The Florida anthropologist

Material Information

The Florida anthropologist
Abbreviated Title:
Fla. anthropol.
Florida Anthropological Society
Place of Publication:
Florida Anthropological Society
Publication Date:
Quarterly[<Mar. 1975- >]
Two no. a year[ FORMER 1948-]
Physical Description:
v. : ill. ; 24 cm.


Subjects / Keywords:
Indians of North America -- Antiquities -- Periodicals -- Florida ( lcsh )
Antiquities -- Periodicals -- Florida ( lcsh )
serial ( sobekcm )
periodical ( marcgt )


Contains papers of the Annual Conference on Historic Site Archeology.
Dates or Sequential Designation:
v. 1- May 1948-

Record Information

Source Institution:
University of Florida
Holding Location:
Department of Special Collections and Area Studies, George A. Smathers Libraries, University of Florida
Rights Management:
All applicable rights reserved by the source institution and holding location.
Resource Identifier:
01569447 ( OCLC )
56028409 ( LCCN )
0015-3893 ( ISSN )


This item has the following downloads:

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.




MARCH, 1965


a publication of the florida anthropological society

Volume XVIII, No. 1

March, 1965

The Marco Midden, Marco Island, Florida
John C. and Linda M. Van Beck
Animal bones from Marco Island Elizabeth Wing
A Dalton Complex from Florida
Lyman 0. Warren and Ripley P. Bullen
A 17th Century Wreck off Cape Canaveral C. B. Harnett
Land Subsidence on the Gulf Coast William C. Lazarus
Atypical Man in South Florida John and Eunice Williams
Proposed Antiquities Law C. H. Fairbanks

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

1st V. Pres.:
2nd V. Pres.:

Officers of the Society 1965
Charles W. Arnade, University of South Florid
Roger T. Grange, University of South Florid:
James W. Covington, University of Tampa
J. Floyd Monk 1960 SW 61st Court Miami 5-
David S. Phelps, Florida State Univertit'
Charles H. Fairbanks, University of Flo':idi

Executive Committeemen 1965
Carl A. Benson, 2310 Resthaven Drive, Orlandi
James A. Ford, Florida State Museum, Gainesvilli
Cliff E. Mattox, P. 0. Box 521, Cocoa Bead
William H. Sears, Florida Atlantic University, Boca Ratoi
Charleton W. Tebeau, University of Miami, Coral Gable


John C. and Linda M. Van Beck

The Marco Midden is located on the northern end of Mar-
co Island, the largest of the 10,000 islands which extend a-
round the southwesternmost coast of Florida. The central
part of the key is arid, with rolling hills covered with typ-
ical scrub oak and palmetto. Most of the inner coastal area
is covered with dense tropical vegetation and bordered by
mud flats and mangrove swamps.

The only available records of previous archeological ac-
tivity in the immediate site area are those of Frank H. Cush-
ing in "The Court of the Pile Dweller" (Cushing 1897). The
area has also been inspected by Clarence B. Moore (Moore
1900) and A. Hrdlicka (Hrdlicka 1922). The Marco Midden was
not mentioned by these early workers although it is only ap-
proximately 850 feet east of the Cushing site. The sites
are separated by a narrow inlet. Other sites on Marco Is-
land that have been the scene of considerable archeological
activity are the Caxambas site, 4 3/4 miles to the south,
and Goodland Point site 6 3/4 miles to the southeast (Goggin

The mound is located on the northeast end of the key in
the center of the settlement known as Marco, which is actu-
ally part of Collier City. It is in Block 13 and centered
over Lot 38. This places the southern edge of the site ap-
proximately 220 feet north of the Marco Post Office, and im-
mediately adjacent to the west side of State Route 951. As
it exists today, the midden is a semi-circle or arc rising
to a height of 14 feet and measuring 235 feet north to south,
61 feet east to west. Road building construction many years
ago completely removed probably more than one half of the
eastern portion of the midden, leaving the western end un-
disturbed. The profile of this western end slopes without
the leveling that one would expect at the summit. This seems
to indicate that only the edge of the mound now exists. Thus
there is reason to believe that the midden may have extended
to Marco Pass on the east. The midden is presently covered
with a few large trees, young saplings, heavy vine, and soft
tropical growth.


Three test pits were dug in six inch levels and screened
with hardware cloth.
Test A- Located 60 feet south of the northern boundary dirt

Florida Anthropologist, Vol. XVIII, No. 1, March 1965 1


rl --7



road and 38 feet west of S.R.951. This pit was 3x5 feet and
was excavated to a depth of 90 inches where a solid layer of
large busycon shells were encountered. Due to a shortage of
time, the pit walls had been allowed to slope inward and the
bone and shell material could not be fully salvaged.
Test B- Located north and immediately adjacent to Pit A.
Pit size of 5x5 feet was maintained to a depth of 120 inches.
Sterile beach sand was found at 116 inches.
Test C- A 3x3 pit located 30 feet west of S.R.951 and approx-
imately 105 feet from the northern boundary. Excavated to a
depth of 3 feet, the pit yielded only pure oyster shell.

Test A and B had a similar stratigraphic structure of
layered brown sand, fire beds, crushed bone, and white ash.
These layers varied from 14 to 4 inches in thickness. An
area of hard grey clay in which bone and pottery were em-
bedded appeared in Level 11 and continued through part of
Level 13. There was a 4 inch layer of wet brownish clay im-
mediately above the sterile sand at the bottom of Pit B.

The percentage of shell to total material volume by
level is as follows: Levels 1 through 8, 5%; Level 9, 15%;
Level 10, 10%; Levels 11 through 13, 5%; Level 14, 2%; Level
15, Pit A, solid Busycon (Pit A terminates), Pit B, 5%; Lev-
els 16 through 18, 2%; Levels 19 and 20, none. Ceramic ma-
terials were distributed without respect to specific compo-
sition of midden levels.

Food Remains

The animal remains indicate a fairly stable ratio of
shellfish to vertebrate food sources throughout the habita-
tion period. Shellfish were probably not as important a
source of food as vertebrates. This was indicated by the
relative quantities of shell and bone deposits in the midden.
Dr. E. S. Wing's analysis of the faunal remains follows this
report. Pits A and B revealed a total of 26 species of ma-
rine mollusks which are still present in the area. The Ab-
bott Classification has been used (Abbott, 1954).

Artifacts: Ceramic

The bases of levels in Pits A and B were contiguous and
in the same plane through Level 15 where Pit A terminated.
Inasmuch as the pits were immediately adjacent and of simi-
lar stratigraphy, the cultural material has been combined
for discussion purposes throughout the remainder of the re-
port. 3

Table I. Shell Species

Barbatia candida
Barnea costata
Crepidula fornicata
Anomolocardia cuneimeris
Spissula solidissima simelis
Noetia ponderosa
Atrina seminuda
Cardita floridana
Vermicularia Spirata
Conus stimpsoni
Mercenaria campechienois
Macrocallista nimbosa
Pleuroploca gigantea

Littorina angulifera
Anomia simplex
Murex pomum
Fasciolaria hunteria
Aequipectin gibbus
Polinices duplicate
Cantharus tinctus
Balanus Spirata
Dinocardium robustum
Crassostrea virginica
Busycon contrarium
Strombus alatus
Oliva sayana

Most of the sherds are clearly representative specimens
of well known pottery types. Some of the Glades Plain sherds
from the bases of the vessels exceeded the range of thick-
ness listed by Goggin as 0.3 to 1.1 centimeters (Goggin 1949:
35). The average thickness of bottom sherds was 1.1 and
reached a maximum of 1.5 centimeters. A high percentage of
the bottom sherds had caked food remains on them. Strap
holes averaging 0.4 centimeters in diameter were found on 7
Glades Plain sherds.
Goodland Pottery. Sherds are of a buff or cream color on
the interior wall with many having a buff exterior also.
All sherds have a black core with the exception of very thin
ware and thin rim sherds, which are buff throughout. The
past description of Goggin is applicable (Goggin 1949:76-7).
None of the sherds show tool markings. Vessel forms are sim-
ilar to Glades types with thickness range of 0.2 to 0.3 cen-
timeters at the rim and 0.5 to 0.8 centimeters at the base.

Sherds classified as Goodland Red had dry pigment bur-
nished into the nearly dry vessel surface before firing.
The resulting surface is almost glass-smooth. This form of
ceramic finish is not durable; erosion and pitting will oc-
cur rather rapidly unless the ware is protected. While at-
tempting to determine the Goodland vessel form, it was found
that some sherds classified as "red" and "plain" were conti-
guous in the same vessel. Original classifications of Good-
land Red and Goodland Plain have been allowed to stand al-
though it is probable that many sherds classified as plain
are eroded Goodland Red. There is question as to whether
the terms "plain" and "red" are meaningful other than for
use in describing the current condition of Goodland type

Sanibel Incised Variations. There were 40 sherds in Levels

Table II. Distribution of Pottery by Type, Pit A.

1.1 M I V2 3 E-M

p a H V2 QH
a H 0 Z in
m 1- F-4 02 (x I-I .4 0 E
<< Q M < 3

0 O -40 014
xw M t z 0 00 0 O C.
O i1 El | -lO l 0 0 a Z 6 0
s~~~~ o tawSgeit lB 6

14 161 6 5 1 173 193
13 3 4 0 20

21 1 22
15 02 1 1 23

(4)1916 0 8 10 0 41 23 3 8 0 0 7 2016
(1) 131 2 0 0 2 9 10 0 6 18 1 2 181
- u --

14 181 1! 9


(512047 2 8 10 2 50 15 3



13 and 14 from one vessel. Thirty-three of these sherds
were glued together to form part of its rim and side. The
resulting design shows 3 rows of closely spaced ticking that
form a running V design on the rim and shoulder (Fig.2,A).

Five sherds from Level 18 show closely spaced rows of
ticking (Fig.2,C). Goggin stated that this design was re-
lated to Sanibel Incised (Goggin 1950:Fig. 19,M).

Gordons Pass Incised. Most of the geometric patterns found
at Goodland Point (Goggin 1950:Fig. 19, A-K,P) were also
found at this site.

A new pattern found in Levels 11 and 14 has rows of in-
cised lines spaced approximately 2.3 centimeters apart, per-
pendicular to the rim. Rows of deep dot punctations paral-
lel alternate lines. Gordons Pass ticking parallels the re-
maining lines (Fig. 2, F).

One sherd found at Level 10 has a row of ticking that
looks like a row of dashes parallel to the incised lines
(Fig. 2, D).

