• TABLE OF CONTENTS
HIDE
 Copyright
 Table of Contents
 Some Florida Radiocarbon Dates...
 Excavations at Grossman Hammock,...
 The Turner River Site, Collier...
 Test Results at the Paw Paw Mound,...
 Contributors to this Issue
 Membership Information






Group Title: Florida anthropologist
Title: The Florida anthropologist
ALL VOLUMES CITATION PAGE IMAGE ZOOMABLE
Full Citation
STANDARD VIEW MARC VIEW
Permanent Link: http://ufdc.ufl.edu/UF00027829/00122
 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-]
quarterly
regular
 Subjects
Subject: Indians of North America -- Antiquities -- Periodicals -- Florida   ( lcsh )
Antiquities -- Periodicals -- Florida   ( lcsh )
Genre: periodical   ( marcgt )
 Notes
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: VID00122
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
    Copyright
        Copyright
    Table of Contents
        Table of Contents
    Some Florida Radiocarbon Dates and their Significance
        Page 31
        Page 32
        Page 33
        Page 34
        Page 35
        Page 36
    Excavations at Grossman Hammock, Dade County, Florida
        Page 37
        Page 38
        Page 39
        Page 40
        Page 41
        Page 42
        Page 43
        Page 44
        Page 45
        Page 46
    The Turner River Site, Collier County, Florida
        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
    Test Results at the Paw Paw Mound, Brevard County, Florida
        Page 61
        Page 62
        Page 63
        Page 64
        Page 65
    Contributors to this Issue
        Page 66
    Membership Information
        Page 67
        Page 68
Full Text





COPYRIGHT NOTICE

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-Le 9t0oL na c4ndtIofogi




Vol. IX June, 1956 No. 2







CONTENTS



SOME FLORIDA RADIOCARBON DATES AND
THEIR SIGNIFICANCE ...................... Ripley P. Bullen 31

EXCAVATIONS AT GROSSMAN HAMMOCK,
DADE COUNTY, FLORIDA ................... Marvin J. Brooks, Jr. 37

THE TURNER RIVER SITE, COLLIER COUNTY, FLORIDA William H. Sears 47


TEST RESULTS AT THE PAW PAW MOUND,
BREVARD COUNTY, FLORIDA .................... Carl A. Benson 61

CONTRIBUTORS TO THIS ISSUE .............................. 66













SOME FLORIDA RADIOCARBON DATES AND
THEIR SIGNIFICANCE

Ripley P. Bullen

In June, 1953, I conducted archaeological excavations beside the
Chattahoochee River in northwest Florida for the Florida State Museum.
Field work was sponsored by the National Park Service under contract
14-10-0100-134.

One of the sites investigated, known as J-5, was composed of super-
imposed Archaic (pre-ceramic), Orange (fiber-tempered pottery), Deptford,
and Ft Walton zones. This site was in the present natural river levee;
sterile waterlaid deposits separated the cultural zones. We secured suf-
ficient charcoal for radiocarbon dating from the Orange and Ft. Walton
zones. About 3 miles away we tested a small, Kolomoki period, shell
midden and collected carbon for dating purposes. This second site is
known as Ja-63.

Late in the spring of 1955, some money being available, we secured
radiocarbon dates on these three components from the University of Michi-
gan Memorial-Phoenix Project Radiocarbon Laboratory through the courtesy
of Dr. James B. Griffin. The results were as follows: Site J-5, Ft. Walton
zone 1400 A.D. plus or minus 200 years; Site Ja-63, Kolomoki period
- 350 A.D. plus or minus 250 years; and Site J-5, Orange period 1200
B.C. plus or minus 250 years.

Before discussing these dates, I wish to mention two Georgia radio-
carbon dates which seem pertinent. One is Waring's date of approximately
3750 years ago (1800 B.C.) for a plain fiber-tempered period on Sapelo
Island (Griffin, 1952, p. 366). Waring has told me the charcoal for this
date was taken from the middle, not the base, of the plain, fiber-tempered
deposit. Hence, it seems reasonable to use 2000 B.C. as an approximate
date for the beginning of the undecorated portion of the Orange period.

The other Georgia date is 150 B.C. plus or minus 140 years for Site
9 HL 64, located in the Buford Reservoir (Fairbanks, 1954, p. 20). This
site produced simple-stamped, linear check-stamped, and check-stamped





sherds. Tetrapods were present in reasonable quantities. Although this
site is located a fairly long distance from Florida, it may be suggested
that the site represents a time period equivalent to late Deptford times.

The Florida dates need a little explanation. The one from the Ft.
Walton zone at Site J-5 would apply, I believe, to a middle or late middle
part of the Ft. Walton period. This is based on ceramic considerations
(Bullen, n.d.a).

The date for the fiber-tempered zone at Site J-5 must apply to terminal
or final fiber-tempered times (Orange 5: Bullen, 1954, p. 47) because we
uncovered in the shallow midden deposit not only fiber-tempered sherds
and steatite vessel fragments but also several St. Johns Plain and three
St. Johns Incised sherds (Bullen, 1954, p. 26, Pl. IV; n.d.a). Presumedly,
this occurrence represents the first, or nearly the first, introduction of St.
Johns ceramics into the Chattahoochee region. Apparently, St. Johns
ceramics had their origin in peninsular Florida. Logically, this origin
occurred before chalky pottery was introduced into northwest Florida. The
resultant date for the start of St. Johns ceramics is, if the radiocarbon
date is correct, substantially earlier than previous estimates (Goggin,
1950; 1952, p. 36; Bullen, 1952, p. 5; 1955, p. 55).

The pottery complex at Site Ja-63, the small Kolomoki period midden,
is similar to that found in Kolomoki period middens at the Kolomoki site
in southwestern Georgia (Sears, 195 a). Out of about twenty-five hundred
sherds, 34 per cent are Kolomoki Complicated Stamped; the balance of the
decorated sherds, less than 1 per cent of the total, includes Sears's "free-
incised" variety of Weeden Island Incised. This assemblage differs sub-
stantially from pottery found in the caches in the burial mounds at the
Kolomoki site (Sears, 1951b).

Such a difference between pottery from a village site and from a cache
in a burial mound is a good Weeden Island trait. However, I think the ap-
parent lack at Ja-63 of anything suggesting Middle Mississippian influences,
such as were included in burial mound caches at the Kolomoki site, is
significant. All Kolomoki Complicated Stamped sherds from Ja-63 appear
to be from straight-sided vessels. In this respect they differ from most
Kolomoki Complicated Stamped sherds found at the Kolomoki site. These
differences suggest, to me, time changes within the Kolomoki period of
southwestern Georgia.

If pottery like that at Ja-63 had been found before.Sears's work at the
Kolomoki site, it would undoubtedly have been classified as a Weeden
Island I complex. The radiocarbon date of about 350 A.D. would seem
to fit an early Weeden Island date rather well. Apparently, the Kolomoki





period of southwestern Georgia lasted for a long time at least long enough
to receive some Mississippian influences as evidenced by some of the
vessels in the burial mound cache (Sears, 1951b).

