|
4ii
CONTRIBUTIONS
OF THE
FLORIDA STATE MUSEUM
SOCIAL SCIENCES
Number 5
TWO WEEDEN ISLAND PERIOD BURIAL MOUNDS, FLORIDA
William H. Sears
UNIVERSITY OF FLORIDA
Gainesville
1959
S.QINS COLLEGE
LIBRARY
CONTRIBUTIONS
OF THE
FLORIDA STATE MUSEUM
SOCIAL SCIENCES
Number 5
TWO WEEDEN ISLAND PERIOD BURIAL MOUNDS, FLORIDA
The W. H. Browne Mound, Duval County
The MacKenzie Mound, Marion County
William H. Sears
\
-^
UNIVERSITY OF FLORIDA
Gainesville
1959
ii I
",
~ruwsas*lrur~
PREFACE
This issue of the Contributions of the Florida State Museum
records the excavation of two previously undisturbed burial mounds.
This is the first time that undisturbed burial mounds in Florida
have been carefully excavated using modern techniques. Such
structures are rapidly disappearing from the Florida landscape and
we are fortunate in having the opportunity to record and present
this data from the past.
Because of the undisturbed nature of the 8rowne and Mac-
Kenzie mounds, Dr. Sears has been able to reconstruct in great
detail the building of these mounds and the order of burial and, to
some extent, to understand the attendant ceremonies. He brings
stimulating ideas to the controversial question of interpretation.
To gain an understanding of regional and chronological varia-
tions, we need many more examples of carefully recorded field
work of undisturbed, or nearly undisturbed, burial mounds. Let us
hope we can salvage more of this story before it is too late.
Ripley P. Bullen
Curator of Social Sciences
CONTENTS
Page
INTRODUCTION ................... .. ........ 1
BROWNE BURIAL MOUND................ ..... .... 3
Excavation ........ ........... ....
Burials. .. .. .. .............
Pottery ..... ............ .......
Non-ceramic Artifacts ................
Mound Construction and Sequence .........
Test Pits in Village Shell Middens ........
Village Site Chronology . . ......
........
........ 11
. .. .. 14
. . 14
MACKENZIE BURIAL MOUND....................... 18
Excavation ... ...... ............. ... ....... 19
Burials ....... ............ .......... ...... 22
Pottery............ .... ....... . . . .23
Non-ceramic Artifacts. ................ . . 27
Mound Construction and Sequence. .. . . . ..... ..28
SUMMARY AND CONCLUSIONS. .. . . . ..... ..31
REFERENCES CITED. ........... ....... ......... 36
LIST OF FIGURES
Page
1. Map of Florida Locating the Browne and MacKenzie Mounds. . 2
2. Map of Site Du-62 ................. ... ........... 5
3. Excavation Plan, Browne Mound. .. . . ..... ... 7
4. Cultural Sequences in the Lower St. Johns
and Neighboring Areas. ...... . . .. . .16
5. Apparent Changes through Time in Major Ceramic Types at the
Mouth of the St. Johns River. . . . ..... 17
6. Excavation Plan, MacKenzie Mound. . . . . .20
7. Profile, MacKenzie Mound .........................21
LIST OF PLATES
I. Sherds from Du-62 and the Browne Mound
II. Sherds from Du-62 and the Browne Mound
III. Artifacts, Plain Sherds, and Burials, Du-62 and the Browne Mound
IV. Aerial View of Bird Island
V. Small Vessels from the MacKenzie Mound
VI. Small Vessels and Pottery Deposit Sherds, MacKenzie Mound
VII. Sherds from Mound Fill and Pottery Deposit, MacKenzie Mound
VIII. Weeden Island Series Sherds from Mound Fill and Pottery Deposit,
MacKenzie Mound
IX. Shell Beads, Projectile Points, and Pottery Deposit, MacKenzie
Mound
INTRODUCTION
In 1957 the writer had the unusual opportunity of excavating,
for the Florida State Museum, two small burial mounds within a
four-month period. Both structures were the low, amorphous sand
mounds which are quite common in Florida but which, when first
visited by professional archaeologists, usually look like the sites
of gopher colonies. Most fortunately, neither of these mounds
showed any signs of previous disturbance.
One, herein named the Browne mound, is located at site Du-62
near the mouth of the St. Johns River. It was tested as part of our
1955 program on the W. H. Browne tract (Sears, 1957). The other,
called the MacKenzie mound, is situated on a small island in Lake
Weir in the lake region of central Florida. Figure 1 gives the loca-
tions of both mounds.
Surface material from village middens associated with the
mounds and pottery included in them indicate that both are assign-
able to the general St. Johns IIa or the late Weeden Island periods
as these have been defined for peninsular Florida. Acceptable
dates would fall somewhere in the A.D. 800 to 1200 range. This
contemporaneity gives us an opportunity to compare mortuary prac-
tices and the implied ceremonialism in this period for two distinct
areas, the Northern St. Johns and the southern end of the Central
Florida area.
In this report each mound and its contents are described sepa-
rately. In each case there is some information on associated vil-
lage midden materials. The concluding section of the report offers,
very speculatively, some interpretations based on data from the
two mounds individually and jointly.
We owe our particular thanks to Mr. W. H. Browne for permis-
jsion to excavate the mound, now named after him, on his land. He
'also assisted in the excavation and in the surveying necessary to
map all of site Du-62.
TWO WEEDEN ISLAND MOUNDS
Fig. 1. Map of Florida locating the Browne and MacKenzie mounds.
The MacKenzie brothers of Ocala, who were in the process of
developing Bird Island in Lake Weir for real estate purposes, con-
tacted us to suggest that we excavate a mound on the island before
it was necessarily removed or disturbed. I would like to thank them
for their wonderful co-operation, material assistance, and many
courtesies. I would also like to thank Mr. A. Willis and Mr. and
Mrs. R. V. Ott. They suggested to the MacKenzies that they con-
tact us, and their many visits to the site during excavation were
very welcome.
Charles R. Solt, of the Museum staff, drew all figures except
Figure 4.
BROWNE BURIAL MOUND
A single 5-foot-wide trench was run entirely through the
Browne burial mound from east to west in 1955, and two test pits
(Fig. 2, 1-2) were dug in nearby shell-midden deposits. One of
these tests yielded 126 sherds, chiefly of the Colorinda complex
(Sears, 1957, p. 29). The other produced 114 sherds, predominantly
of an earlier period characterized by a sand-tempered plain ware,
although Colorinda Plain appeared at higher levels (Sears, 1957,
pp. 18-20; Fig. 9-10).
The trench through the mound revealed shell layers around the
mound periphery, a number of ourials, and the probable location of
the old ground surface on which the mound was built. A profile
summarizing this information was published (Sears, 1957, Fig. 8).
Artifacts from the mound consisted only of sherds, ranging
from a limestone-tempered version of Early Swift Creek Compli-
cated Stamped to Swift Creek II Complicated Stamped and Weeden
Island series specimens. All came from mound fill. Most of the
burials encountered seemed to have their upper surfaces not more
than 18 inches below the surface of the mound.
In two weeks of work in 1957 we completed the excavation of
the Browne mound, dug six more test pits in shell-midden areas,
and completed a topographic map of the site (Fig. 2).