Unclassified. Of particular interest is a sherd from
Level 19 that shows a combination of Gordons Pass and Sani-
bel Incised features (Fig. 2, G).

Englewood-like Incised. Nine sherds (Fig. 2, H) from
Level 9 have a soft paste tempered with fine sand. Decora-
tion consists of lines forming V shapes filled with rows of
punctations similar in execution to Englewood Incised. Sears
found similar sherds at the Turner River site and indicated
that they did not have cultural or chronological signifi-
cance there (Sears 1956:55).

Unclassified Incised. Most of the sherds have one to three
lines of incision and are of Glades paste. Two sherds in
Levels 9 and 14 have a V shaped design (Fig. 2, I) and other
exceptions are illustrated in Figures 2 and 3.

One variant from Level 15 is probably related to Mate-
cumbe Incised (Fig. 2, J).

Another variant found in Levels 3 and 5 is probably Key
Largo Incised type found in the Northern Fort Walton Period
(Fig. 3, D).

From Level 9 came another sherd that has a clearly
spaced, running design of chevrons which are approximately
1.2 centimeters wide (Fig. 2, K).

0 0 00


J2 7 KIZ L
Figure 2. Pottery
A-C Sanibel Incised. D-F Gordons Pass. G. Unclassified. H. Englewood-like
Incised. I-L. Unclassified Incised





Figure 3. Pottery
A-D. Unolasified Incised E,F,K. Unclassified Punctate G-J. Fort Walton Influenced

Table III. Distribution of Pottery by Type, Pit BR

la 2 i .

BY] | ela! a l(@lele ellg | Biles | ae
aiel 4 a HIATT Ja Ble| 2 2I5;/ 4) BIT sl.
< af] slelalalg Oo] & Loe. ote a ae
a iF] & je) BIS] Sele /8 elelalal@lEle &) 3

a lH] oO a|=]ua 3/2/32 a Pils SEA aed! «
a | Bm | ais 9/s;e/9)8 3/|3/8 |} ef HAT AT &
io} a 4 3 ata alel|id/d] 2] 2
8| 3/2! celal gialelgial 8] 8lel ale] eiele) 81] é
2] sib] BIB] BIS/EIBls] S|] SIE |S] a] 2128/8] eae] &


wo wo Wo no










nny ru




~ feb lelelefelelelelelabarlalelebalaPal= lo"
( 26 12 105.
noru o]sose | 2 |g |s [se fs [> [> [avs [ass | [1 [o> [2s [ee Joose [207] sms

yVN Pr FO Wu No OF

Unclassified Punctate. All of the sherd are of Glades paste.
Two sherds in Levels 9 and 14 have a single row of puncta-
tions and one sherd in Level 9 has a semi-circle of puncta-
tions (Fig. 3, F).

A sherd from Level 15 may be similar in design to one
listed by Goggin (Goggin and Sommer 1949:P1. 2,L). There is
a single row of small raised bumps parallel to the rim.
Each bump has a pin-hole size punctation in its center and
there is also a hole near the bottom of the sherd. The ves-
sel may be a shallow saucer with a hole in its base (Fig. 3,

Fort Walton Influenced. Five sherds from Levels 1,5,6 and 9
show Fort Walton influence in their shape and punctation de-
signs. The paste is typically Glades and the vessels were
probably locally made (Fig. 3, G-J).

Descriptive Categories of Worked Rims. None of the rim
sherds listed in Table IV show any thickening or clay build-
up at the rim. The designs which are on Glades Plain ware
seem continuous. The Type D fluted rims (Fig. 4, D) have
been mentioned by Sears (Sears 1956:Fig. 6,C) and Willey
(Willey 1949:Pl. 14, A).

The grooved flat lips, Type C (Fig. 4, C) are similar
to two found by Willey (1949:Pl. 14, A, and P1. 3, D). Types
A and B (Fig. 4, A and B) all have fine incising at the rim
which is similar to the Fort Walton type of marking. Type E
(Fig. 4, E) is mentioned by Goggin (1950:245). Type F (Fig.
4, F) has fine grooves on the outer surface of the vessel
perpendicular to the rim. The grooves in Type G are on the
edge of flat lipped vessels. Starting at the inner lip edge,
they cut deeper as they reach the outer surface (Fig. 4, G).
These grooves were also found of the Matecumbe Incised Vari-
ant (Fig. 2, J) which is not included in Table II.

Artifacts: Nonceramic

Shell Tools. The range of shell tools (Tables V and VI)
fell mainly into forms listed by Goggin (1950:77-81). There
were a large number of notched shell weights and perforated
Notia ponderosa (Arca) which Cushing found to be used as net
weights (Cushing 1897:366). One perforated Crassostrea vir-
ginica Gmelin (Oyster) was found at Level 7, and two perfora-
ted Aequipectin gibbus Linne (Scallop) at Levels 3 and 5
(Fig. 5, L,M). Two awls were found in Levels 1 and 3 (Fig.
5, F,I). A shell hone with well worn surface (Fig. 5, J)
and a shell pin fragment (Fig. 5, K) were found in Level 13.
Two Pleuroploca gigantea short hammers had three hafting



Figure k. Worked Rims
A. Lines on top of rim. B. Lines on top and side. C. Grooved, flat lips.
D. Fluted. E. Notched. F. Groves perpendicular to rim. G. Grooves on lip edge






Figure 5. Stone and Bone Objects
A-E. Limestone Plummts. F,I. Shall Awa. 0,H. Bone Pina J. Shell Hoe.
K. Shell Pin Fragment. L. Perforated Scallop. M. Perforated Oyater


Table IV. Worked Rims from Combined Pits *

A Lines on top of rim Levels 6(3), 7(2) 5
B Lines on top and side Levels 2(3), 3(1),
15(1) 5
C Grooved, flat lips Levels 1(2), 2(2), 4(1),
701), 803), 13@ _ 11
D Fluted rims Levels 10(1), 13(1) 2
E Notched Levels 1(1), 2(3), 33),
F Grooves perpendicular to rim Levels 18(2), 19(1) 3
G Grooves on lip edge Levels 2(1), 3(1), 4(2),
6(2), 15(1) 7
*Number of specimens in brackets Total 41

Bone. Several perforated fish vertebra beads were found in
Levels 3 through 15. In Level 12 there was a set of five
matching vertebrae that were finely grooved, A broken bone
pin of unidentifiable long bone was found in Level 15 (Fig.
5, H) and a pin fashioned from the tibiotarsus of Casmerodi-
us albus (American egret) was found in Level 11 (Fig. 5, G).

Stone. The five limestone plummets found in Levels 2, 3,
and 7 are typical in configuration to those of the Glades
Area, and are roughly shaped (Fig. 5, A-E). With the excep-
tion of two limestone grinders, the bulk of the remaining
worked stone is not native to the area. In general, these
stones are irregularly shaped with sharp edges, and were
most likely used for cutting, scraping, etc.


As the pottery from Cushing's site has not been ana-
lized, the nearest site one can compare this midden with is
Goodland Point. The general distribution and types of pot-
tery are the same, with Glades Plain being the most numerous
type. The percentage of Goodland type ware at Goodland
Point was 11%, and at the, Marco Midden it was 5%. The peak
was reached in Level 14 and Goggin found the peak at Good-
land Point to be in the Glades Ila and IIb Period (Goggin

Period markers were found in small quantities at both
sites. Biscayne Stamped was not found at Marco. On the
other hand, 50 soft paste, sandy sherds were found in Levels
1-3 and classified as St. Johns, At the Turner River site,
St. Johns Plain was located in the Glades IIa Period (Sears
1956:53-58). Belle Glade Plain was found in Levels 1-9 and
has been placed in the Glades IIc-IIIa Period (Goggin 1949:
87). Surfside Incised, a Glades IIIa marker was found in
Levels 3 and 5. Two sherds which are probably related to


Table V. Distribution of Shell Toola, Pit A.

Levels 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 Total












2 2

1 1 1


1 2


1 12

1 1

Total 4 6 4 3 1 1 6 1 3 5 1 3 6 1 45

Table VI. Distribution of shell Tools, Pit B.

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 Total



















4 1 5 3 2

4 2 8 1

3 5 10 1 3

2 1 2

2 1 1

1 1




1 1 1 2

6 2 3 11

3 1 4 10 4

5 2 4 2

3 7 2


3 1



2 11


1 1 1 1

5 5 33 5 2 1

3 7 3 1


1 1 2

2 2


10 3 2


2 7 2 4 2 5 2 1 3 6 1 16 7 6




3 4

1 1


1 65


Total 13 32 42 21 19 22 23 18 36 40 22 14 47 49 14 3 1 1 417

. ..

the Key Largo Incised type were also in Levels 3 and 5. Key
Largo is considered a Glades IIb marker. The Englewood-like
incised in Level 9 was found in conjunction with Key Largo
Incised at the Turner River site (Sears 1956:55-59) and in
general Glades II Period in Miami (Laxson 1962:6). Miami In-
cised, a Glades la marker was found in Levels 1-9.

Gordons Pass Incised (Glades IIa-IIb) began at Level 9,
reached its peak in Level 14 and continued to Level 19. San-
ibel Incised (Glades IIa-IIb) began at Level 11, reached its
peak in Level 13, ard continued to Level 18. A sherd combin-
ing both of these types was found in Level 19. One sherd of
Fort Drum Incised was found at Level 17. It has been rela-
ted that Glades II at Gordons Pass (Goggin 1939) and to ear-
ly Glades IIa (Goggin 1950:240).

It should be noted that many of the sherds listed as
unclassified wares have been found in the Glades area when a
large sample was available. As yet, there are not enough ex-
amples of these sherds to indicate any definite classifica-
tion categories.

None of the previous samples in the Glades area re-
vealed the variety or number of unclassified worked rims
that were found at Marco. Seven forms of execution were no-
ted and categorized for discussion purposes (Table IV).
Type C, with grooved, flat lips, (Levels 1-13) comes closest
to resembling Glades Tooled. The notched rim, Type E, which
was found in Levels 1-5, does not appear to be Pinellas
Notched. Type E may be also considered as a possible fore-
runner of Glades Tooled which is found in the Glades III Per-
iod. Goggin found a similar type first appearing in late
Glades IIb (1950:245). The similarity of Types A, B, and F
to Fort Walton Safety Harbor is a factor of interest. The
worked rims are distributed throughout the midden.

The Pleuroploca gigantea long hammers were found in
Levels 1-7. Goggin indicated that the long hammer form was
relatively late and placed it in the Glades IIc-IIIa Period.
The other tools (Tables V and VI) were distributed randomly.

