The Florida anthropologist

Material Information

The Florida anthropologist
Abbreviated Title:
Fla. anthropol.
Florida Anthropological Society
Conference on Historic Site Archaeology
Place of Publication:
Florida Anthropological Society.
Publication Date:
Quarterly[<Mar. 1975- >]
Two no. a year[ FORMER 1948-]
v.28 no. 3, pt. 2, September, 1975
Physical Description:
v. : ill. ; 24 cm.


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


Contains papers of the Annual Conference on Historic Site Archeology.
Dates or Sequential Designation:
v. 1- May 1948-
General Note:
Florida Anthropological Society Publications: Number 7.

Record Information

Source Institution:
University of Florida
Holding Location:
University of Florida
Rights Management:
Copyright Florida Anthropologist Society, Inc. Permission granted to University of Florida to digitize and display this item for non-profit research and educational purposes. Any reuse of this item in excess of fair use or other copyright exemptions requires permission of the copyright holder.
Resource Identifier:
01569447 ( OCLC )
56028409 ( LCCN )
0015-3893 ( ISSN )

Full Text
Carl J. Clausen, H. K. Brooks, and A. B. Wesolowsky
Edited by-
Ripley P. Bullen
The Florida Anthropologist, Volume 28, Number 3, Part 2
19 7 5

THE FLORIDA ANTHROPOLOGIST is published quarterly in March, June,
September, and December by the Florida Anthropological Society, Inc. c/o
Room 102, Florida State Museum, University of Florida, Gainesville, FL 32611.
Subscription is by membership in the Society for individuals or institutions in
terested in the aims of the Society. Annual dues are $6. 00; student members
$4. 00. Requests for memberships and general inquiries should be addressed to
the secretary; subscriptions, dues, changes of address and orders for back issues
to the treasurer; manuscripts for publication to the editor; and newsletter
items to the president. Second class postage paid at Gainesville, Florida.
President Benjamin I. Waller
4911 NE 7th St., Ocala, FL 32670
Directors at Large
1st Vice President Wilma B. Williams
2511 McKinley St., Hollywood, FL 33020
Three years: Ray C. Robinson
1020 4th Street North
St. Petersburg, FL 33701
2nd Vice President Raymond Williams
Dept, of Anthropology, University of
South Florida, Tampa, FL 33620
Two years: Wesley Coleman
10 NW 124th Avenue
Miami, FL 33126
Secretary Wilburn Cockrell, Division
of Archives, History, and Records
Management, The Capitol, Tallahassee,
FL 32304
One year: J Anthony Paredes
Department of Anthropology,
Florida State University
Tallahassee, FL 32300
Treasurer Jerald T. Milanich
111 SW 23rd Terrace, Gainesville,
FL 32604
Editor-Resident Agent Ripley P. Bullen
102 Florida State Museum, University of
Florida, Gainesville, FL 32611

Introduction .......... 1
Geology and Environment 3
The 13 meter Ledge and Its Sediments 7
Archaeological Investigations 11
Description of Human Skeletal Materials 18
Analysis and Conclusions 20
Interpretation of the Sediments .22
Explanation for the Human Remains in the Site 26
Recommendations 35
Acknowledgements 3 5
References Cited 36

The Florida Anthropological Society, Inc. appreciates a contribution
from the General Development Foundation, Inc., which makes possible the
publication of the following important scientific report on Warm Mineral
Springs. The General Development Foundation, Inc., is a non-profit re
search foundation created by the General Development Corporation of
Miami, Florida.