EXCAVATION
Since it seemed probable that most burials were high in the
mound, we shifted in 1957 to a stripping technique designed to
work out the burial pattern in maximum detail. Whenever possible,
burials were cleaned in place and left for final simultaneous plot-
ting and photographing. The few sherds found were kept sorted by
square and level, although, as it turned out, this was not signifi-
cant. All burials, except those at the bases of multiple bundles
TWO WEEDEN ISLAND MOUNDS
and the deeper ones in the central mass burial, were simultane-
ously plotted this way.
Some portions of the total structure, such as areas around
larger trees, were left unexcavated (Fig. 3). Further work around
the periphery of the mound would have been possible and would
have exposed other segments of the shell ring and perhaps an-
other burial or two. Since the course of the ring, or layer, appeared
predictable, it did not seem probable that the return of information
would warrant the additional expense.
Vertical control was achieved by the use of a transit, tied in
to an arbitrary elevation point. The same point was used for con-
struction of the site map.
BURIALS
In most cases, burials consisted of soft, splintered, badly
fragmented bones. Observation would usually indicate the burial
type. Positions of skulls and of long bone fragments are indicated
in the excavation plan (Fig. 3) except for the central pit. There it
was not possible to leave all of the bones in place after cleaning
or to illustrate all of those which were cleaned. Burials 1 through
6 were excavated, in whole or in part, in 1955 (Sears, 1957, p. 17).
Comments concerning some burials follow:
Burial Comment
1 Skull and long bones from bundle Burials 21-2 lb.
2 Probably part of multiple bundle Burials 15 and 15b. (Not extended on
back as stated in 1955 report.)
3 Adult, male, part of central mass interment.
4 Part of Burial 21.
5 Part of central mass interment.
6 Female, adult, extended on back, shell ring covered body.
7-11 Numbers assigned to bone scrap concentrations found among burials in
northeastern quadrant of mound.
12 Female, young, extended on back, in and on top of shell ring.
Red ochre burial. Long bones, parts of pelvis from one adult, no skull parts
found, mass of red ochre, quantity of several bushels, placed over bones.
BROWNE BURIAL MOUND
13
14
15-15b
Adult, probably bundle burial.
Adult, bundle burial.
Multiple bundle burial of two adults. Burials 2 and 20 were either parts
of this bundle or separate bundles placed in a pile with Burials 15-15b.
FELT
Fig. 2. Map of Site Du-62. Large hachured area is burial mound, smaller
ones are test pits.
TWO WEEDEN ISLAND MOUNDS
Burial under 15. Extended on back, skull was directly under 15, legs extended
towards red ochre burial.
16-17 Multiple bundle. Skulls and long bones from two or more persons were
under those numbered on the excavation plan (Fig. 3). All of the larger
long bone fragments lay in the same direction. Lack of a sand layer
between the highest and lowest skulls and bones seemed to indicate
that the remains of all four persons were wrapped in the same bundle.
The bundle definitely included the remains of an adult male, an adult,
an adolescent, and a child.
18-19 Multiple bundle burial of an adult female and two adult males.
21-21b Bundle burials of adults.
22 Bundle burial of young adult. Near, but not part of multiple bundle
Burials 23-24.
23-24 Multiple bundle burial. Four persons represented, including a young
person, two adults, and one unclassifiable powdery mass of skull frag-
ments.
25 Adult, probably male. Seems to have been extended on back, less than
one foot under ground surface, feet extended over the edge of the cen-
tral mass interment.
26-38 Numbers assigned to some of the burials in the central mass interment.
and Due to crowding and collapse of skulls into one another, few definitive
45-49 specimens could be collected. The following were definitely included:
26, 38, and one under 37 were male adults; 37, adult; 46, young, prob-
ably male; and 47, young, probably child. All of the burials in this
central mass interment were extended on their backs. Burial was in a
bathtub-shaped pit, lined with a thick layer of organic substance which
left a soft dark mass as much as a foot thick. The first interments,
Burials 26 through 29, were placed across the southeast end with their
heads at the top edge of the pit (Fig. 3). Burials whose skulls run
across the center of the mass, including Burial 37, were then placed.
Their heads rested approximately in the laps of the first row. A third
row, towards the northwest end, including Burials 46-48, were then
placed in another overlapping layer with their heads in the laps of the
second row. The row of skulls and bones, Burials 32-35 and 45, along
the southwest edge of the pit, was so compact that no sorting out was
possible. They may have been extended burials placed in compact
shingle fashion along the top edge of the pit, but in the field, as I
dissected the mass, I felt that this was one elongated multiple bundle
with the skulls and bones originally wrapped in some sort of stiff mat
or hide. The remains of at least six persons were included.
39 Adult male, possibly single skull.
40 Adult, possibly single skull.
41 Adult male, bundle burial.
42 Adult female, bundle burial.
43 Adult, bundle burial.
44 Probably adult, may have been a bundle burial.
BROWNE BURIAL MOUND 7
^IL4
Pir'i
C
/ "
o
^/(S)^^^^^
/9^^%y ^ A?^
I c- M
'4 |
^^^^^i oo ? ^"
TWO WEEDEN ISLAND MOUNDS
POTTERY
Records of the square and level locations of sherds were kept
for most of the work, although any one 10-foot square and 12-inch
level rarely yielded more than two or three specimens. Finally,
sherds were segregated only by sand fill-where they would be in
the same context as the burials-or by the shell ring. On analysis,
no differentiation could be made on any basis, except that the two
St. Johns Check Stamped sherds (listed below) were in the top
foot. A count of all sherds from the mound, including those from
the 1955 trench (Sears, 1957, pp. 17-18), follows.
Sherds from Browne Mound
Number Per cent
Sand-tempered plain 182 66.7
St. Johns Plain 4 1.4
Fine limestone-tempered plain 5 1.8
Colorinda Plain 1 .4
Swift Creek II Complicated Stamped (Pl. I) 53 19.4
Crooked River Complicated Stamped (Pl. I, H) 4 1.4
St. Johns Check Stamped 2 .7
Savannah Check Stamped 1 .4
Weeden Island Red 2 .7
Carrabelle Incised 1 .4
Weeden Island Incised 1 .4
Limestone-tempered complicated-stamped 5 1.8
Simple-stamped (apparently not Deptford) 6 2.2
Unique complicated-stamped, sand-t'd (PI. I, J) 1 .4
Unclassifiable incised 1 .4
Deptford Simple Stamped 1 .4
Deptford Check Stamped 3 1.1
Total 273. 100.0
There is some indication of occupancy in the St. Johns Ia
(late) or Hopewellian time period in the form of two pointed-base
complicated-stamped sherds (P1. I, B-C) included above under
Swift Creek II Complicated Stamped. Notched rims on fine lime-
stone-tempered complicated-stamped sherds in the collections from
the first season's work (Sears, 1957, P1. II, G), included in our
BROWNE BURIAL MOUND
total count above, also indicate the existence of this period. Most
of the specimens, however, fall in the St. Johns Ib or early Weeden
Island periods, including the Swift Creek II Complicated Stamped
and the Weeden Island series sherds as well as sand-tempered
plain rim sherds with folded rims (Pl. II, H). A few later speci-
mens, including the St. Johns Check Stamped, Colorinda Plain,
and Savannah Check Stamped sherds are present. These, I believe,
pinpoint mound construction as occurring in the Colorinda period.