With the above points in mind, I have applied these five radiocarbon
dates to the chart below. This chart applies chiefly to the Gulf coast of
Florida, although the Orange and Preceramic periods are better known in
east Florida.


Table I
CULTURE PERIODS AND RADIOCARBON DATES


Ft. Walton ----1400 A.D.

Weeden Island II

Weeden Island I
_350 A.D.
Santa Rosa-
Swift Creek

Deptfo- 150 B.C.
Deptford

Transitional or
Early Post Orange
1200 B.C.
Orange (decorated)

Orange (plain)
--- --2000 B.C.
Preceramic


The use of the above dates raises one problem not previously mentioned.
For the Orange period to have lasted for eight hundred years seems reason-
able; for the Deptford period to cover all the time from about 1200 B.C. until
150 B.C. does not seem likely. Partially to take care of this difficulty but
chiefly to take recognition of other data, I have included a "Transitional
or Early Post Orange" period between the end of the Orange period and the
beginning of Deptford.

Such a period is the Gulf coast equivalent of Rouse's Malabar I
(Rouse, 1951, p. 70) for the Indian River area and Goggin's "Orange late"






(Goggin, 1952, pp. 36, 47) for the St. Johns River region. Time markers
for the central Gulf coast are St. Johns Incised and Pasco Incised. They
illustrate a transfer of Orange Incised decoration onto chalky or limestone-
tempered pottery respectively.

Evidence for such a period was found at Johns Island near the mouth
of the Chassahowitzka River in Hernando County in 1949. There, in two
tests, these time markers were found in levels below those which produced
Deptford ceramics (Bullen and Bullen, 1950, pp. 30-33). Recent information
from Bayport, a little farther south, supports the Johns Island findings
(Coates, 1955). At Bayport semi-fiber-tempered and various Perico deco-
rated pottery types were also present. I would, at least for the time being,
include these pottery types as belonging to the Transitional period (Bullern
n.d.b). Stratigraphic evidence for the inclusion of semi-fiber-tempered
pottery was found at Site N-9 on Amelia Island (Bullen and Griffin, 1952,
p. 49).

At the moment, the Transitional period includes various, somewhat
divergent, pottery complexes all of which had their start very early in
post-Orange times. With more data it will probably be possible to sub-
divide the Transitional period into various phases. Characteristics of
these phases may vary in different parts of Florida.

Originally,, we had presumed a Transitional or Early Post Orange
period to have been of extremely short duration. The radiocarbon date
from Site J-5 for the introduction of such a period is so early it suggests
the period started earlier, lasted longer, and was more important than we
had realized.





LITERATURE CITED



Bullen, Adelaide K., and Ripley P.
1950. "The Johns Island Site, Hernando County, Florida." American
Antiquity, Vol. 16, No. 1, pp. 23-45. Menasha.


Bullen, Ripley P.
1952. "Eleven Archaeological Sites in Hillsborough County, Florida."
Report of Investigations No. 8, Florida Geological Survey.
Tallahassee.






1954. "Culture Changes during the Fiber-Tempered Period in
Florida." Southern Indian Studies, Vol. V, pp. 25-26, 45-48.
Chapel Hill.


1955. "Archeology of the Tampa Bay Area." The Florida Historical
Quarterly, Vol. 34, No. 1, pp. 51-63. Gainesville.

n.d.a Six Sites near the Chattahoochee River, Florida. Manuscript
on file at Bureau of American Ethnology, Smithsonian Institu-
tion. Washington.

n.d.b Excavations on Cape Haze Peninsula, Florida. Manuscript at
Florida State Museum. Gainesville.


Bullen, Ripley P., and John W. Griffin
1952. "An Archaeological Survey of Amelia Island, Florida." The
Florida Anthropologist, Vol. V, Nos. 3-4, pp. 37-62. Gainesville.


Coates, Gordon C.
1955. "Recent Tests at the Battery Point Site, Bayport, Hernando
County, Florida." The Florida Anthropologist, Vol. VIII, No. 1,
pp. 27-30. Gainesville.

Fairbanks, Charles H.
1954. "1953 Excavations at Site 9 HL 64, Buford Reservoir." Florida
State University Studies, No. 16, pp. 1-26. Tallahassee.


Griffin, James B., editor
1952. Archeology of Eastern United States. University of Chicago
Press. Chicago.

Goggin, John M.
1950. "Florida Archeology 1950." The Florida Anthropologist,
Vol. III, Nos. 1-2, pp. 9-20. Gainesville.

1952. "Space and Time Perspective in Northern St. Johns Archeology,
Florida." Yale University Publications in Anthropology, No. 47.
New Haven.


Rouse, Irving
1951. "A Survey of Indian River Archeology, Florida." Yale Univer-
sity Publications in Anthropology, No. 44. New Haven.







Sears, William H.
1951a. "Excavations at Kolomoki, Season I 1948." University of
Georgia Series in Anthropology, No. 2. Athens.


1951b. "Excavations at Kolomoki, Season II 1950." University of
Georgia Series in Anthropology, No. 3. Athens.




FLORIDA STATE MUSEUM
GAINESVILLE, FLORIDA













EXCAVATIONS AT GROSSMAN HAMMOCK,
DADE COUNTY, FLORIDA

Marvin J. Brooks, Jr.

THE SITE

Grossman Hammock includes nearly 100 acres of high, wooded land in
the east edge of the southern Florida Everglades about 25 miles southwest
of Miami. The excavations described in this report are a few hundred feet
east of the center of Section 25, Township 55 south, Range 37 east, Dade
County, Florida.

The hammock averages from 10 to 12 feet above sea level and is dry
the year around. The surrounding rocky glades are from 7 to 7.5 feet above
sea level and are flooded during much of the rainy season.

This area was included in land purchased by the Grossman family in
1917. When Mr. M. L. Grossman first visited the hammock in 1925, he found
sugar cane, banana trees and pumpkins growing along the east edge. The
Seminole occupation evidenced by this growth must have ended several
years earlier as there were no remains of chickees evident.

The family of William McKinley Osceola lived along the east edge of
this hammock and farmed much of the high ground about 1900 (personal
communication, Sept. 25, 1955).

During 1927 Mr. Grossman visited the families of Jimmy Willy and Willy
Tommy, Seminole Indians, who were living on a small hammock three or
four miles to the west.

During 1926 aid 1927 a road was built to the hammock, and the ham-
mock, as well as much of the surrounding low ground, was cultivated. This
cultivation disclosed evidences of occupation along the east edge of the
hammock. These included pottery fragments, rifle barrels, an iron sewing
machine frame, and part of a human jawbone.

In 1943 and again in 1949, the hammock was visited by companies





exploring and drilling for oil. In 1953 the Mineral Springs Corporation
started development of a recreational resort near the south end of the ham-
mock.

There are rumors of parts of human skulls, cannon balls, almost whole
pots, and mysterious chests having been found here, but none of these re-
ports have been verified. Karl Squires' collection is supposed to contain
artifacts from this site, but the writer has not seen this collection. Dr.
John M. Goggin visited the hammock during the 1940's but did not find the
Indian mounds (personal communication, Feb., 1955). So far as is known,
the excavations described in this report are the first attempt at controlled
archaeological excavations here, although there are numerous evidences of
previous digging by treasure hunters.