Carl J. Clausen, H. K. Brooks, and A. B. Wesolowsky
The potential archaeological importance of Warm Mineral Springs,
(8Sol9) formerly a cenote-like solution feature in the southern portion of
Sarasota County in southwest Florida (Fig. 1), was first noted in the profes
sional literature 15 years ago in an article in American Antiquity. It described
the discovery of a human skull--reportedly containing portions of preserved
brain--as well as other human skeletal material and artifacts in sediments
located in a shallow "cave" some 35 to 40 feet underwater. A sample of partly
burned log reportedly buried in sediment near specimens of human skeletal
material returned a radiocarbon date of 10,000 200 years B. P. or 8050 B. C,
(Royal and Clark 1960:286). The writers of that article, Eugenie Clark, a
marine biologist and author of popular books, and William Royal, a retired
Air Force colonel and sport diver, asserted that the 10,000-year-old radio
carbon date was the earliest known for man in Florida and suggested that
early man in this area inhabited limestone caves in now-water-filled sinkholes
(Royal and Clark 1960:285-87).
Unfortunately, Royal and Clark were silent both on the excavation tech
niques utilized and on the exact provenience of the specimens. Most important,
no trained archaeologist was involved in the recover or even witnessed the
specimens in situ. The issue was further clouded when the press and national
television sensationalized the finds, printing and broadcasting stories containing
statements and interpretations concerning the nature and significance of the
recoveries which were unsupported by the evidence (Fellows 1962:20-21, 29,
30; Goggin 1962:80,83). As a consequence of this incongruity, the antiquity of
the 1959 Warm Mineral Springs specimens have been viewed with some skepticism.
In January 1971, Clausen, then associated with the Florida Bureau of
Historic Sites and Properties as an underwater archaeologist, dived at Warm
Mineral Springs in response to requests made in late 1970 by Doris Davis,
historian of the Sarasota County Historical Commission, and George Wheeler
Jr. general manager of the spring. Following other work at the site in 1971,
during which unarticulated human bone was observed in the recent sediment
both in the debris cone at the bottom of the spring and on the shallower ledge
area, a small test was excavated in January 1972. The primary purpose was to
obtain additional samples of sediment for floral, faunal and radiometric analyses,
to supplement the geological data previously gathered by Brooks, and to deter
mine feasibility for controlled excavation in these sediments. Clausen was
assisted in these efforts by Gordon P. Watts, now an archaeological assistant
with the North Carolina Department of Cultural Resources, and by other under
water technicians of the Florida Bureau of Historic Sites and Properties.

Fig. 1. Map of the Charlotte Harbor area of southwest Florida
locating Warm Mineral Springs.

Unfortunately, since the announcement of the 1959 finds, the unique and
scientifically valuable stratified deposits located on the ledge in the spring
have been largely destroyed by many amateur collectors who "excavated" in
the spring. Nevertheless, the test, placed in surviving undisturbed deposits
below the surface of the spring in 1972, produced two fragmentary specimens
of human skeletal material which radiocarbon dates of associated organic
material indicate are approximately 10,000 years old.
Prior to this work at the spring, Brooks-mapped, photographed and ob
tained carefully zoned samples of the sediments in the spring for faunal, floral,
petrographic, and radiometric analyses before the deposits were extensively
damaged. This geological work was accomplished between 1958 and 1962 and
is summarized in the following section. Through cooperation, the scientific
value of the independently obtained archaeological and geological information
has been enhanced. The results are complementary and are presented here
with. Because finds of this nature properly receive close scrutiny from interested
scientists, the circumstances of the recovery of these specimens and the general
environment in the area are presented in special detail, much in the manner of
a site report.
Geology and Environment
Warm Mineral Springs originated as a sinkhole during a glacial stage when
sea level was considerably lower than today. Although there are limestone beds
within the stratigraphic sequence at shallower depths (Fig. 2), it appears that
collapse of a cavern in the Tampa Limestone at depths greater than 40 m below
the present ground surface gave rise to the large vertical shaft.
In form, the spring originally consisted (Figs. 3-4) of a round surface
pond (now modified to tear shape) approximately 73 m in diamter, centered
above a deeper shaft. The average water depth of the debris cone in the spring
directly below the surfact pond is approximately 38 m. The basin-shaped bottom
of the pond slopes gently downward from a depth of 0. 15 m to 0. 20 m at the shal
low periphery to depths of 3 to 6 m, where the bottom drops away forming a
curcular opening approximately 46 m in diameter. This constricting orifice or
ledge at the upper end of the shaft of the spring is attributable to the harder,
sandy limestones which underlie the less lithified (primarily sand) surface
sediments (Fig. 2). Below the lip of the orifice the more-or-less cylindircal
walls of the spring gently recede some 2 to 3 m on all sides as depth increases
due to erosion of softer rock materials underlying the upper limestone. It is
from this upper limestone ledge that huge stalactites, sloping outward toward
the center of the spring, have developed. Outward sloping stalactites are pecu
liar to dripstone formations on sinkhole ledges open to air above and not to
underground caverns.