The Marco site is a single sand midden in close proxim-
ity to the famous Cushing Site. The cultural materials indi-
cate that this site is primarily of a Glades IIa-IIb culture
in Levels 9-19 that expanded into Glades IIIa through a
transitional period in Levels 1-8.

This transitional period is marked by the presence of
Belle Glade Plain (Glades IIc-IIIa). There is an absence of

Table VII. Distribution of Stone Materials.


Levels 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 Total

UNWORKED STONE 1 1 2 1 3 2 1 11





Total 1 1 4 3 1 3 2 1 16


Lavel. 1 3 4 5 6 7 8 9 10 11 12 13 14 15 Total

WORKED STONE 1 1 3 1 5 5 2 3 2 23

UNWORKED STONE 4 1 2 1 1 1 2 1 4 1 2 20



CORAL 1 2 1 4


Total 2 8 5 5 1 7 1 1 7 3 7 3 2 52

Glades Tooled (Glades IIIa marker) and of Gordons Pass In-
cised and Sanibel Incised (Glades IIa, IIb markers). On the
other hand, Surfside Incised (Glades IIIa) Miami Incised
(Glades IIa), Key Largo Incised (Glades II) and St. Johns
(Glades II) were present. The high percentage of Glades
Plain tends to support the theory that Levels 1-8 represent
a transitional period which might be termed Glades IIc.

The midden material, viewed stratigraphically, suggests
a fairly stable population and continuous occupation. There
is no evidence of chronological movement of habitation area
toward water or in concentric circles such as has been noted
at a number of sites on keys or near large bodies of water
in the Glades area. Following Goggin's sequence for the
Glades area, this midden was probably occupied between A.D.
25 and 1500.

The relationship of the Marco Midden to the Cushing
Site across the inlet can only be surmised at this time.
Whether they were populated by the same people at roughly
the same time or whether the midden was occupied first, (the
Cushing Site has been tentatively classified as IIIb), are
conjectures. It is hoped that further analysis of the Cush-
ing Site material will shed some light on this issue.

The authors wish to express their thanks to Dr. William
H. Sears, Dr. Elizabeth S. Wing, Dr. Charles H. Fairbanks,
Dr. Towner B. Root, Dr. John M. Goggin, Mr. Frederick W.
Sleight, and Mrs. Lowell L. Lotspeich.


Abbott, Robert Tucker
1954 American Seashells. Van Nostrand, New York.

Cushing, Frank H.
1897 Exploration of
Gulf Coast of

Ancient Key Dwellers's Remains on
Florida. Proceedings, American
Society, Vol. 35, pp. 329-448.

Goggin, John M.
1939 A Ceramic Sequence in South Florida. New Mexico
Anthropologist, Vol. 3, pp. 36-40. Albuquerque.

1949 Cultural Occupation of Goodland Point. The Flo-
rida Anthropologist, Vol. II, Nos. 3-4, pp. 65-
91. Gainesville.

1950 Stratigraphic Tests in the Everglades National
Park. American Antiquity, Vol. XV, No. 3, pp.
228-46. Menasha.

Goggin, John M. and Frank H. Sommer III
1949 Excavations on Upper Matecumbe Key, Florida.
Yale University Publications in Anthropology, No
41. New Haven.'

Hrdlicka, A.
1922 Anthropology of Florida. Publications of the
Florida State Historical Society, DeLand.

Laxson, D.

Excavations in Dade and Broward Counties. The
Florida Anthropologist, Vol. XV, No. 1, pp. 1 -
10. Gainesville.

Moore, Clarence B.
1905 Miscellaneous Investigations in Florida. Journ-
al of the Academy of Natural Sciences of Phila-
delphia, Vol. XIII, Pt. 2, pp. 299-325.

1907 Notes on the Ten Thousand Islands, Florida.
Journal of the Academy of Natural Sciences of
Philadelphia, Vol. XIII, Pt. 4, pp. 458-470.

Sears, William H.
1956 The Turner River Site, Collier County, Florida.
The Florida Anthropologist, Vol. IV, No. 2, pp.
47-60. Gainesville.

Willey, Gordon R.
1949 Excavations in Southeast Florida.
sity Publication in Anthropology,
University Press, New Haven.

Yale Univer-
No. 41, Yale


Elizabeth Wing

Marco Island, one of the Ten Thousand Islands on Flori-
da's southwest Gulf coast, gained notoriety at the turn of
the century for a remarkable discovery of Indian artifacts
by Dr. Frank H. Cushing (1896). This site is unique in the
southeast for its preservation of organic materials such as
wood and fiber. Since these materials were salvaged in 1896
other excavations have been made. One of particular inter-
est is within a few hundred yards of Gushing's site and was
excavated in 1963 by John and Linda Van Beck. These two
sites differ in the culture represented and methods of exca-
vation therefore caution must be used in their comparison
but complementary data gained from both sites provides infor-
mation on the food economy of the early inhabitants of this
arid island environment.

All the materials collected by Cushing were preserved
in muck. They were salvaged from the muck but their strati-
graphic position was not maintained. I believe there was
some selection of at least the bone saved because of the un-
usual proportion of skeletal elements as for example far
more fish spines than vertebrae whereas the reverse propor-
tion is typical. Furthermore a small portion of the remains
are not available for study. The skeletal remains are of
animals that were undoubtedly used for food. Some of these
bones and teeth have been shaped into tools and ornaments.

The materials excavated by the Van Becks came from two
pits which were extended 14 and 20 six-inch levels into a
sand mound. These materials, consisting of pottery, shell,
and animal remains, were maintained in their stratigraphic
position for study.

The animal species that have been identified and the
relative abundance of each are listed on Table 1. All the
species represented occur in this area now. No significant
changes are evident in the different levels of the two pits
or differences between the two pits of the Van Beck site,
however, the remains found at the two sites are different in
many ways. More mammalian and bird species are represented
in the Cushing collection whereas the fish remains are more
abundant and varied among the Van Beck's material.

As was mentioned caution must be used in comparing the
faunal remains from the two sites because of the differences
Florida Anthropologist, Vol. XVIII, No. 1, March 1965 21

in methods of excavation but some of the remains are worth
discussing in detail.

Mammalian Remains
Among the mammalian remains five individual dogs (Canis
familiaris) are represented in Cushing's material but none
were found by the Van Beck's. Four of these dogs were small,
their measurements corresponding to those from an Alabama
shell heap (Haag, 1948). The remains of one individual were
substantially larger than the small dog remains. Houch
(1951) reports abundant dog remains from South Indian Field
confined to the lower levels of the excavation corresponding
to a very early occupation (Orange Period). She suggests
that these dogs were eaten as many of the individuals were
juveniles. She indicates that these were the small Indian
dog (Allen 1920) and the measurements of the one jaw that
she gives correspond to those of the small dog from Cush-
ing's site and the small dog from Alabama (Haag 1948) (Ta-
ble 2). With the remains of the small dog were a few frag-
ments of a larger dog which she suggests may represent the
large extreme of the small dog size range. In both cases
the data are inconclusive but Haag has indicated (1948) that
two sized dogs were not contemporaneous at any one site and
that the small dog was associated with older cultural hori-

Deer (Odocoileus virginianus) remains are of greatest
interest in regard to the many uses made of their bones. Al-
though worked teeth and bone from the Cushing site will be
the subject of further research a brief mention of the types
of bone used will be made in this report.

Little of the bone from the Van Beck's excavation was
worked but much of that from Cushing's collection was. In
his collection the bones most commonly used are those of
deer and turtle. Of the thirty worked or partially worked
deer bones; eight antler fragments were modified to varying
extents three of which were completed elements of composite
tools; one tibia was smoothed and polished to a point dis-
tally; two phalanges were drilled with a hole laterally
through the proximal end; nineteen metapodals were modified
to varying extents, which generally involved a lateral
split, of those completed one was spatual-like and the
others were pointed. In addition there are eleven needle-
like objects that are too modified to positively identify
the species and element they were made of but I suspect they
are also made of deer metapodals.

Bird Remains
A far greater variety of birds is represented in the

Cushing collection and these are predominantly shore birds.

Reptilian Remains
Of reptilian remains turtle is abundantly represented
at both sites. Turtle shell from the Cushing site is worked
in basically two ways. Several diamond-back terrapin (Mal-
aclemmys terrapin) carapaces and one chicken-turtle (Deiro-
chelys reticularia) carapace with plastrons removed were
smoothed on the inside and holes were drilled in pairs along
the marginals. The other type of worked turtly shell was
cut, smoothed, and polished squares or rectangles on the av-
erage of about 1 3/4" by 2 1/4". Seventeen of these were
made from the posterior portion of the plastron of box tur-
tle (Terrapene carolina), three were made from the anterior
portion of the plastron of diamond-back terrapin, seven were
made from sea turtle shell (Cheloniidae), and eight were
made from soft-shelled turtle (Trionyx ferox) shell.

Remains of alligator were sparce at both sites. Alli-
gator bone fashioned into four pins are part of the Cushing

Fish Remains
Fish remains were most abundantly represented at the
Van Beck's site, constituting over 75 percent of the total
animal remains. Although fishes constitute only 32 percent
of the total animal remains at the Cushing site selection by
the archeologist probably has taken place so that this is
not an accurate indication of the percentage of fish at the
site. The fishes represented are predominantly bottom feed-
ing species that live inshore. Mullet (Mugil sp.), repre-
sented at Cushing's site by 6 skulls, has not been found in
Florida middens before although it is and presumably has
been a very common fish in Florida waters. The worked fish
remains are predominantly teeth of various genera of shark
(Isurus, Charcharhinus, Galeocerdo) and the dentary of barra-
cuda (Sphyraena sp.). The sharks teeth are drilled in 1, 2
or 3 places in the root as though to provide a way of haft-
ing them. The barracuda dentaries had holes drilled proxi-
mally and in two cases a stick was inserted into the cavity
occupied in life by the articular. This again seems to be a
means of hafting.

Invertebrate Remains
In addition to the vertebrate remains from the Van
Beck's excavation there was a small mat of sponge (Hexacto-
nellida) spicules. Sponge spicules have been found previous-
ly in association with archeological sites (Linne, 1925) and
may sometimes be used as clay temper although there seems to
be no evidence of this here.

The particular assemblage of animal species found is in-
fluenced by the animal resources in the area of each site
and the technical ability of man to utilize these resources.
All the animal species represented at the two sites still
exist in the area today. The Gulf waters probably provided
the greatest quantity of protein. This reliance upon ani-
mals living in the sea is seen in the Van Beck's site where
the excavated sample may be considered representative of the
site as a whole.