The sherd collection from the mound consists entirely of ac-
cidental inclusions with the fill. The topographic map (Fig. 2)
and the sherd counts for the test pits (explained later) suggest
that the fill came from the hollow adjacent to the mound. That the
earlier sherds probably came from the west end of this hollow and
the later specimens from its east end is suggested by the sherds
from Tests 2, C, and E on the arced ridge from Tests E to I
(Fig. 2).
NON-CERAMIC ARTIFACTS
Only two specimens were found. Locations for both of these
in the central mass burial are marked on the excavation plan (Fig.
3). The first was a cymbal-shaped copper ornament which has a
single central perforation (PI. III, B). A number of smaller paired
perforations near the rim appear to be from aboriginal repairs. The
second was a celt or adz (Pl. III, A). The specimen, made of a
grayish-green fine-grained stone, is double beveled but has a bev-
el twice as deep on one side as on the other. The badly battered
edge indicates considerable usage since the last sharpening.
MOUND CONSTRUCTION AND SEQUENCE
It was not completely clear from the profiles whether the first
interments in this structure were on the original ground surface or
on a small core mound. Plotted datum depths, the shape of the
central pit, and some hints from the profile do indicate that there
was a small, low core mound about 18 inches high and 20 feet in
diameter. Its center would have been somewhere near the center of
TWO WEEDEN ISLAND MOUNDS
the pit for the central mass burial.
The total mound construction sequence appears to have been
as follows:
I. A small core mound was built, into which a rectangular
bathtub-shaped pit, 15 feet long by 10 feet wide, was excavated to
a point about 18 inches below the original ground surface. Pos-
sibly the core mound was simply the result of piling sand from this
pit around its periphery. If not, and the quantity of sand available
from the pit does not seem adequate, sand from the pit must have
added considerably to the height of the core mound.
II. The central pit was lined with a thick layer of some or-
ganic substance on which extended bodies were placed, shingle
fashion.
III. The central pit was then filled with sand just barely high
enough to cover the burials along its upper edges, apparently fin-
ishing the core mound.
IV. Bundle burials and the cremation were placed on the sur-
face of the core mound. Extended Burial 25, the extended burial
under Burial 15, and the burial with the red ochre were deposited
at the same time.
Possibly some fill was added along with these burials.
This would account for the location of the extended burial under
Burial 15. However, there was contact between these burials, so
that the bundles may have simply been placed in a pile directly
over the extended body.
V. A mantle of sand sufficiently thick to cover all bundles,
etc., was then added. Burial 6, the extended burial at the extreme
western edge of the mound, seems to have been placed at the edge
of the mound at this time.
VI. The shell mantle was deposited. Burial 12, extended with
its legs on top of the pile of red ochre, appears to have been the
last burial placed in the mound. It was covered by only a part of
the shell ring.
I see no real alternative to acceptance of this mound as
the product of a single, continuous, short-term ceremony. The only
break in the construction-deposition process appears to have been
at the completion of the core mound with the sand fill over the
BROWNE BURIAL MOUND
central mass burial, but I do not believe that any real period of
time intervened before completion of the total structure.
TEST PITS IN VILLAGE SHELL MIDDENS
Test pits 1 and 2 (Fig. 2) were excavated in 1955 (Sears,
1957, pp. 18-20). Tests A through F, excavated in 1957, were dug
in arbitrary i-foot levels, measured and numbered from the surface
of the ground. The collections from these units, usually 5- by 10-
foot excavations, are individually recorded in tabular form. In the
pottery distribution tables, a number of abbreviations and contrac-
tions for pottery type names and variants have been used. Tables
giving sherd counts and percentages by level have the count as
the first figure and the percentage as the second figure.
Sherds from Test A
Section 1 Section 2 Section 3
Levels 1 1 2 1
St. Johns Check Stamped 3 4.9 1 3.7
St. Johns Plain 11 18.0 3 21.4 3 20.0 4 14.8
Colorinda Plain 23 37.7 1 7.1 3 11.1
Unique comp. st'd., sand-t'd. 1 1.6
Unique incised and punc'd. 1 3.7
Cord-marked, sand-t'd. 1 7.1
St. Andrews Comp. St'd. 1 6.7
Sand-tempered plain 22 36.1 7 50.0 11 73.3 18 66.7
Plain, grit-tempered 1 1.6 2 14.3
Totals 61 14 15 27
Two 5- by 5-foot units formed sections 1 and 3; while section
2, the central one, was a 5- by 10-foot unit. The complete test, a
5-foot-wide by 20-foot-long trench, ran completely through a small
midden ridge (Fig. 2, A). The shell midden was just over 24 inches
thick in the center, the only section where we could take out two
12-inch levels.
fhe unique complicated-stamped sherd has a completely un-
familiar, complex, angular design (P1. I, K). It is, however, identi-
TWO WEEDEN ISLAND MOUNDS
cal to that on a sherd from Level 1 of Test D. The St. Andrews
Complicated Stamped sherd has a zigzag motif. The incised and
punctated sherd is chalky ware with fine punctations between in-
cised lines (Pl. I, M). The cord-marked specimen might be classi-
fied as West Florida Cord Marked, but the cord impressions seem
unduly heavy and widely spaced.
Sherds from Test B
Number Per cent
St. Johns Plain 6 28.6
Sand-tempered plain 5 23.8
Fine limestone-t'd. plain 5 23.8
Early Swift Creek Comp. St'd-like, fine limestone-t'd. 3 14.3
Unique fingernail punc'd. (Pl. I, L) 1 4.8
Deptford Check St'd. 1 4.8
Total 21 100.1
The midden here was less than 18 inches deep and so was
removed in one level. Actually the old top soil was met at 12 in-
ches, and a few shells and sherds had worked down into the top
six inches of sand.
The fine limestone-tempered stamped and plain wares are
hard, well-smoothed, and thin. Temper is a sparse amount of fine
limestone, now represented by small angular holes (P1. III, D). The
stamped ware appear to be quite typical of Early Swift Creek Com-
plicated Stamped, including the notched rims, except for the paste
used.
Sherds from Test C
Number Per cent
St. Johns Check Stamped 1 3.4
St. Johns Plain 2 6.9
Colorinda Plain 11 37.9
Sand-tempered plain 15 51.7
Totals 29 99.9
BROWNE BURIAL MOUND
As with Test B, midden depth was considerably less than it
appeared to be from the surface, so that the midden was almost
entirely removed in one 12-inch level.
Sherds from Test D
Levels 1 2 3
St. Johns Check St'd. 1 1.8
St. Johns Plain 26 45.6 11 23.9 3 10.7
Savannah Check St'd. 1 1.8
Colorinda Plain 12 21.0 8 17.4 5 17.9
Sand-tempered plain 13 22.8 24 52.2 19 67.0
Hole-tempered plain 1 1.8 3 6.5
Weeden Island Incised 1 .1.8
Cord-marked, sand-t'd. 1 1.8
Unique comp. st'd., sand-t'd. 1 1.8
Unique punc'd. (drag and jab), sand-t'd. 1 3.6
Totals 57 46 28
In this test and in Test E below, due to the greater depth of
the midden, we were able to get three levels.