EXCAVATIONS

The excavations described in this report are in the northernmost of a
group of three middens. Two of these are along the east edge of the ham-
mock. The third, a small near-circular midden, rises 4 or 5 feet above the
glades east of the hammock.

Midden 1, with which this report is concerned, is about 200 feet long,
75 feet wide, and follows the irregular east edge of the hammock. The
deposit seems to be 12 inches deep or deeper only in a central area hardly
more than 60 feet in diameter and concentrated where the slopes to the old
slough to the east are steepest. Most of this central area has been dis-
turbed by recent truck farming and much of the soil has been removed. All
of the excavations described in this report were made in Midden 1.

Excavation 1

Excavation 1, near the northwest extremity of the midden, was made
before the area had been adequately explored; the site was chosen solely
because of ease of access. During August and September, 1954, a 5-foot-
square pit was excavated in two arbitrary 4-inch levels to blue-gray marl.

Excavation 2

Excavation 2 is about 120 feet south of Excavation 1 and is near the
southeast extremity of the midden. It is in the small area of near-level
ground between the disturbed area and the slope marking the edge of the
midden. It was started in October, 1954, as a 2-foot-square hole. By sub-
sequent addition of 2-foot-square sections, it was finished in January,
1955, as a 6-foot-square pit.





Each of the 2-foot-square sections was excavated in 3-inch levels ap-
proximately parallel to the ground surface. The ground slopes about 3
inches diagonally across the pit from the northwest to the southeast. It
should be noted that the levels excavated, although parallel to the ground
surface, were not parallel to the layers of deposition.

Artifacts disappear at a depth of about 12 inches where a gray marl is
reached. About 2 inches of this gray marl overlies the honeycombed Miami
Oolite. In the old slough to the east of the midden there are several inches
of muck on top of the Miami Oolite, but this marl is absent there. The
upper 12 inches in this pit consist of black midden dirt intermixed with
potsherds and bone and shell fragments. The top 2 to 5 inches were lighter
in color, but this slight difference in color could not be seen during exca-
vation; it was observed only after the excavated bank had been exposed
for a day or longer. This thin zone was observed over an area only some
20 feet in diameter.

Excavation 3

Excavation 3, an irregular pit covering about 9 square feet and located
just west of the disturbed area, was made to test the possibility of gather-
ing worthwhile data from this location. The deposit was only 5 inches
thick here.

Excavation 4

Excavation 4 was a 5-foot-square pit in the disturbed area about 25
feet northwest of Excavation 2. It was made to test the possibility of
gathering data from the disturbed area. This location appears to be near
what was once the highest part of the midden. From 9 to 12 inches of the
deposit had been removed previously and only 4 or 5 inches remained.

Excavations 5, 6, and 7

Excavations 5, 6, and 7 were each 5-foot-square pits along a north-
south line about 25 feet east of Excavation 1. Ten feet of unexcavated
space remains between these pits. The nearest is about 65 feet from Ex-
cavation 2: The deposit at all three pits was thin, although the actual
depth to rock varied from 4 inches to 16 inches. None of these pits pro-
duced many artifacts, and the actual number of both potsherds and non-
ceramic artifacts varied greatly from pit to pit.

ARTIFACTS

Excavation 1 produced 483 undecorated sherds, 5 decorated sherds,






1 blue glass bead, and numerous fragments of unworked bone and shell
but no bone or shell artifacts. All five of the decorated sherds came from
the upper 4-inch level. These included 1 Glades Tooled, 1 unique incised,
and 3 miscellaneous incised. The glass bead was also from this level.
The undecorated sherds included 43 rim sherds and 322 body sherds.

Pottery from the lower 4-inch level consisted of 11 plain rim sherds
and 107 plain body sherds.

The unique incised sherd from Excavation 1 is decorated on both sides
and is curved so slightly as to suggest a plate form rather than a bowl
form. The buff-colored interior is decorated with a combination of incised
lines and dash-like punctations. The black exterior is decorated with
irregular faint dash-like punctations. The sherd is too small to determine
the complete pattern.

The 5835 artifacts from Excavation 2 are tabulated by level in Table
1. No attempt at a complete description of these artifacts will be made,
and only points of special interest or significance will be noted.


Pottery Types of Interest

Unclassified Plain. The unclassified plain rim and body sherds are
made of a paste having a very fine sand temper. Most of them are nearer
to Belle Glade Plain paste than they are to Glades Plain Paste. The rim
sherds were separated into Belle Glade Plain, St. Johns Plain, and un-
classified plain. No attempt was made to classify the more than five
thousand plain body sherds; but they were carefully examined to detect
the presence or absence of St. Johns Plain. None were found that com-
pared to the two St. Johns Plain rim sherds or to the single St. Johns
Check Stamped sherd from the surface collection. About one-half of one
per cent were noticeably softer and contained much less temper than the
average, however, and may represent borderline cases.

Glades Tooled. Five of the Glades Tooled sherds from the 9- to 12-
inch level differ in character from the others of this classification. All
five appear to have been modeled with the fingers rather than with a tool.
One is similar to the sherd described by Willey (1949a) as "Glades Plain
fluted lip" and pictured as C in his Plate 14. The other four have a
thickened lip that is notched both on the top and on the sides. This group,
though listed in Table 1 as Glades Tooled, may more properly fall into that
group with rims "grooved, notched, or lightly crimped" that Goggin and
Sommer (1949, p. 90) suggest "may have led to the development of the
boldly tooled rims of Glades Tooled."









Table 1


VERTICAL DISTRIBUTION OF ARTIFACTS FROM EXCAVATION 2,
GROSSMAN HAMMOCK, DADE COUNTY, FLORIDA

TYPE LEVELS (inches below surface)
TOTALS
CERAMIC 0-3 3-6 6-9 9-12 12

Glades Tooled 15 16 3 6 40
Surfside Incised 3 1 4
Englewood-like or
Safety Harbor-like Incised 5 5
Key Largo Incised 21 29 4 54
Matecumbe Incised 2 3 1 6
Miami Incised 1 1
Dade Incised 1 1
Miscellaneous incised 28 30 18 7 83
Belle Glade Plain, rims 26 13 6 6 51
St. Johns Plain, rims 2 2
Plain rim sherds 152 163 90 69 474
Plain body sherds 1450 1809 1206 591 33 5089

TOTAL CERAMIC 1679 2055 1358 685 33 5810

NON-CERAMIC

Glass beads 6 6
Tubular shell bead 1 1
Fragment of fish spine point 1 1
Bone awls 1 1 2
Shark teeth, perforated 1 2 2 5
Shark tooth, edge worn 1 1
Short bone points 2 2
Fragments of bone points 4 2 6
Socketed bone point 1 1

TOTAL NON-CERAMIC 10 8 5 2 25






Surfside Incised. Two of these sherds do not have the typical incised
line or lines parallel to and below the lip but have only a shallow longitu-
dinal groove on top of the lip. These might be more correctly classified
as Glades Plain, but are listed here as Surfside Incised because they are
of the same temporal period as Surfside Incised (Goggin 1950a).