Little is known about these peoples fishing techniques
but something may be surmised from the types of fishes that
were caught and the fishing implements that were excavated
by Gushing. The fishes represented at both sites are pre-
dominantly inshore forms including bottom-dwellers and
schooling fishes. Remains of mullet have for the first time
been in a Florida site, namely the Cushing site. Mullet
abound in Florida waters today and there is no evidence to
suggest that they have newly invaded this area therefore the
reason their remains are not abundant in Indian sites must
be that the Indians did not generally have the means to
catch them. Among the various types of netting found by
Cushing is some described by him (Cushing 1897) as long
coarse gill netting which was probably used to catch the
mullet. In the Cushing collection (Cushing 1897) there is
also fine netting used for dip nets, floats and sinkers for
the long coarse nets, and fish hooks.

I gratefully acknowledge the support of National Sci-
ence Foundation G-17948 which made this study possible.

Table I

Animal species represented at the Cushing and Ven Becks
Sites. The figures on the left indicate the number of speci-
mens and on the right indicate the minimum number of indivi-
duals, a single figure indicates the minimum number of indi-

Van Beck Site Gushing
Animal Species Pit A Pit B Total Site Total

Didelphis marsupialis 1
Sylvilagus sp. -- 1
Canis familiaris 5

Lynx rufus
Procyon lotor
Urocyon cinereoargenteus
gray fox
Odocoileus virginianus 1,2
white-tailed deer
Gavia immer
Moris bassana
Phalacrocorax auritus
Ardea herodias
great blue heron
Casmerodias albus
American egret
Egretta thula
snowy egret
Guara alba
white ibis
Aythya affinis
lesser scaup duck
Lophodytes cucullatus
hooded merganser
Mergus serrator
red-breasted merganser
Cathartes aura
turkey vulture
Coragyps atratus
black vulture
Numenius americanus
long-billed curlew
Larus argentatus
herring gull
Larus delawarensis
ring-billed gull
Thalasseus maximus
Royal tern
Alligator mississipiensis
Kinosternon bauri
mud turtle
Terrapene carolina'72
box turtle
Malaclemmys terrapin
diamond-back terrapin

- 1

1/1 3/3 4 1

1- -3 1

10/7 3/3 10 6

- 1

- 1

- 1

1 1

1 1

- 2


2/2 5/3 5

1/1 2/1 2

1/1 2/1 2 1

1 8


1/1 1 1

- 1/1 1 2


- 1

1/1 1 3

1/1 4/2 3

2/2 3/2 4 18

- 16/5 5 7

Deirochelys reticularia
Gopherus polyphemus
gopher tortoise 1
Cheloniidae cf Chelonia mydas 42/14
green turtle
Trionyx ferox
soft-shelled turtle
Lamnoidei 3/2
sand-mackeral-nurse shark
Isurus cf I. oxyrinchus 1,?2
mako shark
Scyliorhinoidei 49/10
cat-requiem-hammerhead shark
group (some Galeocerdo
Carcharhinus sp. 1,?2
requiem shark
Galeocerdo cuvier
tiger shark
Pristis pectinatus 5/2
saw fish
Aetobates narinari 2/2
spotted eagle ray
eagle rays
Megalops atlantica
Bagre marinus 3/3
gafftopsail catfish
Galeichthys felis 11/9
sea catfish
Centropomis sp. 20/13
Epinephelus sp.
jew fish
Mycteroperca sp.
grouper or gag
Lutianus sp.
Caranx cf. hippos 10/6
Cynoscion nebulosus
spotted se trout
Pogonias cromis 10/8
black drum
Sciaenops ocellata 7/6
red drum
Archosarqus cf A. probato-
cephalus 41/18

5/1 1

8/4 4 1

182/12 26 4

1 1

13/4 6

- 7 2 1

130/17 27 -

3 1

17/7 9 2

3/3 3

1/1 1

13/6 9

446/92 101 8

115/39 52 3

- 1


67/25 31 3

4/4 4

84/28 36 6

87/26 32

819/181 199

Sphyraena sp.1,2
Mugil sp.
Chilomycterus schoepfi
Opsanus sp.
toad fish

1Some bones of these species

2ore specimens from the Cus
tional Museum collections.



2/2 2

2/2 2

1/1 27/16 17

have been worked

;hing site are in the U. S. Na-

Literature Cited

Cushing, F. H.
1897 Exploration of Ancient Key
the Gulf coast of Florida.
Soc. 35 (153): 329-433.

Dwellers' remains on
Proc. Amer. Philos.

Haag, W. G.
1948 An osteometric analysis of some aboriginal dogs.
Univ. of Kentucky Report in Anthro. 7 (3): 107-

Houch M. van W.
1951 Animal remains from South Indian Field. Yale
Univ. Publ. in Anthro. no. 45, appendix p. 51-

Linne, S.
1925 The technique of South American ceramics. G'te-
borgs Kungl. vetenskaps-och vitterhets-samnalles
Handlinger. Fj'rde F'ljden. 29 (5):1-199.

Rostlund, E.
1952 Freshwater fish and fishing in nature. North
American Univ. of Calif. Publ. in Geog. 9:1-313.


Comparison of Dog Measurements

Dimension in mm. Indian Alabama Shell Heap? Cushing's Key Marco
Field? # Mean (range) Small Dog Large Dog

Alveolus 1, - M, 77 51 82.5 (73.7-92.0) 80 , 80

Alveolus C - M3 76.6 60 77.3 (68.5-85.7) 76 , 78

Alveolus Py - M3 62.5 49 64.0 (55.6-71.2) 63 , 64

Alveolus Py - My 58.8 62 60.0 (52.0-67.2) 58 , 60

Alveolus Pp, - M, 49.3 64 51.1 (44,2-57.0) 50 , 51

Alveolus Mj - M, 31 64 30.2 (26.8-33.7) 31 , 34

P,, length 10.8 ll, 11

M, length 19.4 62 18.7 (16.5-21.3) 19 , 19

Alveolus 11 - m2 43. 82.4 (73,0-90.3) 84.7 93.5

Alveolus C - M2 47 68.4 (60.2-75.2) 70.0 79.0

Alveolus P! - mw? 42 56.7 (50.2-61.7) 56.9 62.5

Alveolus P2 - w? 49 50.3 (44.5-55.4) 51.0 55.0

Alveolus mu! - * 50 15.7 (14.0-19.4) 18.0 19.5

p* iength 51 16.1 (14.1-19.0) 16.5 19.2

a Measurements of complete lower jaw from
South Indian Field (Houch 1951:56)
b Haag 1948:146-147


Lyman 0. Warren and Ripley P. Bullen

ABSTRACT: Certain projectile points recently
dredged from Terra Ceia Bay in Manatee County,
Florida, are like those found in the Dalton zone
of the Stanfield-Worley rock shelter of northern
Alabama. Two radiocarbon dates from the Dalton
zone average 7300 B.C. While isolated Dalton
points have been found before in Florida, this is
the first concentration suggesting occupation at a
site during a Dalton period.

The accompanying illustration covers certain projectile
points found from 1958 to 1963 by Warren and sent to Bullen
at the Florida State Museum for identification and photo-
graphing. These points were found eroding from a clay beach
along the northeastern side of Terra Ceia Bay immediately to
the east and west of where US Highway 19 joins the mainland
between Palm View and Lakes Point, shortly south of Rubonia,
Manatee County, Florida. We understand the dirt of this
beach was dredged out of Terra Ceia Bay by the present owner.

Also found on the same beach were plain, sand-tempered,
black sherds; plano-convex turtle back scrapers; a hammer-
stone; stemmed projectile points (undoubtedly of later prov-
enience than the points discussed in this paper); and infre-
quent chert spalls. Well-mineralized Pleistocene fossil re-
mains included bison horn cores, fragments of mammoth teeth;
bison and horse teeth; turtle shells; manatee ribs; small
fragments of mastodon teeth; shark's teeth; and unidentified
bone fragments. Both artifacts and fossils were black or
gray in color. Shellfish included primarily small oysters
and shell clams quahogg, Venus mercenaria). No bone, wood,
shell, or steatite artifacts were found.

Many of the specimens illustrated are similar to those
found in the lower or Dalton zone of the Stanfield-Worley
rock shelter of northern Alabama (DeJarnette, Kurjack and
Cambron 1962). As they have not been named in Florida and
as this is the first instance in which they have been found
in reasonable quantities, we will use Alabama terminology in
describing and referring to them.

All of the points in our illustration exhibit basal
grinding except for the stemmed point (Fig. 1, k) in the
lower right hand corner. This is not a typical Alachua or
Marion stemmed point of the Late Preceramic Archaic period
of Florida in that the tip is much thicker and the edges

more serrated. In Alabama it would be called "Kirk Serrated'.'

The last specimen in the middle row (Fig. 1, h) is a
typical Florida Bolen point with a rubbed base, side to cor-
ner notches, and pronounced unifacial (twisted) beveling
which, however, does not show clearly in the illustration.
In Alabama this specimen would be called a Big Sandy I point.

The first two specimens (Fig. 1, a-b) may at first
glance be thought to be Suwannee points but close examina-
tion indicates such is not the case. They have a noticeable
bend or break in the outline of their blades and extremely
small basal ears with, especially in the case of the first
one, a pronounced flare. These are not Suwannee point char-
acteristics. These points easily fit the Alabama classifi-
cation of Greenbrier Dalton of which DeJarnette, Kurjack and
Cambron (1962:57) write "The distinguishing feature of this
type is the expanded, rounded tangs." The next two speci-
mens (Fig. 1, c-d) are excellent examples of the Nucholls
Dalton type although their bases are more concave than those
illustrated by DeJarnette, Kurjack and Cambron (1962:65).
All four of these Dalton points exhibit basal thinning.

The first three specimens in the middle row (Fig. 1,
e-g) are easily classified as Greenbrier points in the Ala-
bama nomenclature (DeJarnette, Kurjack and Cambron 1962:56).
Two (Fig. 1, e-f) have the serrated edges and pronounced bi-
facial beveling characteristic of Greenbrier points. If
they had opposite bevels we would have to consider them a
variation of the Bolen point.

The first specimen in the bottom row (Fig. 1, i) is a
uniface hafted scraper. The next point (Fig. .) is a narrow
lanceolate-shaped side-notched type which is rare in the
southeast. It's shape and chipping matches exactly DeJar-
nette, Kurjack and Cambron's (1962:47) illustration of a
Beaver Lake point although it is not as thin as their defini-
tion would suggest.