Cord markings on the single sherd here is identical to that on
the one sherd from Test A. As noted under Test A, there is also
identity of the unique-stamped motif. Since the two tests are only
a few feet apart (Fig. 2), the sherds may be from the same pot.
Sherds from Test E
Levels 1 2 3
St. Johns Check St'd. 1 2.3
St. Johns Plain 8 18.6 1 4.8
Savannah Check St'd. 3 14.3
Colorinda Plain 6 14.0 1 4.8
Sand-tempered plain 20 46.5 2 9.5
Hole-tempered plain 2 4.7 1 4.8
Swift Creek II Comp. St'd. 2 4.7 1 4.8
Fine limestone-t'd. plain 4 9.3 12 57.1 3 100
Totals 43 21 3
TWO WEEDEN ISLAND MOUNDS
Savannah Check Stamped sherds in this case might, at least
in one instance, be classified as Wakulla Check Stamped. Two of
the sherds fit the Savannah type well enough; but one specimen
has extremely fine checks, which is not a Savannah characteristic
as I understand the type.
This pit demonstrates rather nicely the precedence in time of
the limestone-tempered plain ware over the sand-tempered plain.
Even with the small samples, and the slight admixture of Colorinda
Plain in two levels, the shift from 46.5 per cent sand-tempered
plain in Level 1 to 57. 1 per cent limestone-tempered plain in Level
2 is as sharp a distinction as could ever be expected.
Sherds from Test F
Levels 1 2
St. Johns Check St'd. 2 3.2
St. Johns Plain 4 6.3
Colorinda Plain 16 25.4
Sand-tempered plain 36 57.1 5 71.4
Hole-tempered plain 1 1.6
Simple-st'd., sand-t'd. (not Deptford) 1 1.6
Unique punc'd. (drag and jab), sand-t'd. 3 4.8
Early Swift Creek Comp. St'd. 1 14.3
Deptford Check St'd. 1 14.3
Totals 63 7
The location of Early Swift Creek and Deptford sherds in the
lower level of this test agrees with the stratigraphic placement of
these ceramics elsewhere (Fig. 4). In the burial mound and in Test
B they were mixed with later pottery. The unique punctated (drag
and jab) sherd is identical to the one found in rest D, Level 3.
VILLAGE SITE CHRONOLOGY
It will be noted that no significant verticalchange in ceramics
is apparent in any test except E. At the same time, there are dif-
ferences from pit to pit and hints of changes in single pits when
the complexes are compared with those from the larger series of
BROWNE BURIAL MOUND
test pits excavated in this area in 1955 (Sears, 1957).
Two known complexes, each a local period marker, show up
in some quantity. These are the complex dominated by the sand-
tempered plain ware and that characterized by the sherd-tempered
Colorinda Plain. Both complexes and periods were defined in the
earlier report (Sears, 1957). Fully in the Colorinda period, or in
mixed Colorinda and sand-tempered plain deposits, are the follow-
ing: Test A, Sec. 1, Level 1 of Sec. 2, and Sec. 3; Tests C, D, F,
and Level 1 of Test E. fest A, Sec. 2, Level 2 represents a pre-
dominantly sand-tempered plain complex.
The pre-Colorinda complex of these units, except Level 2 of
Sec. 2 of Test A, differs somewhat from that defined from such
other units as Test 2, dug in 1955. The characteristic complicated-
stamped and Weeden Island series types are missing. In their
place, especially in the lower levels of Test E and in Test B, we
find a perceptible quantity of thin, hard, well-smoothed sherds
with fine limestone temper. This temper appears otherwise only in
complicated-stamped sherds with notched rims in Test B and in
the burial mound collections.
While our collections are too small to be positive and there is
certainly some mixture, I do think that, at this site at least, the
St. Johns la-late period, the Santa Rosa-Swift Creek or Hopewel-
lian time horizon, is distinguishable from the later St. Johns Ib or
Weeden Island time level by this fine limestone-tempered ware.
It is also clear that there is a slight but consistent occur-
rence of St. Johns Check Stamped sherds in Colorinda period lev-
els. This was also true of Colorinda period levels in Tests l and
2 (Sears, 1957, pp. 18-20). In our preoccupation in 1955 with the
distinction between the Colorinda period complex and clearly de-
marcated St. Johns IIb units with both St. Johns Check Stamped
and Savannah Fine Cord Marked, the significance of these few
check-stamped sherds was overlooked. Generally the St. Johns
Check Stamped sherds which are associated with Colorinda Plain
are made of a soft-paste, often buff-colored, chalky ware, whereas
the later check-stamped sherds are harder and darker. The check-
stamped sherds in the Colorinda levels also seem to average larg-
er checks, as few as four to the inch, which tend to be rectilinear.
TWO WEEDEN ISLAND MOUNDS
It is generally in these deposits that Savannah Check Stamped
sherds are first found, making them another marker for the Colo-
rinda period.
As noted earlier, in the section above on the burial mound,
the Colorinda Plain, St. Johns Check Stamped, and Savannah Check
Stamped sherds in the burial mound fill indicate that it was built
in this period.
I would think, then, that the Colorinda period is the regional
equivalent of St. Johns IIa as well as late St. Johns Ib, as indi-
cated previously (Sears, 1957, Fig. 12). This is based on accept-
ance of St. Johns Check Stamped as the marker for the St. Johns
II period. This adjustment has been made on the culture-chronolog-
ical chart (Fig. 4). Equally necessary adjustments have been made
with the Georgia coast column.
NORTHERN LOWER N.E.GEORGIA CENTRAL N.W. FLORIDA
ST.JOHNS ST.JOHNS COAST FLORIDA COAST
isoo STAUGUSTINE LEONJEFFERSON
100 STJOHN S c _FORT WALTON
4oo STJOHNS rB STJOHNS I11 IRENE
300o C LATE
too1 ST. JOHNS L SAVANNAH MAC KENZIE
to1100 A BROW NE MOUND WEEDEN
1ooo MOUNDI N W ME LTON
9oo D I MOUND
*oo ST.JOHNS A L ISLAND
700 IB SAND I
soo TEMPERED EARLY
400 PLAIN T
400
soo ST. JOHNS LIMESTONE 0 SANTA ROSA-
2oo IA TEMPERED N SWIFT CREEK
100 LATE PLAIN
00
loo ST. JOHNS
zoo
3oo 1A DEPTFORD DEPTFORD DEPTFORD
4oo EARLY
500
600
700 ORANGE ORANGE STALLINGS FIBER-TEMPER
Fig. 4. Cultural sequences in the Lower St. Johns and neighboring areas.
Fig. 4. Cultural sequences in the Lower St. Johns and neighboring areas.
BROWNE BURIAL MOUND
The Wilmington period seems to have been at least partially
coeval with the Colorinda period. This is based on the clay tem-
pering of both complexes and on Wilmington Heavy Cord Marked
sherds found in Colorinda period levels in Test 3 in nearby site
Du-59 (Sears, 1957, p. 11). The Savannah Check Stamped sherds,
grit-tempered in a number of cases, indicate the Colorinda period
to be at least partially coeval with the Savannah period too. Re-
lationship with this latter period is somewhat obscured, since
Caldwell has divided it into two parts: Savannah I and Savannah
II. He states that on the Georgia coast cord marking appears be-
fore check stamping (Caldwell, 1952, pp. 317-19), whereas we
seem to have here some indication of the reverse.