At a mean depth of approximately 13.4 m (Fig. 2), the walls of the spring
suddenly contract to form a sloping ledge wich completely, though not uniformly,
rings the cavity, The width of this 13 m ledge, as we shall refer to this feature,
varies from perhaps 1 to a maximum of 4 to 6 m. In some areas, horizontal
solution fissures lead back from this ledge into the surrounding rock. These
fissures are found primarily on the northeast and southwest sides of the spring.
Through a relatively short period in the recent geological history of the spring,
when the water level in the cavity was lower than at present, the more hori
zontal portions of this ledge nearest the walls, particularly along the northern
side of the spring and to some extent the exposed mouths of the shallow fissures,
caught and held a variety of sediments which settled from above. The human
skeletal remains described in this report were found in this deposit (Fig. 2).
In the pristine state, the walls of the spring above the 13 m ledge to a depth of
5 to 6 m below the present water surface, including the dripstone formations,
were covered with a feathery tufa crust composed of calcite up to 0. 10 m thick.
Where available light is present, a layer of green-to-purplish slime of sulfur
bacteria and blue-green filamentous algae covers the rock surfaces and the sedi
Below the 13 m ledge, the diameter of the spring cavity expands rapidly
through irregularly to form a large chamber. (This flaring is a characteristic
of most underwater collapsed sinkholes in Florida. ) The floor of this chamber
is at minus 38 m. A large, sloping, cavern-like chamber extends downward and
north to northeasterly (Fig. 4). The floor of this sloping chamber is a talus
slope. Depths of up to 80 m have been explored. It is possible that this chamber
extends into the Suwannee and Ocala limestones.
It is into this chamber at depths of 60 to 70 m that cool meteoric water
from the Tampa Limestone and hot connate waters from deeper in the earth
enter the throat of the spring. More than 9 cool and 2 major hot streams, which
enter the chamber along the juncture of the sloping bottom sediments with the
northern wall of the spring, have been identified. These waters ascend through
fissures from below, probably along a fault zone. (It is known that upward mi
gration of hot salty water along fracture zones also occurs in Lee County to the
south.) The maximum temperature of the connate waters is about 37 C. while
the meteoric water, reflecting the mean annual temperature for this area, is
approximately 23 C. Intermixing of the two sources of water occurs to produce
a year-round uniform temperature of approximately 30. 5 C. found at least in
the upper two-thirds of the water column.
Water issuing from the spring is heavily charged with dissolved minerals
and reeks of hydrogen sulfide. Table 1 summarizes the results of a 1972 analysis
of the water by the United States Geological Survey. An unusual photochemical
reaction occurs in the upper meter or two of spring water when exposed to the

i *w
Fig. 3. Aerial view of Warm Mineral Springs and immediate surroundings
looking northwest. Includes some recent extensive changes.