Englewood-like or Safety Harbor-like Incised. These five sherds from
the 0-to-3-inch level are too small to type exactly. They are pictured in
Figure 1 as E through 1. They are about 3/16 of an inch thick and vary in
color from buff through black. The paste is sand-tempered. The inside of
the sherds is pictured here. They are also decorated on the outside with
fine random crossing incised lines.

Key Largo Incised. The fifty-four sherds in this group vary greatly in
size of pattern and execution of pattern. (Fig. 1, J-0). An attempt was
made, by measuring the diameters of the arches, to discover any temporal
significance in the variations in arch diameter. The only conclusion that
could be reached, however, was that the older sherds showed more varia-
tion in pattern size and in execution of pattern than did the more recent
sherds.

Matecumbe Incised. Typical Matecumbe Incised sherds are pictured
as A and B in Figure 1. One variant (Fig. 1, D) is similar to another such
variant pictured by Goggin and Sommer (1949) as D in their Plate 1.

Miscellaneous Incised. This group includes those sherds with a single
incised line and those too small to classify. Some are no doubt fragments
of known types. One sherd might represent Miami Incised, two appear
similar to Opa Locka Incised, and one group of eight body sherds with two
nearly parallel incised lines might be either Key Largo Incised or variants
of Matecumbe Incised. In these cases, however, the sherds were too small
to allow positive identification. One rim sherd from the 0- to 3-inch level
is suggestive of a Weeden Island influence. The flattened lip is notched
and a single incised line parallels the lip. Other slanting incised lines
terminate at this line. One rim sherd (Fig. 1, C) from the 3- to 6-inch
level is of particular interest as it is similar to one excavated in Hialeah
by Laxson and pictured in an article by Bullen and Laxson (1954) entitled
"Some Incised Pottery from Cuba and Florida." The sherd found here was
in a mixed level containing elements of both Glades II and Glades III. The
sherd from Hialeah was found in a Glades II level.

Non-Ceramic Artifacts

The glass beads are all faceted, some crudely. Two are dull green,
the remainder blue. All are from 3/16 of an inch to Y of an inch in diame-

















G H I
1:4 1!


-l :t 1 1 tt r) 1 ... 3 1, ,


Fig. 1. Decorated pottery from Grossman Hammock.
A-B, Matecumbe Incised; C-D, unique incised; E-F, Englewood Incised;
G, Safety Harbor-like Incised; H-1, Englewood-like Incised; 1-0, Key Largo
Incised.






ter. The straight-sided tubular shell bead is 3/16 of an inch in diameter
by 1 inch in length and is drilled from both ends. The fragment of a fish
spine point is 1 inch long and shows traces of an asphalt-like material.
The bone awl fragments are both 3/8 of an inch wide and are % of an inch
and IV inches long. The shark teeth vary in size from Y by 7/8 of an inch
to 7/8 of an inch by 1% inches. The short (bipointed) bone points are 1'
inches long. The socketed point is 1% inches long. The fish spine point,
the bone awls, and the socketed point are fragmentary.

Artifacts from Excavation 4 are comparable to those from the 9- to 12-
inch level of Excavation 2 with the addition of ten unclassified incised
sherds. These have incised parallel lines and are similar to one found by
Goggin in the Everglades National Park (Goggin, 1950a, Fig. 78, D). Non-
ceramic artifacts noted only in Excavation 4 included two bone pins and
one columella chisel.

Artifacts from the thin deposits of Excavations 5, 6, and 7 are not
tabulated here. They include Glades Tooled, Surfside Incised, Key Largo
Incised, and one doubtful Matecumbe Incised. The distribution, both as
to numbers and types, was erratic from excavation to excavation and from
level to level. Of interest are three sherds with a modeled decoration
similar to some from upper Matecumbe Key (Goggin and Sommer, 1948,
P1. 2, A). These were in a provenience with Key Largo Incised as the
only other decorated type. Non-ceramic artifacts from these three excava-
tions include three Strombus celt fragments and one perforated shell
sinker.

The surface collection from the disturbed area at Midden 1 includes
only one decorated type not represented in the excavations, a single St.
Johns Check Stamped sherd. Non-ceramic artifacts from the surface col-
lection include four Busycon picks, Type A.


DISCUSSION

Excavations were made in a small (75 foot by 200 foot) black dirt
midden where the remaining deepest deposits are little over 1 foot thick.
The smaller area of these deeper deposits as deep as 1 foot seem
confined to the area adjacent to an old slough. Most of these deeper de-
posits have been destroyed by removal of much of the rich soil from the
center of the midden. Only Excavation 2 was in these deeper deposits.
It produced a relatively large number of potsherds and a relatively small
number of non-ceramic artifacts.

As even the "thicker" deposits are actually thin, the stratigraphy






shown is on a gradual percentage change basis rather than on a basis of
sharp separation (Table 1). The glass beads and the Glades Tooled and
Surfside Incised ceramics in the upper level evidence occupation not only
by Seminoles but during the major part of Glades III times.

The next level contains both Glades III and Glades II types, but the
lower levels are clearly Glades II except for the presence of a few Glades
Tooled sherds. Glades IIc, with its predominately undecorated pottery, is
not observed here; but sherd types from both Glades IIb and Glades IIa are
seen: Matecumbe Incised, Miami Incised, Dade Incised, and miscellaneous
incised similar to Opa Locka Incised all accompanied by Key Largo
Incised, the marker for the Glades II period.


The very low percentage of St. Johns Plain (chalky ware) seems ex-
plained only by consideration of the fact that this midden is 17 miles west
of the Atlantic Ocean and some 6 or 7 miles west of the west edge of the
"Miami Rock Ridge." It should be noted, however, that the types marking
the Glades II period in the Calusa sub-area to the west, where St. Johns
Plain is at a minimum, are not represented here either in the excavations
or in the surface collection.


Influence from the northwest during Glades III times is seen, however,
in the five sherds from the upper level that are similar to either Engle-
wood, Sarasota, or Safety Harbor Incised.


With the exception of the excavation in the disturbed area, all the other
excavations were in the thin deposits some distance away from the old
slough. This part of the midden, while occupied during both Glades II and
Glades III, must have been occupied only intermittently; as not only are
the deposits thinner, but even the number of artifacts recovered are less
per volume excavated.




ACKNOWLEDGMENTS


I wish to express my appreciation to D. D. Laxson, of Hialeah, for
invaluable assistance and advice; to Ripley P. Bullen, of The Florida
State Museum, for his advice and for his assistance with typology, with-
out which this paper would not have been possible; and to Mark L. Gross-
man for assistance at the site and for much information concerning the
history of the site.






LITERATURE CITED


Bullen, Ripley P., and D. D. Laxson
1954. "Some Incised Pottery from Cuba and Florida." The Florida
Anthropologist, Vol. VII, No. 1, pp. 23-25. Gainesville.

Goggin, John M.
1950a. "Stratigraphic Tests in the Everglades National Park." Ameri-
can Antiquity, Vol. 15, No. 3, pp. 228-46. Menasha.

1950b. "The Snapper Creek Site." The Florida Anthropologist, Vol.
III, Nos. 3-4, pp. 50-66. Gainesville.