The material of these points is interesting because of
its variation. Two (Fig. 1, b, d) are made of a semi-trans-
lucent chert not uncommon in the Tampa Bay area. Three
(Fig. !, g, i, k) are of a blackish chert rare in peninsula
Florida but more common towards the northwest. The balance
are of whitish or gray patinated chert, containing many vari-
ously shaped inclusions. One of these (Fig. 1, f) shows,
where broken, an interior brown color.

In their report on the work at the Stanfield-Worley
shelter, DeJarnette, Kurjack and Cambron (1962:84) delineate

a Dalton complex of projectile points. Six of ten points
from the Terra Ceia beach collection represent four of the
seven types illustrated for this Alabama Dalton complex.
The remaining points are the three Greenbrier and the
stemmed Archaic-like point. The latter we have mentioned as
similar to a Kirk Serrated.

DeJarnette, Kurjack and Cambron (1962:91-109) present
ten stratigraphic provenience tables. In all cases Big
Sandy I (Bolen-like) points are present in the lower or Dal-
ton zone and have the same vertical distribution as do the
various types of Dalton points. Greenbrier points concen-
trate at the top of the Dalton zone while Kirk Serrated
points, when present, are always a little higher.

It seems evident that there were Indians living in Flo-
rida beside or near the Gulf around 7000 B.C. whose projec-
tile points, and presumably culture, was similar to those of
the "Dalton complex" of northern Alabama. It is believed
additional sites of this culture will be found as more work
is done in Florida or under the waters of the Gulf. It is
believed these people lived in a period chronologically af-
ter that of those who made and used Suwannee points. Con-
tinuing to live in Florida, they appear to have given up
their typical Dalton-like points and we recognize them as
the Early Preceramic Archaic makers of Bolen points, a pro-
jectile point type they appear to have introduced into Flo-
St. Petersburg

Florida State Museum

Reference Cited

DeJarnette,David L., Edward B. Kurjack, and James W. Cambron
1962 "Excavations at the Stanfield-Worley bluff shel-
ter." Journal of Alabama Archaeology, vol. viii,
nos. 1-2, pp. 1-124. University.

c d

< f 6 h

Fig. 1. Projectile points from beach at Terra Ceia.

a-b, Greenbrier Dalton; c-d, Nuckolls Dalton; e-g, Greenbrier;
h, Bolen or Big Sandy I; hafted scraper; J,- Beaver Lake;
R, Kirk Serrated.


Charles B. Harnett

After 1560, the principal trade route from America to
Spain followed the Gulf Stream past the Florida Keys up be-
tween Florida and The Bahamas and eastward to Bermuda. The
ten to forty vessels that made up Spain's annual fleets used
this route for 250 years to carry New World riches to Europe.

Often during this period, ships navigating Florida wa-
ters went to the bottom the victims of storms, warships, and
bucaneers. Not long after they sank, some were salvaged by
primitively equipped expeditions. Others, however, could
not be located or worked due to poor underwater conditions,
insufficient information about locations, and inadequate div-
ing and salvage equipment.

Many of these wrecks have remained untouched for the
same reasons that they eluded early salvors. Some were long
ago covered by shifting sand. Others are out of range of
divers. Most are hidden in murky, shark infested waters.

Yet, they exist today, on or in the ocean bottom, veri-
table sunken museums of historical artifacts from one of A-
merica's most exciting and colorful periods.

When the author directed a magnetometer search for such
early shipwrecks (Harnett 1962) several anomalous locations
had been discovered in very murky water off Cape Canaveral.
Lack of time, financing, and equipment prevented further in-
vestigation at that time.

In December of 1963, however, one of the sites was re-
located four miles offshore with a magnetometer. Partici-
pants in the search and later excavation were the author and
two associates, Charles Curry and A. L. Hutchison, of N.Y.C,
all having considerable experience in the type of work to

With visibility of about four feet, the site was visu-
ally examined. It lay at a depth of between 40 and 45 feet
as measured by a lead line, and the bottom consisted of dark
silt and fine shell. Within an area of some 3,000 square
feet, five large encrusted cannon were found, one partially
buried and the others totally exposed. All were of iron.
The only samplings made were at the bore of the four avail-
able cannon and showed lime encrustation of slightly less

than two inches.

Trunions visible on the four exposed pieces were loca-
ted below the longitudinal axis of the barrels, indicating a
design date earlier than 1750 (Manucy 1949:41). Tentative
dating was further narrowed to pre-1730 by the absence of
bell-shaped muzzles.

Three of the four fully visible cannon were measured at
six feet in length and their bore at four inches. Weight of
each was estimated to exceed 4,000 pounds. Cascabels were
large, exceeding ten inches in diameter including encrusta-
tion. The fourth exposed cannon was 12 feet long with a
bore of slightly over five inches.

Cascabel and breech of the partially buried cannon were
visible and appeared similar to the 12-foot long piece.

Protruding from the bottom about five feet was the
fluke of anchor. It was heart-shaped, peculiar to 17th cen-
tury anchor design (Fineham 1800). Part of the anchor ring
was visible 18 feet away. From the anchor's position, it
could be determined that it had had no metal crossbar, cor-
roborating a date of at least 18th century (Fineham 1800).

No other artifacts were visible.

It was first thought that the cannon and anchor might
have been discarded from a vessel stranded on a nearby bar.
Sounding of the immediate vicinity, however, revealed no bar
shallow enough to strand a vessel having the greatest known
draft of that period. Due to this information and the rela-
tive positions of the cannon and anchor (Figure 1), it was
supposed that the site was that of a shipwreck, probably pre-
dating 1730.

To ascertain the length and breadth of the site, a sen-
sor of the search magnetometer was trolled slowly back and
forth over the area in a controlled search pattern (Figure 2)
One man held a line attached to the sensor and thus followed
its path underwater at a distance of 15 feet. This distance
between diver and sensor was necessary due to the sensitivi-
ty of the magnetometer, which is influenced by the proximity
of ferrous metallic objects worn by the diver.

A simple, one-way, boat-to-diver communication was es-
tablished by the topside operator tapping on a bell sub-
merged off the rear of the search vessel. When the tapping
stopped, the diver below understood that the magnetometer
was no longer passing over a ferrous area. Upon cessation

F',lt ~n

IZ- F#ot Canno,

=-, J Bu,+-et Cam.

Sea.rctk Vessel

Fig. 1 Surface Objects on Shipwreck Site

Ancheb. Flute




fM5heton te.r-
srrrh BO~t'tfn

\P. otc



Fig. 2 Search Patterns Outlining Wreck Site

of the tapping signal for 15 seconds (correlating with a
search path distance of about 30 feet), the diver would drop
to the bottom a weighted, yellow plastic dish. The topside
operator then would steer a reciprocal course parallel to
the previous search path. In this way, the site was outlined
with 14 yellow markers.

From this method, it was determined that the site mea-
sured approximately 140 feet in length and 50 feet in width.
Its length lay, roughly, northwest by southeast (Figure 2).

The above was accomplished in the span of two working
days from a 22-foot inboard vessel. Further investigations
had to await availability of time, money, suitable diving
conditions, and development of an air lift improved for use
in muddy bottoms.

The next investigation was carried out during two days
in March, 1964. The site had not been buoyed; therefore,
two hours of searching was required before it was once again
located. Underwater visibility was about five feet and the
submerged markers denoting the site's perimeter were easily
located, except for three which had been covered over by
drifting silt. An area four feet wide and ten feet long was
selected as a test trench some 20 feet northwest of the
southeasternmost end of the site. The area, centered midway
across the site's width (Figure 3), was staked out using
three-foot long aluminum rods forced into the sea floor.

For excavation, a conventional air lift was employed.
The dredge pipe had a four-inch diameter and operated from a
surface compressor providing air at the rate of eight cubic
feet per minute. Shark and large jewfish vanished from the
area after the air lift was started up.

Excavation proceeded slowly because of silt exhausted
by the air lift, with about three cubic feet of silt removed
every hour. By the end of the two-day working period, some
25 cubic feet had been excavated to a depth of one foot, and
four artifacts were uncovered.

First to be found were two wooden planks, one of oak
measuring six inches wide and 1 inches thick; the other of
teak measuring nine inches wide and two inches thick. Both
were less than four feet long and were broken at both ends.

Next, a strip of lead was uncovered from one foot of
silt. It was eleven inches long, five inches wide, and 1/8
inch thick. A one-inch width of one side was lapped over
the main piece. Next to the lead strip was a 10 x 14-inch

A- F;rt test tiench
a Seco.d t"st tIekTch
(Shadte Exava.ti;o Afre.)

Fig. 3 Location of Test Trenches


piece of leather. One edge was smooth as if cut, while the
other edges were ragged, apparently torn away from the ori-
ginal piece. No stitching was visible. Other scraps of
wood and leather were found throughout the test trench be-
fore operations were terminated.

In June, 1964, another trip was made to the site. Only
one underwater marker remained in the area, the others ap-
parently carried away by current or surge. None of the alu-
minum rods that had outlined the test trench were found, and
the excavated test pit was no longer visible. Because an-
other foot of partially buried cannon was exposed, it was
surmised that that much of the bottom had been displaced by
currents during the four-month interval. A day was spent in
once more outlining the perimeter of the site in the manner
described above.

A second test pit was then staked out and excavation be-
gun, parallel to and approximately 20 feet northwest of the
first pit. An improved air lift dredge was employed. Visi-
bility was approximately four feet.

In most cases, sand, silt and debris carried up an air
lift pipe is dumped back into the water, often creating a
worsening of underwater visibility in the area. The im-
proved device, which we call the "Controlift", operates in
the conventional manner -- like an underwater vacuum cleaner
-- except that the descent of silt and debris is controlled
so that it is not scattered about the site. The Controlift
also gives the operator complete decision over the amount
and distribution of suction to be used over the immediate ex-
cavation area.

In the remaining three hours available for excavation,
about 12 cubic feet of bottom were removed. Again, bits of
wood and leather were uncovered, along with small bits of
glass, two lead shot, and a portion of a metal crucifix mea-
suring two inches high by one inch wide (Figure 4). Shortly
before excavation was halted, the Controlift uncovered a
thin, light-colored mass about six by eight inches.

As the Controlift gently removed the silt from the ob-
ject, black letters began to appear. The letters were in
old Gothic type reminiscent of a Gutenberg Bible. It was
evident that we had uncovered part of an early book.