direct rays of the sun. The water becomes turbid as a result of the formation
of either fine particulate sulfur or calcium carbonate. Reportedly 3.4 mega
liters of water issue from the spring each 24 hours. This water discharges at
the southwest side of the spring to form Salt Creek, a tidal estuary for almost
half of its 3. 5 km length. Salt Creek enters the Myakka River 1. 2 km down
stream from the U. S, Highway 41 bridge over the river.
Table 1. Analysis of Warm Mineral Springs water
made 1972 by United States Geological Survey
Total dissolved solids
18,400 ppm
9, 800 mg/1
470 mg/1
600 mg/l
5, 600 mg/l
170 mg/l
10 microgms/l
Total calcium
hardne s s
1,700 mg/l
1.9 mg/l
1 6 mg/l
3,600 mg/l
The heavily mineralized, uniformly warm, waters of Warm Mineral
Springs have been considered curative, particularly for sufferers of rheumatism
and arthritis. A few people swam there when Clausen first visited the site in
the early 1960s. At that time the site was known as Salt Springs and was acces
sible by substandard road and cowpath. Now the area surrounding the spring,
as can be seen in the aerial photograph (Fig. 3), has been extensively developed
as a health spa by the owners, the Warm Mineral Springs Corporation.
The 13 Meter Ledge and Its Sediments
In many places the outer portions of the 13 m ledge, i. e. that part away
from the wall of the spring, which developed on the underlying hard limestone
stratum some 17 to 20 m below the spring surface, are too steep for accumula
tion of sediment (Fig. 2). However, on the northern side of the spring 14 to 15
m below the spring surface, a relatively wide bench developed which accumulated
a variety of debris.
There is a trend in the composition of this material. The lower zone con
tains fall rock weighing up to several metric tons, varying sizes of stalactites
and spalled-off wall encrustation and dripstone (both in place and fallen). Par
ticularly diagnostic of the lower zone is terrestrial vegetation detritus in the form
of leaves, twigs, logs, seeds and charcoal. Calcitic mud layers with snail shells
occur in the lower and middle zones. The middle zone is predominantly of this
material. The upper layer is an algal sludge. Filling the spaces between the


larger masses of fall rock and fallen stalactites and on uncluttered areas of the
ledge is a stratified sequence, the total thickness of which is 0. 9 to 2 m. Three
distinct zones were originally evident, as follows:
Zone 1 An algal slime, the stratigraphically superior layer, ranges from
0. 20 to 0. 50 m in thickness. It is composed of a soft, aqueous, black algal ooze
containing shells of the same small snails still prevalent in the spring, Heleobops
docima and Pyrogophorus platyrhachis. Bones of alligator, tarpon and turtle are
occasionally found in this layer.
Zone 2 Calcitic mud, the middle layer, ranges in thickness from 0. 15 to
0. 50 m. It is predominantly a gray, unconsolidated calcitic silt. There are
some pine bark, oak leaves and other plant debris in this layer. Two distinct
layers of wall tufa represent periods of spalling. Snail shells are common,
especially Physa cbense, Heleobops docima, and Helisoma trivolvis. Verte
brate remains are uncommon and are those of frogs and mice. Radiocarbon
dates on charcoal from the upper and lower portions of this zone are 8, 520 400
years B. P. or 6570 B. C. (W-1241).
Zone 3 A leaf bed, the bottom deposit, varies in thickness from 0. 10 to
0. 80 m. It is the most variable of all the strata, consisting of alternate bands
of terrestrial plant debries (predominantly leaves, twigs, small logs, seeds, and
charcoal) intercalated with calcitic mud layers that contain fresh-water and
terrestrial snail shells as well as fragments of wall tufa. Fallen stalactites
occur as well as fronds of the sinkhole fern Thelypteris normalis encrusted with
a heavy calcitic dripstone layer. Terrestrial snails are more abundant and di
verse in this zone with Helisoma trivolvis and Physa cbense the most common
fresh-water species. The calcitic muds contain an abundance of fresh-water
ostracod shells. Identified plant remains represent the following species:
Pinus elliotii, Sabal palmetto, Quercus virginiana, Q. laurifolia, Ampelopsis
arbrea, Carya sp. Phylocacca rgida, and Thelypteris normalis. The most
common polynomorph is a pyospore of a species of Chlamydomonas or a related
fresh-water algae. Vertebrate remains so far identified consist of man, deer,
opossum, raccoon, rabbit, squirrel, mouse and frog. Radiocarbon dates on
charcoal collected by Brooks from this zone are 9,370 400 years B. P. or
7420 B, C. (W-1245), 9,500 400 years B. P. or 7550 B. C. (W-1212), and
9,870 370 years B. P. or 7920 B, C, (W-1153), from top to bottom respectively.
These dates are approximately 1000 years older than those given above for
Zone 2.
The upper two zones described above are now largely destroyed. When
Clausen worked at the site in January 1972, remnants of the complete sequence
were only observed at one point along the north wall of the springs. Moreover,
the Zone 3 sediments revealed extensive damage. In the intervening years since