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


Laxson, D. D.
1953. "Stratigraphy in a Hialeah Midden." The Florida Anthropologist,
Vol. VI, No. 1, pp. 1-8. Gainesville.

Willey, Gordon R.
1949a. "Excavations in Southeast Florida." Yale University Publi-
cations in Anthropology, No. 42. New Haven.

1949b. "Archaeology of the Florida Gulf Coast." Smithsonian Miscel-
laneous Collections, Vol. 113. Washington.



MIAMI, FLORIDA













THE TURNER RIVER SITE, COLLIER COUNTY. FLORIDA

William H. Sears

The Turner River site has been known for at least fifty-five years.
Predictably enough, C. B. Moore made a brief visit in 1900 at which time
he collected a notched shell sinker (Moore, 1900, pp. 379-80, Fig. 29).
Hrdlicka published the first real description, which included a sketch of
the site in 1918 (Hrdlicka, 1922, p. 380, Fig. 1). He was so impressed by
the site that he stated: "The site is so characteristic, so easily approach-
ed, and probably so important to science, that steps, it would seem, ought
to be taken to preserve it for posterity, which could best be done by making
it a national reservation. The expense of this at present would be insignif-
icant; and little time should be lost in having it carefully surveyed, which
could be done with no great cost or difficulty at a time when the mosquito
pest abates in some measure" ibidd., p. 39). We can only agree whole-
heartedly with this statement and wish that more attention had been paid
to it twenty-three years ago.

The site is now owned by Mr. Ted Smallwood, of Everglades City.
In May, 1955, Mr. Karl A. Bickel, hearing that the potentially valuable
shell of which the mounds are composed might be sold, arranged for a brief
inspection of the site. The party included himself and the Curator of Social
Sciences of the Florida State Museum, Mr. Ripley P. Bullen. As a result
of this inspection, which brought new realization of the importance of the
site, the writer went to the site in October, 1955, representing the Depart-
ment of Social Sciences of the Florida State Museum. One week of work
with several laborers produced data which enable us to understand, to some
degree, the real position of the site in Glades area archaeology.

Practically speaking, the site is defined by the thirty or more mounds
which are its chief characteristic and by the relatively small spaces be-
tween the mounds. The spatial arrangement of these features can best be
understood by reference to Figure 1.

High tide debris and mud indicate that the river inundates the area be-
tween the mounds for some distance inland, perhaps to halfway up between
those in the row closest to the river. Presumably this occurs only during
the spring high tides at present.
















Fig. 1. The Turner River Site. A-E, locations of tests. (Contour map
adapted from original prepared by E. R. Thomasson, surveyor,
of Naples. Used here by courtesy of the Turner River Shell
Corporation.)


















~csc%










/a4




~~If





The development of ridges running out into the water appears to be a
characteristic of the larger sites in this area. Cushing's map of Key Marco
(Cushing, 1896, P1. XXX) documents the existence of this feature there very
well. He also demonstrates that in some areas of the site, at least, the
ridges continued out well under modern water levels ibidd., p. 351 and Pl.
XXXI). A somewhat similar set of features seems to have been present at
Chokoloskee Island, directly opposite the mouth of Turner River. Traces of
this feature are evident on Chokoloskee today and seem to have been quite
prominent in 1900 (Moore, 1900, pp. 9-13).

Figure 2 gives the general location of the site with respect to others
described in the literature for the Glades area.

The Chokoloskee Bay area appears to have been an important prehistor-
ic population center, characterized as it is by the Turner River site, Cho-
koloskee Island itself, and Key Marco, all of them very large sites.


71kf~


TA C
* *^


Fig. 2. Major archaeological sites in the Glades area of southern Florida.






EXCAVATIONS


Five test pits were excavated at various points in the site. Their
locations (Fig. 1, A-E) were selected with the following considerations
in mind:

1. Possible age differences between different parts of the site.
2. Estimated depth of refuse.
3. Availability. In a short period, we could not excavate in areas
requiring extensive clearing nor in those a long distance from
the boat landing. To work in most of the distant areas we would
have had to cut trails.

Test A 5 feet x 10 feet, excavated to 18 inches in 6-inch levels. Water
at 18 inches. The profile, Figure 3, outlines the situation. The shell
basal layer extends at least one foot deeper.

Test B 5 feet x 10 feet. Test initiated in an attempt to determine some-
thing of the nature, function, and structure of the mounds. The excavation
was closed out as the walls began to collapse at 5 feet. To that depth we
had encountered only clean shell, almost entirely small oyster shells. A
few fragments of conch and Venus shells were observed.

Test C 5-foot x 10-foot pit, excavated in 6-inch levels. Sherds were
relatively plentiful in the top 6 inches of black humus. At 6 inches, shell
began to appear. At 12 inches, the deposit was pure shell. A few sherds
were found more or less on top of this shell. The pit was excavated to a
depth of 5 feet, through clean shell for the entire distance below 18 inches.
The water table was reached at 5 feet.

Tests D and E The stratigraphy in these two tests was identical, al-
though Test E had a larger number of sherds. In both cases (see Fig. 1
for locations) we found 14 inches to 18 inches of black muck midden which
had a high pottery, bone, and charcoal content. Shells were, for this site,
quite rare, although Test D was only a few feet from the foot of two mounds
and Test E only a few feet from the base of one of the two.

At 18 inches, two facts became apparent: 1. We had reached the cur-
rent water table. 2. The black muck changed to a sandy gray marl. The
marl appeared to be a sandy waterlaid deposit except that it contained ovoid
or lenticular deposits of midden material to an undetermined depth, observed
only to 1 foot below the water table.

Since the marl is a normal waterlaid marine deposit, tidal flat mud, the
miscellaneous debris which formed the small pockets of midden was being






TEST A

SOUTH WALL EAST WALL







SMUOK-IDDEN I
5'
GRAY MARL

SHELL

Fig. 3. Profile of south and east walls, Test A.

dropped into the water from some sort of a structure elevated above the
normal water level.

A similar story of deposition in water is rather plain in Test A (Fig. 3).
The base of the shell deposit there was too far below the base of the pres-
ent water table for us to find it. However, the gray marl which seems to be
a marine deposit runs up over part of the shell.

I see no way, then, to escape the conclusion that the shell deposit was
under normal high tide level at some point after it was made and before the
black midden containing cultural debris was deposited. It is also certain
that the shell was deposited by man and is not some sort of natural feature.
It is too uniform in its clean, small oyster-shell content, which duplicates
precisely the constructional material of the mounds, to admit of any form of
natural deposition.

Generally, my impression from these tests, from Chokoloskee where
dredging and road cutting aregoing on, and from Cushing's description of
phenomena observed at Key Marco (Cushing, 1896) is that most of the area
of these extensive sites is composed of clean shell deposits placed on
submerged and still-growing mud flats. This is not to imply that, neces-
sarily, dry land or dry areas were being deliberately created. On the con-
trary, the population with its dwellings appears to have kept moving sea-
ward, away from the growing elevated areas. Goggin makes this observation
in connection with Matecumbe Key and Goodland Point (Goggin and Sommer,
1949, pp. 86-87; Goggin, 1949, pp. 89-90), at least insofar as he documents
the movement of a population seaward from a center.