When the sheet of paper was completely uncovered, at-
tempts were made to raise it by hand. Each attempt resulted
in the paper, in an extremely soft condition, tearing away
from the fingers. After a quarter of the paper had been

I OaK -,i

Te o~k
~ LROA Leo~tkt r

F;Yst- -rest Tkr-amch


Leoth.. -e -mc

Se.onct That T'Tnc.I~

Fig. 4 Profile of Excavated Areas

lost, even with the most careful attempts to handle it, it
was decided to leave the remainder in situ, cover it up a-
gain with silt, mark it, and make a planned, more sophisti-
cated attempt to recover it the next day.

That evening, a device was contrived for removing the
paper from the bottom without destroying it (Figure 5). The
following day, when the paper was once more uncovered, the
device was carefully slipped underneath the paper, and the
hinged top was closed. The "end-less" box was then lifted
and ends were attached in place, providing a closed contain-
er for the paper and accompanying silt. After the box was
raised to the surface, it was immediately immersed in a wash-
tub of sea water, where the box was opened and its contents
slipped out into the water and allowed to settle to the bot-

As excavation continued, three more pieces of paper si-
milar to the first were uncovered, raised and placed in the
washtub. Scattered lead shot, leather, and wood also were

When operations had been secured at the end of the day,
it was discovered that the "pages" in the washtub were be-
ginning to come apart due to sloshing as the vessel rolled
at anchor. A fine mesh screen was slipped underneath each
paper and carefully raised out of the water. The paper
clung to the top of the screen as water seeped through the
mesh. Thus, the paper was transferred to plastic containers
of approximately 12 x 8 x 6 inches, which served to restrict
the articles' movement. The fine mesh screen has since be-
come the medium for handling these papers, serving faithful-
ly in this application.

In September 1964, the papers were photographed and
photos sent to the State Archive at Trier, Germany for possi-
ble translation. The Archivist replied that, being under-
staffed, his institution could not take on the painstaking
work of translation and that he had no idea of the papers'
original source. Copies since sent to the Archive at Vienna
and to two specialists in early German documents in London
are now being translated, and attempts are underway to trace
their origin.

Specimens of the paper were sent in November 1964 to
the Hammermill Paper Company, International Paper Company,
Institute of Paper Chemistry and to Dr. Dard Hunter, one of
the foremost experts on early documents.

Prior to a detailed analysis, Harry F. Lewis, Curator

Endle.B" Box

Fig. 5 Device for Excavating and Raising Paper

Fig. 5 Device for Excavating and Raising Paper

of Museums at the Institute of Paper Chemistry advised that
if the wreck is indeed of the late 17th century, the paper
probably would be of European origin and composed largely of
rag fibers, since the use of wood as a source of paper dates
back only to 1840-1850 and even then rags were widely used.

"If the fibers are from wood, however, it would be un-
usual that the paper structure has been retained; this would
indicate that it had been protected from any agitation by
currents of water...some protection would have resulted if a
number of sheets had been found together, say in a book or
pamphlet," he said.

Upon analysis of a specimen, Dr. R. W. Brown, Director
of Central Research for Hammermill Paper Company, reported
that "the paper is made of linen or possibly a mixture of
linen and cotton. It is very mildly beaten or fibrillated
according to modern standards. This means that very little
physical work had been one on the fiber to prepare it for
the paper forming step.

"It is, perhaps, not particularly surprising that the
paper has survived this length of time. Cellulose is gener-
ally resistant to decay when submerged in water. The fib-
rous material from which this was made probably was relative-
ly pure and would not have been degraded in processing. If
we assume that the paper was in some location where it was
not subjected to agitation from wave action, it would prob-
ably not disintegrate. Several people have commented on the
clarity of the printing. It is rather amazing that it has
remained so sharp and intact," he said.

Dr. Hunter gave the following account of his investiga-
tions: "We removed the fragment from the bottle of sea water
and after drying and pressing, it seemed to be in its origi-
nal condition. The paper is handmade and this would account
to at least a minor extent to the endurance. Also the paper
was animal sized, which would add greatly to preservation.
It is extremely doubtful if a modern book would last beyond
a few years under the same condition.

"The analysis of linen and cotton (by Hammermill) is
without doubt correct, and also the observation that the fi-
bers had limited treatment. These factors contribute to en-
durance. The ink is the usual lampblack and vegetable oil."

Results of analyses have not, at this writing (February
12, 1965) been received from the other laboratories.

No further operations have been carried out at the site



Fig. 6 Portion of One Recovered Paper.
Note date of 1636 in Text.

after the near approach of two hurricanes during the summer
of 1964. It is presumed that associated currents have de-
stroyed all markers and signs of earlier work. Lack of fi-
nances and qualified workers have prevented further investi-
gation and excavation of the site.

Although identification of the papers should provide a
more accurate dating for the site than the few artifacts ex-
amined, the following may now be inferred from correlation
of observations with historical data.

Cannon usually provide a reliable meansgof identifying
the period and nationality of a vessel (although pirate ves-
sels often captured and used cannon from whatever ship hap-
pened to be their victim (Esquemeting 1898). Since the use-
ful life of an iron cannon ranged from 25 to 40 years, an
early wreck can be closely dated if sufficient information
can be obtained from detailed investigation of its armament.

Unfortunately, circumstances made it impossible to
raise the cannon from the site, gissolve their encrustation,
and study them in detail. However, measurements made on the
bottom give a fair picture of their size and general makeup.
Clean lines of the pieces indicate a date later than 1650
(Manucy 1949:41). The 12-foot cannon has earmarks of the
English culverin of the 17th century, and the six-footers
resemble the common “'cannon" used by most maritime nations
in the 17th century.

If the vessel were Spanish, it may have been a large
patache or a nao. Both types usually carried iron, rather
than bronze, cannons. However, the Dutch employed cannon
similar to Spanish, French and English armament during this

Dimensions of the site suggest a mid-seventeenth cen-
tury Spanish nao having length of 124 feet and width of 35

The leather which, because of its good state of preser-
vation, must certainly have been tanned may have been part
of a shipment of hides, suggesting a cargo ship such as a
nao or patache.

During rather specific periods, certain countries used
lead sheathing to protect ships' hulls from the teredo, or
ship's worm. The Spanish widely employed lead sheathing
from 1520 until about 1800 (Harding 1918). English vessels
used such sheathing after 1640 until it was replaced by cop-

per sheathing in the 18th century. 45

Most maritime countries planked their vessels with oak
in the 17th century (Fineham 1800), but teak during this per-
iod was used only by the Dutch and Portugese in the Carib-

The heart-shaped fluke of the anchor appears to be of
the pre-1700 period. Size of the anchor's ring -- apparent-
ly 32 inches in diameter -- rules out a date later than 1815.

The crucifix suggests a catholic nationality of at
least someone sailing on the vessel. Spain, of course, was
the focal point of Catholicism during the Spanish Colonial
Period and, of those countries shipping in the New World,
she employed the most vessels. Yet, Catholics were found in
most European countries at that time.

The paper, apparently from a book, appears to be prin-
ted in an early Gothic script in old German. Robert Marx,
who had done considerable original research on the Spanish
fleets at the Archives of the Indies in Seville, Spain, has
opined that the papers are from an old German sailing atlas.

"It is not unusual," Marx said in personal correspon-
dence, "to find German sailing directions on Spanish ships
since many German pilots were used in the flotas (fleets).
At times when there was a lack of ship bottoms in Spain, the
King would order all foreign shipping seized and used in the
flotas, along with their crews. Many Dutch ships thus
seized used German crewmen and pilots. Germans were especi-
ally active in the 16th and 17th centuries as mining engi-
neers in the Spanish Indies. Furthermore, many of the mis-
sionaries sent to the Spanish colonies were German Catholics,
especially the Jesuits and those sent to Mexico to be trans-
ported overland and then on to the Philippines. Since your
wreck is off Cape Canaveral, it probably was a homebound one
and might rule out the missionaries, as they were sent a-
broad for life (unless they committed a serious crime or
were made a bishop).

"The possibilities that this find opens up may be even
greater than you suppose. On advice boats and even on gal-
leons, it was the royal law that all dispatches, mail, mani-
fests and the like should be kept in specially constructed
metal (often brass or lead) chests, waterproofed and well
weighted with lead so they would be flung overboard the mo-
ment an enemy tried to board a ship. Sooner or later, some-
one will find such a chest; the implications are tremendous.

Because of the scarcity of identifiable artifacts from
the site, a definitive determination of the vessel's nation-

al origin cannot be made. On the basis of available data,
however, it can be supposed that it was not French and prob-
ably not English. Portugese vessels were uncommon in this
area during that time. This leaves Spanish, Dutch or a pi-
rate ship as good possibilities.

As to time period, we would consider the shipwreck to
date, from evidence at hand, to the period preceding 1700,
but probably no earlier than 1650.

A good many questions remain about this site. How deep
in the bottom does it extend? Auger tests could give an
idea. Was the ship part of a fleet or a "loner"? Was it
perhaps a Spanish patache or dispatch ship operating between
St. Augustine and Havana? Was it a Dutchman carrying on il-
licit trade with the Spanish colonies, or some other nation-
ality gathering intelligence on the Spanish treasure fleets?
The variety of identifiable artifacts might suggest a pirate
ship; many operated from Amelia Island, the Florida Keys,
and The Bahamas. Some were known to anchor in relatively
sheltered areas off the Florida east coast while awaiting
the passing of unprotected vessels.

Under what circumstances did she sink? Was she de-
stroyed in combat, capsized in a storm, or ripped open on a
nearby shoal? What of the passengers and crew? Were they
saved or did they perish in the catastrophe? Is this wreck
documented historically, or might its excavation and identi-
fication solve a centuries-old mystery of the disappearance
of some sailing ship?

Most important, what artifacts are left in the remains
of the vessel, and what can they add to our knowledge of the

Many of these questions may be answered if the site is
carefully and systematically excavated. To carry out this
work requires a full-time, full-scale operation including
highly qualified divers specifically trained for this work,
or better, professional archaeologists skilled in underwater
excavation. Also needed are a suitable equipped vessel,
other necessary gear, and the funding necessary to carry out
the project over a six-to-twelve month period.

Until such time as these provisions are available, work
will continue as in the past: on a very limited scale when
time and funds permit.

The remarkable find, the printed papers, have already
marked this site as unique. For never before have printed

documents of such size, readability, and abundance been dis-
covered in such an old submerged site. The ramifications
are self-evident. With documents surviving over 250 years
in the bottom of the sea and still remaining legible, a new
horizon for submarine archaeology now becomes available.