Fig. 5. Site plan locating test excavation, coordinates of vertical survey line,
and location of Line a'-b' (Fig. 2). Maximun extent of overhang is
indicated by Arc a. Arcs b and c give locations of spring wall at
depth of 12 m and outer portion of 13 m ledge at 17 to 20 m depth.

I960, Royal has found and removed portions of 7 human skulls and other skeletal
material representing 30 individuals from the Zone 3 sediments (William Royal,
personal communication)
The astonishing state of preservation of the organic materials, particularly
the plants, in these 9-10-thousand-year-old sediments is probably attributable
to their deposition in the hard waters of the cenote, which may have even become
supersaturated on occasion; and, secondly, through later infusion of the deposit
by the heavily mineral-charged waters, with very little dissolved oxygen, which
now issue from the spring. Contributing factors in the preservation of these
materials under the first, and possibly the second, circumstance above might
be a little-understood antibiotic capability of concentrations of certain bacteria
or algal organisms in the spring waters (Breder 1957:132-135).
Except for one species of poorly known snail, there is nothing in the biota
that does not now occur in the area. Several of the snail species do not now
range north of Tampa Bay. Climatic implications are that the temperature ranges
were similar to that now existing in the area. However, considerably more aridity
than now must have occasionally prevailed during deposition of the two lower zones.
This is indicated by the carbonate-encrusted fronds of the sinkhole fern and the
geochemical conditions necessary for deposition of calcitic mud and tufa as both
wall encrustation subaqueously and occasionally as dripstone growths.
Archaeological Investigations
Disturbance of the sedimentary sequence by sport divers made it necessary
to carefully select an area for testing. A survey, completed during the afternoon
of January 18, 1972, indicated that a several-meter-long undisturbed section of
Zone 3 remained on the ledge along the northwest wall of the spring. The site
plan (Fig. 5) indicates the general location of this deposit and the test excavation
described below. There was no evidence of Zones 1 and 2 (The aqueous, algal
ooze, and the gray calcitic silt) at this point, these upper deposits evidently
having been stripped away. The 13 m ledge in this area was overhung some 3 m
by the limestone and mudstone walls which slope inward overhead to meet the
circular orifice above. This rock overhang effectively shielded that portion of
the ledge nearest the wall from debris settling through the water from above,
preventing deposition of recent material -- a significant point we shall discuss
in greater detail further along in this section and later in the conclusion.
The widest portion of the leaf zone deposit on the ledge at this point was
overlaid by a massive growth of tufa weighing an estimated 150 to 200 kilos.
Close examination indicated that this tufa formation was undisturbed and may
have formed directly on top of one of the uppermost leafy strata of Zone 3, ef
fectively covering the underlying sediments and protecting them from any dis
turbance (Fig. 6).

Fig. 6. Thick tufa formation
protecting a portion of the
Zone 3 sediments on the 13
m ledge in test area. White
aluminum rod is in initial
position described in text.
Scale is metric, delineated
in decimeters and centi
A point in front of the tufa formation and a few centimeters back from
the edge of the deposit was selected, through which an approximately 10 mm
in diameter, white-painted aluminum rod 1.5 m long was inserted vertically
through the sediment. In this position the rod was intended to serve, after
removal of the tufa cap, as the near left (most southern) of the four corners
of the proposed excavation. A small area of the ragged exterior of the leaf
zone to the right of the aluminum rod was then faced down vertically to permit
close examination of the sedimentary profile. At this point the leaf bed (Fig.
7a) was just over 0. 70 m thick and consisted primarily of thin alternating
bands of terrestrial plant debris and calcareous sediments (tufa and marl).
The Zone 3 sediments rested in turn upon an irregular zone of fragmented
gray-green clay from the country rock.