Another phenomenon observed at the Turner River site is that midden






areas appear to be confined to the low spots between mounds and to the
basal portion of mounds. On every mound observed, sherds and shell
tools may be found on the lower slopes, perhaps extending upwards 4 or
5 feet. From there on up to the tops, only clean shell can be found on the
surface or in the test. This is also true in the walls and spoil piles of pits
which had been previously dug into ten or fifteen of the twenty or thirty
mounds observed.

It is, of course, quite possible that layers of occupational debris do
occur in these miniature mountains. No such were observed, however; and
my impression is, after digging holes, climbing over the mounds, and look-
ing in older pits and cuts at Turner River and Chokoloskee, that any such
layers which do exist must contain relatively minor proportions of the oc-
cupational debris occurring at the Turner River site.



ARTIFACTS

Figures 4 and 5 give the distribution of ceramics from the several tests.
It may be noted that only Tests C and E yielded samples large enough to be



0' O i h. E


I I a t
Sa Ma




8 C 4 T T

I S IS I 7 T7

P I 8 I I I10 102
3 66 C' oa
A 24 I 2s1 a
D 2 B 3 S 7 a2 mT
a 8 2 3a 10
I ll 30 2 I I 172 2 1T4
2 I 1 5 s 2 I 7 I 4 I I 2 410 2 413

4 6 6 1 I 0 T T 4
DEFRMeI 8 a a 7 o -o a884

Reference Key: 1-Goggin and Sommer, 1949, pp. 33-34; 2-Goggin, 1949, pp. 76-77;
3-Goggin and Sommer, 1949, pp. 36-37; 4-Goggin, 1949, p. 74; 5-ibid.; 6-Goggin,
1950, p. 240; 7-Goggin and Sommer, 1949, pp. 37-38; 8-Goggin, 1949, p. 77; 9-
Griffin, J.B., 1945, p. 220t 10-Goggin, 1948.
Fig. 4. Occurrence of pottery by test and level at Turner River Site.







ORS CANME TI[
PASS SANI-EL PATOH DOOUDiA M
INCISOD I NISED INCIED REOD bRED Ui SAMPLE


Fig. 5. Bar graph showing proportional frequency of various pottery types in each level of Tests
C and E. Plain ware frequencies based on all plain sherds as 100 per cent, decorated
type frequencies based on all decorated sherds as 100 per cent.






considered truly representative of the ceramic complexes in their respective
areas of the site.

By far the greater number of our sherds are clearly representative speci-
mens of well-known and adequately described types. References to the
original type descriptions are included in Figure 4. A few words of com-
ment on unclassified specimens may be appropriate here.

Unclassified Punctate. There are two sorts of these sherds, both from Test
E, Level 4. Sherds of each are illustrated in Figure 6. One variant, G and
H, is a sort of miniature Sanibel Incised. The rows are composed of closely
spaced, very short dashes placed at right angles to the row. The other sort
(Fig. 6, 1) has a particularly Weeden Island-like appearance with its rows
of circular punctates running down from the rim and with the rim set off by
the encircling row of small angular punctations below it.

Englewood-like Incised. Nine sherds (Fig. 6, K), from Levels 2 and 3 of
Test D, have a rather soft paste tempered with fine sand. Decoration con-
sists of straight, fine, incised lines bordered by punctations. They seem
to be close to Englewood Incised in execution, but cannot properly be clas-
sified under that type. It is not believed that this resemblance had any
cultural or chronological significance for the Turner River site.

Weeden Island Incised. One sherd from Test E, Level 2, is an incised
sherd with an odd sandy paste. Incision is fine, with small punctations
spaced more or less equidistantly along it.

Shell Tools. Shell tools found in the excavations are classified below,
following Goggin (1949, pp. 77-8 1).

Busycon hammer C ............ Test C, Levels 1-5, 1 each level
Test B, 1 specimen
Busycon hammer, fragment. ....... Test E, Level 1, 1
Fasciolaria hammer, long ....... .Test E, Level 2, 1
Fasciolaria hammer, fragment ..... Test E, Level 1, 1
Busycon tool, fragments . ... Test A, Levels 1 and 2, 1 each level
Test B, 1
Test C, Level 1, 1

Figure 5, which depicts graphically the relative importance of the
various ceramic types at various levels in Tests E and C, outlines the
evidence for cultural change at this site quite well. A number of cultural
chronological points are apparent. Sequence within the time span repre-
sented by the Test E midden is indicated by the decrease in relative im-
portance, reading from the bottom up, of Sanibel Incised; its replacement









































JK











M N

Fig. 6. Sherds from the Turner River Site.
A-B, Key Largo Incised; C, Glades Plain fluted rim; D-E, Gordons Pass
Incised; F, Sanibel Incised; G-I, unclassified punctated; J, Fort Drum
Punctated; K, Englewood-like Incised; L-N, unclassified incised.





by Gordons Pass Incised to a considerable extent in Level 3, where Key
Largo Incised also appears; Key Largo Incised gaining at the expense of
the other two types in Level 2; and the final dominance of this type in
Level 1.

Due to the relatively small number of decorated sherds, this trend can-
not be presumed to have been documented absolutely, but it is indicative.
It seems to fall in line rather well with the sequence worked out by Goggin
(1949, p. 85) at Goodland Point for the Sanibel-Gordons Pass duo.

The changes in the decorated ware in Test E are not accompanied by
any significant changes in the relative importance of the two plain ware
types. A change in plain ware frequencies is immediately obvious, how-
ever, if we look at Test C, located at the other end of the same mound.
Here we find a top level with 78.7 per cent of Glades Plain. Levels 2 and
3 reverse this with the Glades Plain dropping to 36.4 and 33.7 per cent, the
Goodland Plain type increasing proportionately. Clearly Level 1, although
the sample is too small for many decorated sherds to have occurred, is
representative of the Test E complex. Below this, and hence earlier, we
have a plain-ware level with Goodland Plain as the dominant type.

My interpretation is that, in this area of the site at least, the more
recent occupation is toward the river, and site growth is along the shell
ridges toward the water. The ridges are actually, of course, built up as
the village moves constantly toward the river which is the dominant factor
in the aboriginal adaptation to the ecology of the area. As noted above,
similar movement toward the water, and into it, has been documented at
Key Marco (Cushing, 1898), Goodland Point (Goggin, 1949, pp. 86-87), and
at Matecumbe Key (Goggin and Sommer, 1949, pp. 89-90).

THE TURNER RIVER SITE AND THE GLADES AREA
CULTURE SEQUENCE

It is, I believe, quite clear that the large sherd collection from Test E
is predominantly representative of the Key Largo complex (Goggin and
Sommer, 1949, p. 80) and consequently of the Glades II period. More pre-
cise placement than this in the chronological framework for the area is
possible, although a bit difficult. This is due, no doubt, to the very large
size of the Glades area and to the fact that, while the area seems remarkably
consistent internally, there is regional variation in the types used as period
markers. For example, we lack completely Miami and Matecumbe Incised,
Glades II markers farther southeast.