Savannah
Sand Limestone Deptford Compli- St. Johns Fine
St. Johns Tempered Tempered Colorinda Check cated Check Cord
Plain Plain Plain Plain Stamped Stamped Stamped Marked
St. Johns
Colorinda
Sand
Tempreed
Deptford
Fig. 5. Apparent changes through time in major ceramic types at the
the mouth of the St. Johns River.
Figure 5 summarizes in idealized form the changes in the
major ceramic types through time for this area at the mouth of the
St. Johns River. This schematic drawing is based on percentages
derived from all of our test pits excavated in 1955 and 1957 in
sites Du-58 through Du-62 and in site Du-66. The temporal rela-
tionships of the limestone-tempered plain ware and the beginning
and end points of all types are in part inferential.
MACKENZIE BURIAL MOUND
The lake region of central Florida has suffered from a more
than normal scarcity of archaeological excavation and reporting.
Except for C. B. Moore's unusually brief comments on mounds
which he excavated in the 1390's (Moore, 1895), we have only a
report by Sleight (1949)on a mound which had been badly disturbed
by the time the investigators arrived and one by the Cabeens(1955)
on a surface collection from a village site on an island in Lake
Harris.
The map (Fig. 1) and Plate IV, an aerial photograph of Bird
Island, indicate, respectively, the location of Lake Weir and the
position of the MacKenzie mound on Bird Island. It is almost cer-
tain that this little 45-acre patch of land was an island when it
was occupied by the Indians.
Surface material is nowhere abundant, but a few sherds can
be picked up almost any place where the land is not swampy. The
apparent lack of any midden concentrations dissuaded us from at-
tempts at village excavation. A listing of our surface collection
follows:
Safety Harbor Incised 1
St. Johns Check Stamped 6
St. Johns Plain 22
Dunns Creek Red 2
Pasco Plain 2
Unclassifiable complicated-st'd., sand-t'd. 1
Sand-tempered plain 9
Projectile point, Pinellas type, broken 1
Except for the Safety Harbor Incised sherd, this assemblage
is in agreement with that from the burial mound (see below) for a
late Weeden Island (early St. Johns II) dating. The Safety Harbor
Incised sherd was collected an the mound and was probably as-
sociated with a single, badly disturbed, intrusive burial (Burial 23)
MACKENZIE BURIAL MOUND
which had a fragment of blue glass bead in definite association.
The mound stood at the western edge of the village area. It
was about 3 feet high and roughly 75 feet in diameter. There were
perhaps a few more sherds on its surface than was to be expected
for the island generally.
EXCAVATION
The mound was staked out in a grid system, with the east-
west axis bisecting the apparent center of the structure and the
north-south axis of the grid along the west edge of the mound
(Fig. 6). Initially, 5-foot squares were excavated in arbitrary 6-
inch levels. This practice was adhered to as far as possible,
particularly for recording the occasional sherds in mound fill. It
was, of course, necessary to resort to stripping techniques at
many points.
Burials, sherd concentrations, and artifacts other than iso-
lated sherds were precisely located by grid coordinates and depths
below an arbitrary datum plane. Burials were photographed and,
due to their poor condition, reburied or left alone, depending on
specific conditions and locations.
A few sherds were removed from the top of the pottery cache
in the east end of the R-5 trench before its existence was ascer-
tained. Except for these, the pottery deposit was excavated and
recorded by cleaning off as many sherds as possible in a 5-foot
wide strip and then taking a number of elevations at key points,
followed by a photograph. Sherds were then sacked by apparently
related groups, and the process was repeated a second time. Twice
over in this fashion was sufficient for adequate recording and re-
moval in all cases.
The excavation plan and profile (Figs. 6 and 7) present basic
structural and location data. Features of some interest are the
the concentration of burials near the center of the mound and on
the old ground surface, the small pottery vessels generally above
the burials in the fill, the mass pottery cache in a shallow pit at
the east end of the mound, and two small sherd concentrations on
the west side of the structure.
TWO WEEDEN ISLAND MOUNDS
S Single Skull
X Bundle Burial
6 Pottery Sherds
0 Complete Vessel
a Charred Logs
CONTOUR INTERVAL-I"
Fig. 6. Excavation plan, MacKenzie mound.
MACKENZIE BURIAL MOUND 21
inu
0 -
0 0 0
S o a: r
Lm
C)C
P OE
: 3
09 .$
010
xl .:4:
B '- ,r O .[;'. E
TWO WEEDEN ISLAND MOUNDS
It is clear on the excavation plan (Fig. 6) that although we
did not completely excavate the mound, we do have the bulk of
the structural data, burials, and pottery. Further excavation, which
would have involved us with trees, would not have given us any
different structural data and would have added only three or four
bundle burials and a few hundred sherds, presumably of pottery
types already found.
BURIALS
All burials were in extremely bad shape, usually only a little,
in any, harder than the earth surrounding them. None could be re-
moved for further study. Classification of, and occasional com-
ments on, the twenty-three burials encountered follow. All burials
appeared to be those of adults, with the exception of Burial 1.
Locations are shown in Figures 6 and 7.
Burial Comment
1 Infant, shell beads were intermingled with scraps of infant bone, Ves-
sel 1 also in association.
2 Bundle burial, bones in very poor condition.
3 Flexed burial (?), bones in very poor condition, position not known.
4-5 Bundle (?) burials.
6 Bundle burial, possibly flexed as long bones extended rather far out
for a bundle burial.
7 Bundle burial.
8 Bundle burial, skull on center of pile of bones.
9-10 Flexed burials, on left sides, heads towards southeast.
11 Bundle burial.
12 Bundle burial, chert flake between teeth, some charring of long bones,
number of vertebrae still articulated.
13 Bundle burial, skull centered and inverted, bundle oriented northwest-
southeast.
14 Bundle burial, oriented northwest-southeast.
15 Single skull.
16 Bundle burial, oriented north-south.
17 Two skulls, not a bundle burial.
18-19 Bundle burials, two skulls, disordered long bones.
20-22 Bundle burials.
23 Bundle burial, half a blue glass bead in association, most of bundle
removed by plow.
MACKENZIE BURIAL MOUND
were found at various points in the sand fill which formed the bulk
of the mound and covered most burials. Precise locations hori-
zontally and vertically are recorded in Figures 6 and 7, and they
are illustrated in Plates V and VI.
Pottery Cache
SHERDS
Number shi
St. Johns Plain 964
body
rims
direct-rounded
flattened
tapered
broad fold
narrow fold
tall, vertical, direct-rounded
constricted, bottle neck(?)(Pl. VI, C)
bird effigy tail(?) (P1. VI, B)
Dunns Creek Red 177
body
rims
direct-rounded
tapered
flared
killed base, convex (PI. VI, A)
Pasco Plain 8
body (Pl. VII, B)
Sand-tempered plain 37
body
rim, direct-rounded
St. Johns Check Stamped
Weeden Island Incised 11
Carrebelle Incised
Others: 8
Keith Incised (very fine lines)
Miscellaneous incised, chalky
Zoned red, chalky
Modeled, rim point or crude effigy
Alachua Cob Marked (from Level 1)
Safety Harbor Incised (from Level 1)
Totals 1205
From
erd lots Totals Per cent
121 1085 69.3
86 263 16.7
52 60 3.8
37 2.3
365 1579 100.0
TWO WEEDEN ISLAND MOUNDS
The collection from the mass pottery cache is presented in
terms of sherds (above), sherd lots representing significant por-
tions of vessels whose shapes could be restored, and restored or
restorable vessels. Totals from the sherd lot list have been in-
cluded in the sherd tabulation to show that the percentages of
sherds by types are practically the same for the pottery cache as
for sherds from the mound fill and sherd concentrations listed
earlier.