Three substantial specimens of wood imbedded in the exposed face of the
leaf zone were selected, photographed in place and removed for radiocarbon
dating (Fig. 7a, circles). It was intended that these samples serve as a control
for the materials that would be secured for dating purposes from the individual
levels of the test excavation to follow. Three samples (WMS-14501 from 0. 10 m,
WMS-14502 from 0.37 m, and WMS-14500 from approximately 0.70 m below the
top of the deposit) were submitted to the radiocarbon dating facility at Gakushuin
University in Tokyo, Japan, and returned dates of 8,920 190 years B. P or
6970 B. Co (Gak-3992) for the top sample, 9,350 190 years B. P. or 7400 B. C.
(Gak-3993) for the middle sample, and 9,220 180 years B. P. or 7270 B. C.
(Gak-3991) for the stratigraphically lowest sample.
Colonel William Royal, who was still diving at the spring in 1972, and
another diver assisted in the removal of the tufa formation. Under Clausen's
supervision, this formation was floated free and disposed of to permit direct
examination of the surface of the leaf zone from above. It was determined that
a slight adjustment in the position of the aluminum rod would permit a test exca
vation a meter long by one-half meter wide. Consequently, the rod was with
drawn from its original position and re-established approximately 0. 14 m to the
west (Fig. 7b). The approximate Universal Transverse Mercator grid coordinates
for this rod, the south corner of the test, calculated from the coordinates of a
vertical survey line projected from the U. S Department of the Army Corps of
Engineers map, Myakka River Quadrangle, 7.5 minute series, are 374,982.7rn
East and 2,993,523.1 North, Zone 17. The upper surface of the leaf zone de
fined by the intersection of the top of the zone with the rod marking the south
corner was 12.70 m below the air/water interface of the spring on January 20,
1972, 10.70 m below MSL. The remaining three corners of the test were then
established in standard fashion using rods set vertically through the deposit and
into the gray-green clay below.
The proximity of the test to the sold claystone wall of the spring at this
depth, and the relatively delicate nature of the sediments, dictated excavation of
the test from the side nearest the center of the spring rather than from above.
By increasing negative buoyancy with lead weights we could balance on our knees
on the exposed sloping outer portion of the ledge with the dark void of the deep
central portion of the spring to our backs and the area to be excavated to our
immediate front (Fig. 8). To facilitate excavation from this position, it was
first necessary to face down the ragged exterior of the deposit to conform to
the vertical meter-long southeast side of the test. In the event scientifically
valuable material outside of the test might be encountered in the facing-down
operation, the sediment was carefully removed and examined by hand. In the
course of this work, the fragmentary left ilium (Fig. lib) of a human juvenile was
encountered outside the test at a point 0.254 m below the upper surface of the leaf
zone, 12.95 m below the air/water interface (10.95 m below MSL), 0.40 m north
east (measured along the southeast face of the test), and 0.15 m southeast

1 9,4 20i 1 50
7 4 70 B.C.
B -
Fig. 7. Views of Zone 3 sediments,
a, photographic mosiac of deposit after dressing face with
rod in initial position. Circles locate proveniences of wood
samples for radiocarbon dating. Note capping tufa formation
and zone of irregular clay at bottom, b, left side of south
east face. Rod is in second position. The 0.10 m excavation
levels shown at right with radiocarbon dates obtained from
materials in levels as indicated.