However, by jumping around a bit and by matching what we do have with
the available data, some relationships are clear. The top level in our Test





E, with Key Largo Incised as the dominant decorated type, would be Glades
IIb, Matecumbe I in its closest relationships (Goggin and Sommer, 1949, pp.
82-83; Goggin, 1950, p. 245) even though the typical Matecumbe Incised is
missing completely. The second and third levels, where Gordons Pass In-
cised replaces Key Largo Incised, fall in the Glades IIa period (Goggin,
1950, p. 233, Table 36, and p. 245). Apparently the next level down repre-
sents a poorly known early Glades Ha level similar to that found in the
Bear Lake midden. Sanibel Incised is dominant here, a type believed on the
basis of other evidence to be ancestral to Gordons Pass Incised (Goggin,
1950, p. 233-34). Quite clearly, then, the collection from the lower part of
Test C is Glades I, perhaps late Glades I in view of the single sherd of
Cane Patch Incised.

There are no traces in our collections of any later occupation. Surfside
Incised, a Glades IIIa marker (Goggin and Sommer, 1949, pp. 80-98), is com-
pletely absent from our collections, as are check-stamped and Glades Tool-
ed sherds which would, if present, represent the later Glades III levels.


OTHER COLLECTIONS

Several other collections from the site are known and/or are available
to us. William Plowden (personal communication) made a surface collection
in 1954. He has only Glades Plain and Key Largo Incised, indicative of the
same Glades II period. His collection came from an area at the river edge.

Mr. Hilton Leech made another collection well to the rear of the site,
near its southeastern edge. This collection includes Glades Plain and a
number of unclassifiable sherds that are more similar to Miami Incised than
to anything else, although certainly not so classifiable (Fig. 6, N). This
collection, too, seems in line with our late Glades I-Glades II dating for
the site.

Mr. Hilton Leech and Mr. Karl A. Bickel, of Sarasota, visited the site
again in March, 1956. They collected in two areas at the northern, up-stream
end of the site. Pottery types found are restricted to the same Glades II
complex as found in our test, with Key Largo Incised, Gordons Pass Incised,
and Sanibel Incised. This, then, corroborates the Glades II dating for the
entire site.


SUMMARY AND CONCLUSIONS

The Turner River site, covering some thirty acres with its thirty or
more linear mounds and other features, is predominantly a Glades II period






site. Its beginnings go back into the Glades I period in the area farthest
from the river (towards the south), but it does not extend into the Glades III
period at all on the basis of the available evidence. This gives us a dating,
following Goggin's sequence for the Glades area, of from a century or two
B.C. until some time around eight or nine hundred A.D., allowing for the
fact that the Glades IIc period, last of the Glades II sub-divisions, may
also be absent (see Goggin, 1952, p. 36, Fig. 2).

My own impression, based on these few sherds, is that the decorated
wares of the Glades Ia and IIb periods have a general Weeden Island feel
and that perhaps the extension of Glades II back into a time period equiva-
lent to the Santa Rosa-Swift Creek horizon is overstretching a bit. However,
this "feeling" is only that and can be substantiated or dismissed only on
the basis of more adequate data.

A point which may be of some significance in our ultimate understand-
ing of this site and of other large sites in the area is the apparent growth -
movement really toward the water, with some indications of not only water-
edge, but over-water, habitation. The indications of growth toward the
water, substantiated by evidence at Key Marco, Goodland Point, and Mate-
cumbe Key, are perhaps not too surprising in view of the environment and
the ecology of the area, including the rather obvious fact that the sea pro-
vided the bulk of the food. To a people adapted to such an area, dry land,
which is at a minimum even today, might well be considered as much of a
nuisance as anything else, certainly not an important factor culturally. In
any event, the indications are that they were moving constantly away from
the higher areas, including those which they created, and toward the water
which supplied their food.





LITERATURE CITED

Cushing, Frank H.
1896. "Exploration of Ancient Key Dwellers' Remains on the Gulf
Coast of Florida." Proceedings, American Philosophical Society,
Vol. 35, pp. 329-448. Philadelphia.

Goggin, John M.
1948. "Some Pottery Types from Central Florida." Gainesville An-
thropological Association, Bulletin Number 1. Gainesville.

1949. "Cultural Occupation at Goodland Point, Florida." The Florida
Anthropologist, Vol. II, Nos. 3-4, pp. 65-91. Gainesville.





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

1952. "Space and Time Perspective in Northern St. Johns Archeology."
Yale University Publications in Anthropology, No. 47. New
Haven.

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

Griffin, James B.
1945. "The Significance of the Fiber-Tempered Pottery of the St.
Johns Area in Florida." Journal of the Washington Academy
of Sciences, Vol. 35, No. 7, pp. 218-23. Menasha.

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

Moore, Clarence B.
1900. "Certain Antiquities of the Florida West Coast." Journal of
the Academy of Natural Sciences of Philadelphia, Vol. 11, pp.
349-94. Philadelphia.


FLORIDA STATE MUSEUM
GAINESVILLE, FLORIDA













TEST RESULTS AT THE PAW PAW MOUND,
BREVARD COUNTY. FLORIDA

Carl A. Benson

The Paw Paw Mound is a shell midden located on the east bank of the
St. Johns River in Brevard County, approximately 2V miles southeast of
Highway 50. The site is accessible by water in dry seasons when channel
boundaries in the river are visible. A convenient slough branches off from
the main channel and carries the water-traveler up to the midden in all
seasons except for severe dry spells.

The midden is owned by the heirs of the late Dr. George Crofford of
Orlando, Florida. In previous years Dr. Crofford cleared the site of many
cabbage palms and rearranged the surface for farming. However, no great
changes were made and the profile is approximately the same as it was
when abandoned by aborigines. Publications studied by the. author regard-
ing this region contain no mention of the Paw Paw site.

The shell midden measures 270 feet north and south and 291 feet east
and west. The highest part, located on the west side of the mound, rises
about 11 feet above the water. This high ridge slopes off towards both
ends to form a lower ridge around the site. This results in a low pocket of
muck-like soil in the center of the site. There guava trees are now growing.

Two stratigraphic tests were dug in the Paw Paw midden. Cultural
materials were removed by 1-foot levels. Results are presented in Table 1.
Selected sherds are illustrated in Figure 1.


TEST I

The first test pit, 5 feet square, was dug to a depth of 5 feet in the
southeastern slope of the highest ridge. Solidification of shells hindered
digging to a greater depth.

As shown in Table 1, St. Johns sherds predominated in the first four
levels. In the fifth level, Orange series sherds outnumbered St. Johns






Table I
VERTICAL DISTRIBUTION OF POTTERY AT THE PAW PAW MOUND




a 3

22 39
0 'a














3 .32 1 a

4 18 1 1
5 4 1 6 9
a r-a





TEST I








1 3 36
2 3928 1

3 3








4 18 10

5 194 1 6 9




7 1 36 10

8 13 42
3 35

4 10













10 10






























C D


Fig. 1. Sherds from Paw Paw shell midden.
A-B, St. Johns Check Stamped; C, Tick Island Incised; D-H, Orange
Incised; I, Orange Plain.





