2
Pinellas Round
Rim Punctated 2
Pinellas Incised 2 4 2
St. John's Incised 2
St. John's Check
Stamped 3
Pinellas Plain,
Grooved Flat Rim 2
Spanish Olive
Jar Sherds 2 4
Iron Nails 3


0-6 6-12 12-18 18-24 24-30 30-36 36-42 42-48 48-54 54-60 60-66 66-72
Fish Hemal Bones 6 16 18 16 20 4 20 4 8 6 16 14
Antler Fragments 2 2
Deer Bone Frags. 6 12 22 8 24 14 12 2 8 6 2
Turtle Shell Frags. 12 22 24 43 28 14 6 4 4 6
Deer Phalanges 2 4 2 2 3
Turtle Bone 8 56 22 80 84 36 18 6 12 12 4
Stone Crab Claws 4 5 8 2 3
Deer Vertebra 2
Fish Spines 2 14 10 16 10 4 8 18 2 1
Busycon Columella
Chisel or Gouge 1
Fish Vertebra
Undrilled 4 6 12 8 29 13 2 5 10 5 2 3
Sting Ray Spine
Projectile Points 2
Bone Pins 1 2 1
Alligator Vert. 1 1 1 1
Ark Shells
Perforated 2 1 1
Strombus Celt 1
Opossum Jaw
(Mandible) 1
Deer Molar 1
Dear Astragulus 1 1 2
Racoon Ulna 1
Busycon Hammer 1
Olive Bead 1
Busycon Hoe 1
Busycon Bowl 1
Busycon Columella
Beads (Barrel) 1 2
Busycon Columella
Plummet 1
Antler Tine 1


0-6 6-12 12-18 18-24 24-30 30-36 36-42 42-48 48-54 54-60 60-66 66-72
Sand Stone Frag. 3 1 2 1 4 1 1
Flint Chips 3 1 3 2 6 1 1
Utilized Flakes 2
Sand Stone Hones 1 1 1 5 1
Steatite Lump 1
Quartz Pebble
Plummet 1
Flint Core 1
Fossil Manatee
Rib. Frag. 1
Limestone Disk
Fragment 1
Coral-Unworked 2
Limestone Plummet 1
Flint Hammer
Stone (Ball) 2
Pinellas Type
Projectile Point 1 1


The Narveez Midden would seem at first glance (consider-
ing that only the one test was made) to represent a rather
typical midden of the Safety Harbor period. The presence of
notched rims throughout most of the tests would tend to veri-
fy this conclusion.

It is interesting to speculate that the upper two feet,
having Spanish material and a very good grade of native cera-
mics would represent a time phase or period missing at the
Maximo Point site, while the lower levels at Narvaez, having
a poorer grade ware come closer to matching Maximo both in
quality of material and in time. This speculation would
tend to corroborate the opinion held by the author and other
local amateurs that Maximo was constructed early in the evo-
lution of the Safety Harbor period the presence of "Weedon
.Island-like" sherds at the bottom of that site being additi-
onal evidence (one Majolica sherd that came from the surface
at Maximo is here being considered as not being sufficient
evidence to change the author's opinion dealing with the age
of that site this sherd is of a type that would normally
be associated with the early 1800's).

It is probable that the Narvaez Midden and the "type
site" at Phillipi Park, Safety Harbor, Florida are of the
same age. The Narvaez Midden might possibly. be associated
with a slightly longer time span if one were to consider the
depth of the deposit alone.

The Narvaez Midden does not at the present have an asso-
ciated burial mound. It is reported by Mr. Anderson, how-
ever, that a cemetery existed just to the north of the site
but was destroyed by years of road building and real estate
developing. It should be noted that off shore and to the
west of the midden was the so-called "John's Pass Mound"- a
burial mound of the Safety Harbor period now covered over by
land-fill (Moore, 1903; Ostrander, 1960).

Finally, I wish to express my thanks to the Andersons
for permission to test and to April Caldwell and Bud Spence
for their assistance.


Bushnell, Frank F.
1962 The Maximo Point Site 1962. The Florida An-
thropologist. Vol. XV. No. 4, December, 1962.
Moore, Clarence B.
1903 Certain Aboriginal Mounds of the Florida Cent-
ral West Coast. Journal of the Academy of Natu-
ral Sciences of Philadelp Ea, Vol. 12, 7 1T3

Morris, Percy A.
1958 A Field Guide to the Shells of Our Atlantic and
Gulf Coast. Houghton Mifflin Co., Boston, 1958

Ostrander, Ozzie
1960 The John's Pass Mound., The Florida Anthropolo-
gist., Vol. XIII, Nos. 2-3, September, 1960.

Sears, William H.
1956 The Turner River Site, Collier County, Florida.
The Florida Anthropologist, Vol. IX, No. 2,
June, 1956.

1958 The Maximo Point Site., The Florida Anthropolo-
gist, Vol. II, No. I, February, 1958.

Biology Department
St. Petersburg Junior College

Back Issues

Volumes 15 through 18 may be purchased from the Treasurer of
the Society at $4.00 per volume.

Volumes 1 through 14 may be purchased from Johnson Reprint
Corporation, 111 Fifth Avenue, New York, N. Y. 10003.
=Prices are as follows:
Vols. 1, 5, 10 $6.00 per volume
Vols. 2-4, 6-9, 1-14 $8.00 per volume
The Florida Anthropological Society Publications nos. 1-5 are
also available from Johnson Reprint Corporation.
No. 1 & 3 $2.00
No. 2 2.50
No. 4 4.50
No. 5 3.00

University of Florida Home Page
© 2004 - 2010 University of Florida George A. Smathers Libraries.
All rights reserved.

Acceptable Use, Copyright, and Disclaimer Statement
Last updated October 10, 2010 - - mvs