I have no doubt at all that the bulk of the St. Johns Plain
and Dunns Creek Red sherds are from large, simple, open or com-
pressed globular bowls, killed before being thrown into the pottery
SHERD LOTS
St. Johns Plain (usually direct-rounded rims)
Open bowl
Open bowl, apparently very tall
Large pot or bowl, flared rim
Large pot, flared rim
Compressed globular bowl, tapered rim
Compressed globular bowl, convex base
Vertical-walled vessel
Dunns Creek Red
Deep open bowl, direct-rounded rim
Compressed globular bowl, direct-rounded rim
Compressed globular bowl, flattened rim
Very large comp'd. globular bowl, direct-rounded rim
Pasco Plain
Large comp'd. globular bowl, direct-rounded rim
Vertical-walled pot, direct-rounded rim
St. Johns Check Stamped
Vertical-walled pot, direct-rounded rim
1 to 3 large jars, flattened and direct-rounded flared
rims, 5-6 checks per inch
Weeden Island Incised
Vertical zones, fine line fill at 450 to zone edge
Zoned punctation, folded rim imitated by incised line
Zoned punctation, rim flared at sharp angle
Fine zoned punctuation, deep incision, basal part of
vessel wall plain
Carrebelle Incised
Compressed globular bowl, flared rim
Number Illustrations
L21
3
7
5
16
5
62
23
86
37
12
25
12
52
14
37
44
7 P1. VII, C
37 P1. VII, D
53
4 P1. VIII, H
24 PI. VIII, A-B
19 P1. VIII, D-F
6
9
9 P1. VIII, I-J
MACKENZIE BURIAL MOUND
VESSELS-Restored or Restorable
St. Johns Plain, large compressed globular bowl, 10 inches high, 14 inches in
diameter
St. Johns Plain, small compressed globular bowl, about 6-inch diameter
St. Johns Plain, shallow open bowl, 6 1/2 inch diameter
Dunns Creek Red, open bowl, 10-inch diameter
Weeden Island Red, open bowl, thickened wedge-shaped rim, 8-inch diameter
deposit. Probably the mass of sherds in the count of single sherds
represents ten or fiftden such vessels. Presumably they could be
restored by an infinite process of matching the edges of nearly
uniform sherds, but I do not see where such a procedure would be
profitable.
I do not see any need for extended discussion of this ceramic
assemblage. It seems to be a normal St. Johns II burial mound com-
plex for the Florida peninsula and can be matched by many of C.B.
Moore's collections from mounds in the St. Johns River drainage.
Perhaps the only unusual point is the definite concentration of
vessels, most of them complete when deposited, into a cache on
the east side of the mound. As I have pointed out elsewhere (Sears,
1958), such caches are otherwise unknown south of the Aucilla
River.
NON-CERAMIC ARTIFACTS
Our inventory of non-ceramic artifacts is extremely limited,
both in kind and in quantity. The following is a full catalogue.
Projectile Points
Four projectile points, all notched with expanding stems (PI.
IX, B-D), were found in the sand fill, the same context as the
small pottery vessels. The remains of pitch on the stems of two of
the specimens (P1. IX, C), indicating its use as a cement in haft-
ing, is of some interest.
Mica Fragments
Several spots in the sand fill produced minor concentrations of
TWO WEEDEN ISLAND MOUNDS
sheet mica. Sheets were 3 to 4 inches in their maximum dimension.
None of them showed any traces of workmanship.
Conch Shell
A number of fragments of Busycon shell were found concen-
trated around Burial 1, the infant. Although it had been badly bro-
ken and disturbed by the plow, there are some indications in the
parts present that the shell had been fashioned into a dipper.
Conch Shell Beads
Shell beads (PI. IX, A) were found only in association with
Burial 1. Forty specimens were recovered, representing perhaps
one-half of those actually present. Bead varieties were as follows:
5 barrel-shaped, 1/2 by 3/4 inches
7 small disc-shaped, 3/16 by 1/16 inches
28 large disc-shaped, 1/2 by 1/8 inches
Numbers of the disc-shaped beads, both large and small, were
adhering in rows with the perforations lined up, but none of the
rows were adhering to each other. I would guess from this that all
of the beads were from a single string.
Glass Beads
One-half of one bead, the light-blue Ichtucknee type (Goggin,
1952) Spanish-period trade bead, was found with Burial 23. Since
most of this burial had been removed by the plow, there can be
little doubt of its intrustive status. It will not be referred to fur-
ther.
MOUND CONSTRUCTION AND SEQUENCE
The major concern in excavating a burial mound is reconstruc-
tion of the mound-building methods and the sequence of events.
Unless one does this, contexts and associations can be badly
misinterpreted. With this information, both a meaningful descrip-
MACKENZIE BURIAL MOUND
tion of the structure and some informed guesses as to its function
and the ceremonialism involved in its construction are possible.
Apparently, the ground was not cleaned for construction of the
MacKenzie mound. Instead, fourteen bodies (in the area excavated),
thirteen bundles and one flexed (?) burial, were placed in shallow
pits in the humus. Flexed Burials 9 and 10 and bundle Burials 2,
19, and 20, were laid on the surface. This simultaneous mass
placement of bodies, mostly secondary burials, is, I think a defi-
nite indication that ,there was a community charnel house, perhaps
one which also served as the temple. The two flexed bodies may
represent the last two deaths, which, for some ceremonial reason,
caused through the initiation of mound-building ceremonies the con-
struction of this permanent place for the deposition of the accumu-
lated remains.
I would think that the charred logs and charcoal, perhaps from
a sacred temple fire, along with the first-mentioned sherd concen-
tration were placed at the west edge of the burial area at this time.
The third phase was construction of the body of the mound.
Clean sand, probably from a nearby beach area, was brought in and
piled over the burials. The small pots, the sloped west-side sherd
concentration, and the single and double skull burials were placed
at this time as the fill was being piled up. Possible exceptions are
Vessels 5 and 7 on the old humus level. The last items placed in
the body of the mound were Burial 1, the infant, accompanied by a
large number of conch shell beads, Vessel 1, and a conch shell
dipper. I would think that these items represent, together, a single
ceremonial deposit. The infant was probably sacrificed at this
point in the ceremonies, infant sacrifice not being too uncommon
in the Southeast. The Saturiba Timucua were one group who sacri-
ficed children (Swanton, 1946, p. 762).
Quite probably, since they are the only ones of their kind, the
single skull burial (Burial 15) and the double skull burial (Burial
17) had some special ceremonial function. They may have been
treated as ceremonial objects rather than as burials per se.
The profile is confused between stakes 50 and 60 (Fig. 7).
This confusion and the slope of the west end of the pottery cache
certify that the main, or core, mound was complete before place-
TWO WEEDEN ISLAND MOUNDS
ment of the pottery cache.