Plain sherds about four to one. In a somewhat intermediate or transitional
position were a few sherds of chalky paste which, by the presence of vari-
ous holes, suggested the inclusion of some fibrous material.

Various animal and fish bones, as well as snail and mussel shells,
were also found in Test I. Shells were mixed with a large amount of black
dirt. Occasionally, thin strata of charcoal and burned shells were also
present.


TEST II

The second pit, 4 feet square, was excavated to a depth of 10 feet. It
was located at the top of the high ridge.

The first level produced three St. Johns Check Stamped and many St.
Johns Plain sherds (Table 1). Also in this level was a columella chisel.
The second through the fifth level produced about the same material except
for the lack of any St. Johns Check Stamped pottery. In the sixth level was
a chert chip, a St. Johns Incised sherd, the first Orange series pottery, and
nearly the last of the St. Johns Plain sherds. Interestingly, the St. Johns
Incised sherd contained various holes suggesting the inclusion of some
fibrous material. Ceramics of the Orange period dominated the lowest four
levels.

Fragments of food bones were scattered throughout the debris. These
included those of turtles, alligators, fish, and warm-blooded creatures such
as raccoons, deer, and birds. A possible suggestion of cannibalism was
found in Level 8. It consisted of the front portion of a human mandible en-
cased in cemented fragments of shells. No other human bones were found.
The last foot of the test was excavated under water.


SUMMARY

Tests at the Paw Paw Mound indicate occupation by aborigines during
the early fiber-tempered or Orange period and during a later period when
pottery was made of a chalky or St. Johns paste. The lowest levels of
Test II may belong to a plain fiber-tempered period which was followed by
one during which pottery was decorated by incising. In Test I, a few semi-
chalky, semi-fiber-tempered sherds are suggestive of an intermediate period
between Orange times and St. Johns times. Such a period is also suggest-
ed by the one St. Johns Incised sherd from Level 6 of Test II. This level
should correlate with Rouse's Malabar I period (Rouse, 1951, pp. 244-47;
Goggin, 1952, pp. 44-45).






Higher levels in both tests, except for Level 1 of Test II, would appear
to represent a time when St. Johns pottery was made but not decorated. To
the north Goggin calls such a period St. Johns I (Goggin, 1952, pp. 47-53)
while to the south Rouse calls it Malabar I'. The highest level of Test II
with St. Johns Check Stamped sherds is representative of St. Johns II or
Malabar II period. Surface collecting produced many more St. Johns Check
Stamped sherds.

Thus, our small tests may have spanned five archaeological periods.
They document for this area the accepted chronological sequences previ-
ously determined for regions a short distance to the north and to the south
Goggin (1952, p. 45) has commented on the lack of detailed stratigraphic
data for the St. Johns River area. It is hoped these tests at the Paw Paw
Mound will be helpful in partially filling this need.


ACKNOWLEDGMENTS

I wish to express appreciation to Mr. Ripley P. Bullen, Curator of
Social Sciences at the Florida State Museum, whose encouragement and
enthusiastic help made this paper possible; to Mr. Louie Albritton, Jr.,
who donated his photographic abilities; to Mr. Robert Amos, who assisted
with the excavation of the first pit; and to my wife, Sara, whose help in
cleaning and cataloging of specimens is greatly appreciated.


LITERATURE CITED

Goggin, John M.
1952. "Space and Time Perspective in Northern St. Johns Archeology,
Florida." Yale University Publications in Anthropology, No.
47. New Haven.

Rouse, Irving
1951. "A Survey of Indian River Archeology, Florida." Yale Uni-
versity Publications in Anthropology, No. 44. New Haven.

ORLANDO, FLORIDA














CONTRIBUTORS TO THIS ISSUE


Ripley P. Bullen, curator of social sciences at the Florida State Museum
and for six years treasurer of the Florida Anthropological Society, is well
known to our readers. The paper on radiocarbon dates in this issue was
originally given as a paper at the Annual Meeting of the Society at Rainbow
Springs, February 25, 1956.


Marvin J. Brooks, Jr., is secretary of the Florida Anthropological Society.
He is doing archaeological digging in the Miami area.


William H. Sears is assistant curator of social sciences at the Florida
State Museum in Gainesville. He came to Florida from Georgia, where he
dug the famous Kolomoki site.


Carl A. Benson, a member of our Society, originally presented this paper
on the Paw Paw Mound as a television program from Orlando, Florida.





FLORIDA ANTHROPOLOGICAL SOCIETY

OFFICERS



President: Charles H. Fairbanks, Tallahassee


First Vice President:


Second Vice President:


Secretary:


Treasurer:


Editor:


Executive Committeemen:


John W. Griffin, St. Augustine


J. E. Dovell, Gainesville


Marvin J. Brooks, Miami


Ripley P. Bullen, Gainesville


Adelaide K. Bullen, Gainesville


Wilfred T. Neill, Silver Springs


H. James Gut, Sanford

Charlton W. Tebeau, Miami


NOTICES


Membership in the Florida Anthropological Society is open to everyone
interested in its aims. Dues are $3.00 per year. Student Membership $1.50.
Members receive the Florida Anthropologist, the Newsletter, and other pub-
lications of the Society. Applications should be sent to the Treasurer, who
should be addressed also concerning receipt of publications. His address
is 103 Seagle Building, Gainesville.

General inquiries concerning the Society should be addressed to the
Secretary.

Manuscripts and publications for review should be sent to the Editor at
the Florida State Museum, Gainesville.

Address items for the Newsletter to the President, Department of
Anthropology and Archaeology, Florida State University, Tallahassee.





LIST OF PUBLICATIONS

FLORIDA ANTHROPOLOGICAL SOCIETY


The Florida Anthrop<
Vol. I, Nos. 1-2, 3-4
Vol. II, Nos. 1-2, 3-4

Vol. III, Nos. 1-2, 3-
Vol. IV, Nos. 1-2, 3-


ologist
Vol. V, Nos. 1-2, 3-4
Vol. VI, Nos. 1, 2,3,4

4 Vol. VII, Nos. 1, 2, 3, 4
-4 Vol. VIII, Nos. 1, 2, 3, 4

Vol. IX, Nos. 1, 2
Each Single No. Each Double No.


Members $ .50 $1.00
Non-Members .75 1.50

Newsletter, Florida Anthropological Society

Nos. 1-33 ........................15 cents each

Publications, Florida Anthropological Society

No. 1 "Two Archaeological Sites in Brevard County, Florida,"
by Hale G. Smith. 32 pages, 4 plates . .50 cents

No. 2 "The Safety Harbor Site, Pinellas County, Florida,"
by John W. Griffin and Ripley P. Bullen. 42 pages, 4
plates ...........................50 cents

No. 3 "The Terra Ceia Site, Manatee County, Florida," by
Ripley P. Bullen. 48 pages, 7 plates . 50 cents

No. 4 "The European and the Indian," by Hale G. Smith. 150 pages,
frontispiece, 6 maps ..................... $2.00


All publications may be ordered from the Treasurer:

Ripley P. Bullen
103 Seagle Building
Gainesville, Florida




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