Books, ship's logs, legal documents and state papers
may, with care, be recovered to fill important gaps in man's
knowledge of the historic past. This possibility sheds an
exciting new light on the slowly emerging science of sub-
marine archaeology.


Exquemelin, Alexandre Olivier
1892 The buccaneers and marooners of America. 2nd ed.
Howard Pyle, ed. London, T. Fisher Unwin.

Fineham, John
1800-1802 A history of naval architecture. 3 vols.

Haring, C. H.
1918 Trade and navigation between Spain and the In-
dies. Harvard Economic Studies. Cambridge, Har-
vard Univ. Press.

Harnett, Charles B.

1962 Operation Lodestone. Preliminary report. A ma-
rine archaeological exploration off Cape Canaver-
al and the Florida Keys utilizing a proton magne-
tometer to determine locations of early ship-
wrecks, Orlando, Florida.

Manucy, Albert
1949 Artillery through the ages. (St. Augustine, Fla)
National Park Service, interpretive series, His-
tory, no. 3, Washington, U.S. Govt. Printing Of-



William C. Lazarus

Numerous aboriginal and historical sites are located on
and in the tidal waters of the Florida Gulf Coast. Meticu-
lous attention is given to land surveying of sites, record-
ing elevations to the nearest hundreth of a foot, but only
generalizations have been advanced as to the effects of land
subsidence and sea level changes with time. The latter two
factors can be of great significance to an understanding of
site topography at time of occupation.

Generalized data on sea level changes with time have
been published by Fairbridge (1960) in the Scientific Ameri-
can magazine. Sufficient data on rise and fall of the water
and rise and fall of the land is presented to permit some a-
nalysis of site conditions in specific areas.

This paper therefore presents no new data but simply ap-
plies the Fairbridge data in convenient form to Gulf Coast
sites in Florida or wherever it can be assumed that a land
subsidence rate of 0.1 mm per year (1.0 inch per 25 years)
has been consistently maintained. Fairbridge relies upon
tide-gauge data gathered over the past 50 years. These mea-
surements are obviously relative (i.e. composed of compon-
ents of sea level changes and land level changes). Fair-
bridge applies a rationale which differentiates between
these. He shows that "gauges located near large deltas or
regions of heavy sedimentation consistently show a subsi-
dence" and assigns a value of 0.1 mm per year subsidence for
the Gulf Coast from the mouth of the Mississippi to Key West.

In Florida there are tide-gauges at Key West, Cedar
Keys and Pensacola with records going back to at least 1930.
(Marmer 1952). Fifty years is hardly a satisfactory time
base upon which to extrapolate a subsidence rate in terms of
thousands of years into the past. However, using a differ-
ent approach to the subject of subsidence on the Gulf Coast,
LeBlanc and Bernard (1954) state that "surface mapping and
subsurface data along the coastal plains prove conclusively
that coastward tilting of this region has persisted since
the early Tertiary." Bernard (personal communication 1964)
has identified 5 factors which determine the movement of the
boundary between the land and sea. Three of these deal with
factors influencing the rise and fall of the water while two
deal with subsidence. He states the latter two to be:
Florida Anthropologist, Vol. XVIII, No. 1, March 1965 49


WE m. +5







-15 -- I -15


-25 -25

-30 0

0 500

1000 1500 2000 2500

3000 3500 4000

4500 5000 5500 6000 6500

All I, .




+20 4-O -n







n N


vwVr -80

-100 -100

-120 -120


-140-A -140

-160 VALDERS 160

-180 -180

-200 RKONA- -200

-220 HURON -220

-240 -0An

6000 7000 8000 9000 10,000 11,000

12,000 13,000
12,000 13,000



r 1 I



_LT __17







- A nn-- I


_ \fN _








TAZEWELL------ -380

-- -400
00 13,000 14,000 15,000 16,000 17,000 18,000 19,000







1) subsidence or uplift of the land caused by earth
movements including isostatic rebound resulting from loading
and unloading parts of the earth by glaciation, deglaciation,
erosion and deposition, and volcanic activity.

2) subsidence of the land caused by compaction of sedi-

Based on the geography of the Gulf of Mexico over the
past 15,000 years, the subsidence would appear largely at-
tributable to the sedimentation of the Mississippi and other
river systems flowing into the Gulf. During the period of
glacial run-off the sedimentation rate, and hence the subsi-
dence rate may have been accelerated. Conservatively then,
a uniform subsidence rate equal to the present has been used
in preparing Charts Nos. 1, 2 & 3.

These charts are in essence, a second derivative of the
Fairbridge data. They were obtained by first superimposing
the local subsidence rate on the oscillations of sea level
water on a worldwide base as these varied with time. The
combined differences were then measured and replotted on the
final charts with all measurements based on the current mean
sea level.

The charts may be used in two ways:

For sites presently above sea level and where occupa-
tion dates are known, it is possible to read an elevation
correction factor which can be added or subtracted to cur-
rently surveyed elevations to provide the true elevation at
the time of the occupation. For example: the historic town-
site of Pensacola on Santa Rosa Island (ES-22) is in the pro-
cess of excavation. The topological survey showed many
parts of the site to be very low relative to current mean
sea level. Artifacts were found down into the present water
table which lead to the quick assumption that the water ta-
ble must have risen but with no explanation as to why. How-
ever, by the use of Chart No. 1, it is immediately apparent
that this site has sunk approximately 2.0 feet relative to
mean sea level over the past 250 years and that this is the
cause of the water table rise. The two foot correction fac-
tor must be added to all contours and elevations to portray
the site as it was during occupation. This aids materially
in understanding the site and the people who built it.

The correction factor for a site currently above sea
level is obtained by reading out on the absyssa to the date
of occupation then down (or up for a period between 330 and
480 years ago) to obtain the correction factor.

For sites currently submerged, it is possible to use
the chart to determine in what time period or periods the
site was actually exposed above the then current sea level
and to determine the elevation which it had at that time.
When the depth of the site below current mean sea level has
been established, read horizontally until an "exposed" area
is reached. The dates of exposure can then be read on the
absyssa and the elevation of the site at that time can be
read vertically.

Table No. 1 presents generalizations on the flooding of
sites along the Gulf Coast. It must be borne in mind that
local anomalies in subsidence and uplift may have occurred to
disturb the assumed uniform subsidence rate used in these
calculations. As in all dating techniques which deal with
the distant past, judicious use of this data must be made.
Obviously, storm tide flooding has not been taken into ac-
count. This latter type of short term flooding could occur
on sites which are as much as 10 feet above present day sea
levels, as demonstrated by recent hurricanes.

Validity of the Sea Level dating technique for sub-
merged sites can be grossly demonstrated now for two loca-
tions on the Florida Gulf Coast. A comparison of date in
one instance is possible through known ceramic types which
have been dated. At the second submerged site compatibility
can be demonstrated with a carbon-14 derived date on the
site. Further, the submerged portion of Garcia Site in
Southeastern Louisiana (which is outside the geographic li-
mits set for this paper but where the 1.0 mm per year subsi-
dence applies) appears to be amenable to dating by this
method. No underwater sites are known which lack compatibil-
ity with this technique.


When site elevation Its occupation must
relative to M.S.L. date from at least-
is -
-3.0' 600 years ago
-5.0' 1,100 years ago
-7.0' 1,675 years ago
-10.0' 2,630 years ago
-16.0' 4,080 years ago
-21.0' 6,000 years ago
-37.0' 6,250 years ago
-56.0' 7,000 years ago
-88.0' 7,800 years ago
-112.0' 9,500 years ago


When site elevation Its occupation must
relative to M.S.L. date from at least-
is -
-130.0' 10,000 years ago
-158.0' 11,600 years ago
-210.0' 12,000 years ago
-248.0' 13,300 years ago
-342.0' 16,300 years ago
-367.0' 16,900 years ago

At depths of 380 feet and below, human occupation sites
appear to be impossible in this area based on currently
available data. The late Wisconsin Glaciation appears
to have produced the greatest drawn-down in sea levels
which occurred within the past 75,000 years.

The "One Fathom Site"
Wilford T. Neill of the Flarida State Museum, presented
a paper titled "The One Fathom Site, A Midden Beneath the
Sea" at the 1964 Annual Meeting of the Florida Anthropologi-
cal Society. This site is in the Gulf of Mexico over half
a mile from the present shoreline in the vicinity of New
Port Richie, Fla. From this site he produced a collection
of lithic materials and Deptford Period sherds. There were
no ceramics or artifacts present which could be associated
with later periods. All indications are that occupation of
this site ended sometime during the Deptford Period.

A Deptford Period Mound in Jefferson County (Je-53),
which is 190 miles northwest of the submerged midden afford9
the nearest dated Deptford site. A Carbon-14 date of 28501
110 years before present has been recorded. The Deptford
Period is generally considered as extending from 2,000 B.P.
to 3,000 B.P.

According to Chart No. 1, the One Fathom Site, lying 6
feet below present mean sea level would have been exposed
and very suitable for habitation during the time interval
3,400 to 2,600 B.P. It would then have been flooded until
2050 B.P. and then been barely above sea level until 1850 B.
P. From 1850 B.P. to 1650 B.P. it was exposed with an ele-
vation of about 3 feet.

The sea level dating technique is thoroughly compatible
with the ceramic materials recovered from the One Fathom
Site and favors the occupation being in the early part of
the Deptford Period. In addition the Sea Level Dating Tech-

nique is harmonious with the Archaic materials from this
site since it was dry in Archaic times except between 3,400
and 3,900 B.P.

The Warm Mineral Springs Site (So-18)

In the springhead at Warm Mineral Springs (also called
Little Salt Spring), Sarasota County, Fla., there is a cave
now submerged 80 feet below the surface. This cave has sta-
lactites which do not form under water. Obviously at some
period in the past this cave was dry for a considerable per-
iod of time. A burnt log was found in the cave which when
subjected to radiocarbon testing yielded a date of 10,000 -
200 years B.P. Bones of seven human beings have been found
in this cave but to date no artifacts have been reported
from it. No association between the bones and the log has
been demonstrated.

Obviously the log was burned in air about 10,000 years
ago, and Chart No. 2 shows that the cave was dry at that
time. This must be weighed against the probability that a
random burned log of just the right age found its way into
the spring at a much later date (after the cave was flooded
circa 7,000 years ago), became waterlogged at an appropriate
spot on the surface so that a convenient current would carry
it into the cave as it settled toward the bottom.