The last steps in the completion of this mound were the scoop-
ing of a shallow depression in the topsoil adjacent to the east side
of the core mound, the placement of the pottery cache in this de-
pression, and finally the covering of it with the topsoil which had
been removed to open the pit. Since the cache lay on a slight slope,
covering the pots with humus added humus to the slopes of the
of the core mound on this eastern side.
This mound can only be interpreted as being the product of a
single continuous ceremony. This is particularly well documented
by the complete lack of any observable surfaces in the profiles of
the main fill and by the evidence for the existence of such a sur-
face afforded by one of the small pottery concentrations on the
west side. If the surface on which this pottery lay-a surface which
must have been in existence about halfway through the core-mound
construction-had been exposed for even a few seasons, some hu-
mus should have developed and weathering and erosion would have
taken place. This weathering would also have disturbed the align-
ment of the sherd concentrations.
The only possibly break in the continuous construction proc-
ess is after the completion of the core mound and the deposition
and covering of the pottery. It would appear that any reasonable
interpretation of the evidence would eliminate consideration of a
break here too, since a mass deposit must be related, ceremonially,
to the entire mound and its contents. Deposition of such a cache
is not something which would be done years later.
Since the ceramics are predominantly of the St. Johns series,
the mound can be placed in time by the presence of St. Johns Check
Stamped sherds, the marker for the St. Johns II period. The accom-
panying Weeden Island series sherds, as well as the Weeden Island
culture-type pottery deposit, make an early St. Johns II, or late
Weeden Island, dating quite certain. This placement is shown in
Figure 4.
SUMMARY AND CONCLUSIONS
Descriptions of the Browne and MacKenzie burial mounds
have been included in a single report to take advantage of the op-
portunities for comparison. The mounds are roughly contempora-
neous; they share some of the same pottery types; burials in both
appear to come predominantly from storage in charnel houses; and
each is interpreted as being the product of a single ceremony. I
have little doubt that the people buried in them all spoke early dia-
lects of Timucua and may well have been, indirectly at least, in
contact.
Yet there are, for the student of culture-change and history,
differences in the two structures, the implied ceremonies, and the
indicated cultural contacts which are of considerable interest.
Except for possibly the two primary flexed burials, a charnel
house would appear to have been the immediate source of all of the
burials in the MacKenzie mound. This is not so obviously the case
with the Browne mound, where many primary burials were found.
The first interments made in the Browne mound were the twelve to
eighteen extended bodies placed in the central pit.
There are two probably explanations for this mass of bodies:
1. Complete bodies were kept in a charnel house. This is re-
corded graphically by White for the Powhatan, an Indian tribe of
Virginia (Swanton, 1946, P. 86). If this is true, then two quite dis-
tinct social groups are reflected in the many bundle burials scat-
tered around as against the centralized mass of extended bodies.
In such a case, two social classes are the most obvious, and prob-
able, explanation.
2. These bodies are those of persons all of whom died at once.
This implies sacrifice, probably initiated at the death of some one
specific person whose remains are included in the pit with the
others. There are, of course, theories concerned with battles and
epidemics which need not, I think, concern us in efforts to compre-
TWO WEEDEN ISLAND MOUNDS
hend the nature of a ceremonial structure.
Assuming that the Indians had cultural traditions, which would
be at least as strongly represented in normally conservative burial
customs as in ceramic decoration styles, differences in social and
ceremonial status are represented in the Browne mound. That is, if
the resident Indians had social traditions, then the mound was
built and the bodies interred in accordance with ceremonies which
were an extension and representation of these traditions; they were
not simply stuffed into the mound at convenient times and places.
The two other extended burials at the west and north edges of
the mound, represent still a different phase of ceremonialsim and a
different social and ceremonial status. I think that sacrifice is
definitely implied here, since they were placed at these cardinal
points in roughly the same phase of mound construction. Both of
them are female, and it hardly seems likely that both ladies died
natural deaths simultaneously at such a convenient time. If their
bodies had been stored in a charnel house, the positions and late
interment still document special treatment.
Ceremonies in the Browne mound, then, commenced with the
placement of twelve to eighteen bodies in a special central tomb.
These bodies were those of persons grouped together by their soci-
ety in some one specific social status. Either the bodies had been
preserved as complete bodies in a special charnel house, or spe-
cial section of one, or some of the persons had been sacrificed.
The former seems most probable in this case. After filling this
tomb, a number of bundles, each containing the bones of one or
more persons, were placed on the surface around the tomb. Finally,
the two females, presumably sacrificed, were placed at the west
and north edges of the mound, and the entire mass was covered
with a blanket of sand.
The end product here is quite clearly a good example of the
mass burial or charnel house type of mound (Sears, 1958). The
central tomb is a new feature, but in one of the two alternate hy-
potheses offered, it is not necessarily out of line. Referring to the
Choctaw, our major sources for charnel house use, the bodies of
chiefs were placed in a separate charnel house (Swanton, 1946,
p. 726). Nor are the two possible female sacrifices out of place.
SUMMARY AND CONCLUSIONS
Rather, I think that we are here getting further insight into the fu-
neral, as well as social and religious, customs of societies using
charnel houses for the temporary disposition of human remains. The
extra details filling out the picture are only some of the many pos-
sible variations in detail which we may expect with full excava-
tion. Presumably, we may expect informative consistencies in these
additional traits.
In the first report, on excavations in the Browne tract, I went
into some detail to point out that the direction of cultural relation-
ships, contacts, and influences was probably from interior south-
eastern Georgia rather than from peninsular Florida. Certainly ex-
tensive contacts with cultures in the St. Johns drainage upstream
are not indicated for the sand-tempered plain or Colorinda period
cultures. Hence, parallels, as well as sources and origins, for the
set of mortuary phenomena found at the Browne mound are to be
expected on the Atlantic and Gulf coastal plains.
The builders of the MacKenzie mound omitted all preliminary
preparation. Bundles from a charnel house were carried to the se-
lected spot and interred simultaneously on or slightly below the
surface. After a small amount of humus had been placed or re-
placed over these, two complete, apparently primary, flexed bodies
were placed more or less in the center of the area. The presumption
is, again, that a special burial type combined with centralized or
other obviously selected position is indicative of special status in
the funeral ceremonies. In this case, with only two bodies, I in-
cline to the belief that they are important interments. Their deaths
would then have initiated the series of ceremonies which resulted
at their conclusion in the completed mound.
The charred logs and charcoal were placed at the west edge of
the area at about this point in the ceremonies. Since the fire did
not burn there, a sacred temple or charnel house fire would appear
to have been their most likely source. If so, this is of real signifi-
cance, since combined with cleaning out of the charnel house and
other features described below, there was a clearly demarcated
break in community ceremonialism with the building of this mound.
The small vessels-miniatures really-placed in the fill area
over the bodies presumably had some special function. They are
TWO WEEDEN ISLAND MOUNDS
specialized in size and were treated differently in the ceremonies
from the mass of pots in the east-side cache. Possibly they were
temple furniture or, equally possible, they were made especially
for these funeral ceremonies. The assortment of a bottle, a dipper,
a pot, and a number of bowls or cups leads one to think in terms of
a black drink ceremony, but this can only be speculation at this
time.
A final act in this phase of the ceremonies was the apparent
sacrifice of the infant, Burial 1, and the placement of its body, a
conch shell dipper, and a small bowl or cup near the top center of
the nearly completed mound. In this final sacrifice, the MacKenzie
mound ceremonies are again very similar to those which took place
at the Browne mound but differ in the type of personnel involved.
Surely the selection of an infant in one case and of two adult fe-
males in the other with the placement at mound center in the first
instance and at the peripheries at cardinal points in the second
implies social-cultural differences of the first order.
The final ceremony at the MacKenzie mound, placement and
covering of the east-side pottery cache, causes its classification
as a pottery deposit type of mound and points very definitely to
religious influences from the Florida northwest coastal area. The
pottery vessels, mostly large plain bowls, were probably a part of
the ceremonial furniture of a temple. Quite possibly they were ves-
sels in which black drink was brewed for community ceremonies.
The few decorated Weeden Island series sherds placed in the
cache are strong evidence for the special regard held for this ware
and are indicative of its particular ceremonial significance. These
few sherds must have been retained in spite of their fragmentary
nature. Certainly they were not simply everyday containers before
they were broken.
This final feature of a mass east-side pottery cache places
the MacKenzie mound in the "Patterned Mound, pottery deposit
type" class (Sears, 1958). The east-west orientation of various
features fits quite well, although the predominance of secondary
bundle burials, emphasizing the use of a charnel house, and the
lack of a special central burial are divergent.
This is the only mound of the patterned, pottery deposit type
SUMMARY AND CONCLUSIONS
south of the Aucilla River to the best of my knowledge. I would
suggest that it represents a partial shift during the late Weeden Is-
land II period from mortuary practices originating in charnel houses,
with a lack of evidence for a strongly differentiated social struc-
ture, toward the northwest Florida coast patterned mound type,
which indicates very clearly a strongly stratified social structure
(Sears, 1958). That is, while one mound is admittedly inadequate
evidence and only permits the establishment of a working hypothe-
sis, I see here an early stage of a shift in social-religious struc-
ture as expressed in mortuary ceremonialism from a somewhat un-
differentiated to a strongly stratified variety.
Since most of peninsular Florida appears to have been occu-
pied by strongly stratified societies in the sixteenth and seven-
teenth centuries, judging by French and Spanish accounts, it would
seem that the shift to this type of social organization began, in
central Florida at least, early in the St. Johns IIa or Weeden Is-
land II periods. In this connection, it may well be if the evidence
for a break in community ceremonialism-indicated by possible
extinguishment of a temple fire (normally perpetual), the emptying
of the charnel house, and the disposal of temple furniture in the
mound-is acceptable that we have in this mound signs of the end
of one social system and its accompanying religious structure and
the first signs of a shift into another. Proof we do not have, but I
think that the hypothesis merits further consideration and further
work.
REFERENCES CITED
Cabeen, Paul and Grace
1955. "The Horseshoe Island Site, Lake County, Florida." Florida Anthro-
pologist, Vol. VIII, No. 1, pp. 23-26.
Caldwell, Joseph R.
1952. "The Archeology of Eastern Georgia and South Carolina." In Ar-
cheology of Eastern United States (James B. Griffin, ed.). Chicago.
Goggin, John M.
1952. "An Introductory Outline of Timucua Archeology." Mimeographed.
University of Florida. Gainesville.
Moore, Clarence B.
1895. "Certain Sand Mounds of the Oklawaha River, Florida," Journal of
the Academy of Natural Sciences of Philadelphia, Vol. 10. pp. 518-43.
Sears, William H.
1957. "Excavations on the Lower St. Johns River, Florida," Contributions
of the Florida State Museum, Social Sciences, No. 2.
1958. "Burial Mounds on the Gulf Coastal Plain." American Antiquity, Vol.
23, No. 3, pp. 274-84.
Sleight, Frederick W.
1949. "Notes Concerning an Historic Site of Central Florida." Florida
Anthropologist, Vol. II, Nos. 1-2, pp. 26-30.
Swanton, John R.
1946. "The Indians of the Southeastern United States." Bureau of American
Ethnology, Bulletin 137.
C
F G
K
H
--
L N
Plate I. Sherds from Du-62 and the Browne Mound
A, Early Swift Creek Complicated Stamped, notched rim; B, Early Swift
Creek Complicated Stamped, pointed base; C, Early Swift Creek or Swift
Creek II Complicated Stamped, pointed base; D-G, Swift Creek II Compli-
cated Stamped; H, Crooked River Complicated Stamped; I-K, unique com-
plicated stamped, sand-tempered (j has been used as a hone); L, unique
fingernail punctated, sand-tempered; M, unique incised and punctated,
sand-tempered; N, incised, sand-tempered.
Plate II. Sherds from Du-62 and the Browne Mound
A,C-F, St.Johns Check Stamped; B,G, Savannah Check Stamped; H,I, dif-
ferent rim treatments, sand-tempered plain; J, pottery hone.
A
&,,
Plate III. Artifacts, Plain Sherds, and Burials, Du-62 and the Browne Mound
upper, A, celt or adz; I, cymbal-shaped copper ornament; C, hole-
te.npered plain; I), limestone-tempered plain; F, Colorinda Plain.
lower, central mass burial, Browne mound.
s+
~ I~ -
-o
C
U
0
-C
C
-C
C
Cr
-u
0
cl
C
3 'fl
Cr
UN
-,.-)
C
C
u
.,
a U
E
0
E
C
o -
-
C-
o -o
C
rVi)
r C
Cr
r. Cr
Cr
01]
I to
Plate VII. Sherds from Mound Fill and Pottery Deposit, MacKenzie Mound
A, St. Johns Plain; B, Pasco Plain; C-D, St. Johns Check Stamped;
E, incised, chalky ware; F-H, Oklawaha Plain, red painted.
A
(6~'
Nq *IV VFW
D
SJ
K L M
Plate VIII. Weeden Island Series Sherds from Mound Fill and Pottery
Deposit, MacKenzie Mound
A-I,L, \Veeden Island Incised; I-K, Carrabelle Incised; Af, Weeden Island
Red.
0
0
0
B
Plate IX. Shell Beads,
Mound
Projectile Points and Pottery Deposit, MacKenzie
A, Small disc, barrel-shaped, and large disc beads, full size, some ad-
hering in rows as found; B-D, projectile points (note pitch on C), half
size; E, part of pottery cache looking west.
CONTRIBUTIONS OF THE FLORIDA STATE MUSEUM
SOCIAL SCIENCES
number 1. Excavations on Cape Haze Peninsula, Florida, by Ripley P.
and Adelaide K. Bullen. Price $.75.
iber 2. Excavations on Lower St. Johns River, Florida, by William H.
Sears. Price $.75.
nber 3. Eight Tarascan Legends by Maurice Boyd. Price $.75.
iber 4. The Bolen Bluff Site on Paynes Prairie, Florida, by Ripley
P. Bullen. Price $.75.
iber 5. Two Weeden Island Period Burial Mounds, Florida, by William
H. Sears. Price $.75.
(Plus sales tax in Florida)
Order from Florida State Museum,
Seagle Building, Gainesville, Florida
TMLLS MEMORAN FROAR
MILLS MEMORIAL UBRAR.
. i
|
RSS
HIDE