From Chart No. 2 it will be noted that the floor of
this cave would have been as much as 50 feet above sea level
10,000 years ago. The hydrostatic head of the spring and
the character of the run-off channel from the spring to the
Gulf may have modified this 50 feet of "free board" to some
extent but probably not the whole amount based on the evi-
dence that it was dry at one time.

Chart No. 2 clearly indicated that the cave would indeed
have been dry 10,000 years ago but that about 9,000 years
ago it became flooded for a span of about 500 years and then
could have been marginally dry until about 7,300 years ago.
From that date to the present the chart indicates continuous
flooding. The chart is therefore compatible with the carbon
14 evidence from this site.

Discussion and Conclusions:
From the Charts it is obvious that any site in the area
under consideration which is now more than one foot above
current sea level has not been subjected to long term flood-
ing during the past 75,000 years. This date marks the end
of the Sangamon Interglacial Period the one which pro-
ceeded the Wisconsin Glaciation. When land subsidence and

sea levels are considered, it is within the realm of possi-
bility to have human occupation sites on the Florida Gulf
Coast which are now submerged as much as 367 feet. Inhabi-
table caverns at the spring heads of Florida's big springs
are entirely plausible. The data presented herein is compa-
tible with archaeological data on the cavern sites at Silver
Springs, Warm Mineral Springs and even the 200-220 foot
depths does not appear to be valid if we consider a time
period of 13,000 to 15,000 years ago.



Rhodes W.
The Changing Level of the Sea. Scientific Amer-
ican, Vol 202, No. 5, p 70, New York.

Gagliano, Sherwood M.
1963 A Survey of Preceramic Occupations in Portions
of South Louisiana and South Mississippi. The
Florida Anthropologist, Vol. XVI, No. 4, p 105

LeBlanc, Rufus J. and Bernard, Hugh A.
1954 Resume of Late Recent Geological History of the
Gulf Coast, Geologie en Mijnbouw, Nr. 6, (NW.
Ser.) 16 e Jaargang Page. 185-194 June 1954.

Marmer, H. A.
1952 Changes in Sea-Level Determined from Tide Obser-
vations. Council on Wave Research of the Engi-
neering Foundation, Univ. Calif., Berkeley, Proc
2nd Conf. on Coastal Engineering, Cahpter 6, -
pp 62-67.

Morell, L.

No. 4, p

Mound (Je-53), Florida: A Preliminary
The Florida Anthropologist, Vol. XIII,
101. Gainesville.

Neill, Wilford T.
1964 The Association of Suwannee Points with Extinct
Animals of Florida. The Florida Anthropologist,
Vol. XVII, No. 1, p. 17. Gainesville.

Olsen, Stanley J.
1958 The Wakulla Cave. Natural History, Vol. 67, No.
7, p. 396, New York.

Royal, Williams and Eugenie Clark
1960 Natural Preservation of Human Brain, Warm Miner-
al Springs, Florida. American Antiquity, Vol.
26, No. 2, p 285. Salt Lake City.


John B. and Eunice Williams

During 1955-1965, while hunting Calusa artifacts in the
Chokoloskee area of Southwest Florida, we heard,through some
of the older residents, that there was evidence of the form-
er existence of a strange, "primitive'' race of man on Gopher
Key in the Ten Thousand Islands to the south. The report
was that this human was not an Indian but of a "lower mongo-
loid" type and, judging by the number of burned human bones
present at the site, may have been cannibalistic. We have
subsequently collected many such bones.

"Doc" Clarence Brown, a native of the area and an ex-
perienced guide, was particularly knowledgeable and enthusi-
astic about this interesting subject. We had no difficulty
in persuading him to take us by boat to Gopher Key where we
spent a day (April 27, 1956) unearthing bones and artifacts,
and fighting mosquitos. The site, a low brushy key of con-
siderable area surrounded by mangrove swamp, is now within
the confines of the Everglades National Park, a fact which
precludes further amateur bone hunting.

According to an earlier report, skulls associated with
this type of man have been found separated from the body and
apparently were buried in a circle. Those described in the
present writing, however, did not seem to be so disposed, a
large assortment of skeletal remains being located in the
same area with no obvious arrangement. Possibly they had
been disturbed. Generally speaking, these bone fragments
indicate a man of abnormally tall stature, the skulls and
teeth pointing definitely toward an unusually large, though
flattened head. There is absolutely no evidence that such
cranial compression was in anyway artificially produced.

The two skulls collected on this expedition were dis-
covered close to the surface and more or less intact, but
immediately fell to pieces when we removed them from the
damp ground. We were, however, able to reassemble one, vir-
tually complete skull-cap with frontal and parietals, and it
is this adult specimen which is presented herewith as pri-
mary evidence for a very distinctive, not to say astonishing,
human type. Another incomplete cap (fig. B-1) composed of
the disc of the frontal bone plus parietal fragments, is
quite similar and most certainly related.

According to these particular specimens (fig. A-l, 2 &3
and B-1) man of this race possessed a cranium distinctly

wider than that of the typical Calusa Indian taken from the
same locality (fig. C-1 and 2). This is especially notice-
able in the distance across the zygomatic processes of the
frontal bone (orbital span) which is considerably wider in
the new skull. Also the zygomatic processes are heavier and
the nasal process relatively wide as compared with the Calu-
sa and other skulls (figs. A-1,2,3 C-1,2 and D). But the
most obvious point of difference between the two types is
the abrupt facial angle giving the new skull an extremely
sloping, though arched, forehead. This unique feature is re-
flected in profile and face-on views as an exceedingly low
dome with retreating frons. The cranial bones also differ
in being more massive and, together with great width of the
head in the parietal region, the large thickened jaw and big
teeth, combine to suggest a man of powerful build.

It is hoped that we will be able to add to these skull
caps various other parts of the skulls including mandible
and teeth.

There is, as yet, no artifactual evidence for a culture
peculiar to Gopher Key and therefore possibly ascribable to
this strange colony. However, we feel that our find in
other respects is of sufficient interest and importance to
present in this brief form to the Florida Anthropological
Society for inclusion in its publication, The Florida Anthro-

We are deeply indebted to Dr. J. Manson Valentine, of
Miami, well known zoologist, for his keen interest in, and
study of, the subject of this paper and for his comments and
help in bringing out salient points.


Fig. A-i Frontal view of atypical skull fragments.

Fig. 2 Left: Normal Indian skull from Gopher Key.
Right: Atypical skull from Gopher Key showing
extremely low vault. View from front.

Fig. 3 Oblique views of normal Indian skull (left) and
atypical skull (right) from Gopher Key.


The following law was proposed by Dr. Robert 0. Vernon
of the Florida Geological Survey at a meeting December 10,
1964. Members of the Florida Anthropological Society are
advised of this in case they choose to mobilize before the
legislature meets. Dr. Vernon proposed that the act be ex-
tended to apply to all lands, public and private.

AN ACT relating to objects of antiquity on lands owned by
state; defining objects of antiquity; vesting title to said
objects in state; providing for issuance of permits; provid-
ing for disposition of objects collected; prohibiting the re-
moval, taking, use, sale, destruction or injury of objects
of antiquity; providing a penalty; providing an effective

Be It Enacted by the Legislature of the State of Florida:

Section I. Objects of antiquity for the purposes of
this act shall be defined as meaning any property which is
situated on lands owned by this state, including submerged
lands within the boundaries of the state, which has been
lost or abandoned for a period of twenty (20) years or more.
Such property shall include, but not be limited to, arti-
facts, treasure trove, structures, derelict property of any
nature, ships, their cargo, parts and fittings and any ob-
ject of scientific, historical or archaeological value.

Section 2. Ownership of said objects of antiquity
shall vest in the state.

Section 3. The governor of Florida is hereby autho-
rized on recommendation of the director of the state board
of conservation to declare by public proclamation that his-
toric and prehistoric structures and sites, and other ob-
jects of scientific interest that are situated upon the
lands owned or controlled by the state, shall be state parks
or preserves, and may reserve as a part thereof such parcels
of land as may be necessary to the proper care and manage-
ment of the objects to be protected.

Section 4. Permits for the examination of ruins, the
excavation of archaeological sites, and the gathering of ob-
jects of antiquity or general scientific interest, may be
granted by the director of the state board of conservation
to institutions which they may deem properly qualified to
conduct such examination, excavation, or gathering, subject
to such rules and regulations as the aforesaid board may pre-
scribe; provided, that the examination and gatherings are

undertaken for the benefit of reputable museums, universi-
ties, colleges, or other recognized scientific or education-
al institutions as the director of the state board of conser-
vation may prescribe, and;
provided further that not less than fifty per cent (50%) of
all specimens so collected by nonresident institutions shall
be retained as the property of the state, unless the board
of conservation shall expressly accept a smaller proportion,
as meeting the requirements of this act. This act shall not
interfere with the making of natural history collections by
individuals for scientific purposes only, provided that such
individuals secure permission as prescribed in this section.

Section 5. Unless other locations be designated by the
director of the state board of conservation, the museum of
the division of geology of the state board of conservation
shall be the depository for all collections made under the
provisions of this act and shall distribute materials from
such collections to other state universities and museums on
request of the governing bodies of said institutions, when
in the opinion of the director of the state board of conser-
vation, proper arrangements are made for the safe custodian-
ship and public exhibition of such material.

Section 7. Any person who shall appropriate, excavate,
injure or destroy any historic or prehistoric ruin, site, or
any object of historical, archaeological or scientific value,
situated on lands owned or controlled by the state or its
institutions, without a permit or recommendation of the di-
rector of the state board of conservation, shall be fined in
a sum of not more than two hundred dollars ($200.00) or be
imprisoned for a period of not more than sixty (60) days in
the county jail or shall suffer both fine and imprisonment
in the discretion of the court; and it shall be the duty of
any employee of the state board of conservation and the
Florida fresh water fish and game commission, or any peace
officer, including sheriffs, to proceed against any viola-
tion of this law, and the duty of state attorneys and county
solicitors of this state to prosecute anyone violating the
provisions of this act.

Section 8. All acts or parts of acts in conflict with
or within the purview of this act, are hereby repealed.

Section 9. This act shall take effect immediately upon
becoming a law.


Price of Back Issues

The price of back issues of The Florida Anthropologist
is $1.00 for each single issue, $2.00 for each double issue.
Double numbers are Volumes, I through V; IX, Nos. 3-4; X;
XIII, 2-3, XIV.

Out of Print Numbers

Volumes I, Nos. 1-2, 6:4, 8:2, 8:4, 10:1-2,11:4, 13:1,

and publicationn No. 4.

Full Text