(measured at right angles to the side of the test) This speciment will be
discussed later in this report.
Since the principal purpose of this test excavation was to secure samples
of faunal and botanical specimens for laboratory analyses, and the deposition
of the strata in the deposit appeared essentially level (Fig. 7b), we elected to
excavate this first test by arbitrarily imposed 0.10 m levels. This approach
was adhered to even after it was determined in the course of excavation that
the strata within the deposit sloped downward slightly from north to south, or,
when viewed from the southeast, from right to left. Our efforts represented the
first time that controlled archaeological excavation in this type of finely stratified
deposit had been undertaken underwater; thus we were unsure of our ability to
excavate by natural stratigraphy. In retrospect, it appears our concern was
unwarranted and that this shift might have been accomplished successfully,
Levels of the excavation were assigned identifying numbers in serial order
from the top down, starting with 1. A photograph (Ektachrome positive) was
made of the horizontal upper surface of each succeeding level. The material
in each level was loosened in small sections with short stiletto-type divers'
knives (much in the manner a trowel is employed in land sites) lifted out, and
carefully separated in the hand using the fingers. All readily identifiable verte
brate faunal material and major botanical specimens such as acorns, hickory
nuts, and budding twigs, as well as portions of charred wood or other organic
matter suitable for later radiocarbon dating, were bagged in clear polyethylene
sacks and identified by level.
In addition, large samples of sediment were taken to insure preservation
of a representative sample of the minute skeletal remains of smaller vertebrates
which we knew from examination were included in the deposit, but which were
difficult to see in the dim light and reduced visibility at this depth, and to secure
a full range of botanical specimens including pollen. This was accomplished by
carefully cutting free and removing intact from the south corner of each level a
block sample of the sediment measuring 0.30 m long, 0.20 m wide, and to cor
respond with the excavational units, 0.10 m thick. Maintaining the stratigraphic
orientation of each of these samples, the individual blocks of sediment were
placed in plastic sacks, sealed, identified, and removed to the surface where
they were quick-frozen. The five of these 6,000 cc samples, WMS-14507, -14510,
-14512, -14514 and -14521, taken from Levels 1 5 of the test, can be stacked
in order, reintegrating a virtually continuous column of original deposit approx
imately 0.50 m in depth, A sixth strata sample, WMS-14517, taken from the very
incomplete Level 6, was secured in the usual manner samples are taken, bagged,
but not frozen. No sample was taken for Level 7 of the test. Macroscopic
botanical or faunal remains were not encountered in excavating Levels 6 and 7.
The sediment samples, level bags and other specimens were turned over to the

Fig. 8. General view of 13 m ledge area during facing down of
Zone 3 sediments. Four aluminum rods delineate the
four corners of the test. Distortion is attributable to
extreme wide-angle lens required to obtain clear
photographs in the murky eaters of the spring. Dark
area to lower left is the drop-off into greater depths.

preservation laboratory operated by the Bureau of Historic Sites and Properties,
Florida Department of State, for storage and analysis.
Excavation of the test progressed uneventfully until Level 4. In that level,
at a point 0.35 m below the upper surface of the leaf zone, 13.05 m below the air/
water interface (11.05 m below MSL), 0.35 m northeast (measured along the
southeast face of the test) and 0.06 m northwest (measured at right angles to
the side of the test) an almost intact first sacral vertebra (Fig. 11) of a human
juvenile was discovered. No artifacts were recovered in the test. No breaks
or disturbances were noted in any of the discrete sediment layers in the leaf
zone during excavation.
Samples of wood, in one instance slightly charred, were isolated from
each of the upper five levels of the excavation and submitted to the Gakushuin
University radiocarbon laboratory with the control samples previously mentioned.
The laboratory was informed of the circumstances under which the samples had
been collected and was specifically instructed to take all necessary steps to in
sure that any antique carbonate compounds they might contain would be purged.
Results (GAK-3995/6/7/8/9) for the five levels have been plotted in Figure 7b.
They indicate that the sediments in Level 4, the level containing the human
vertebra, and the next two superior levels ( 3 and 2 ) of the test were more than
10,000 years old (Fib. 7b).
Two anomalous features which came to light during the course of the exca
vation warrant comment. In order of appearance, these were a second tufa
formation which intruded into the excavation from the northeast; and a fallen
stalactite, lying horizontally with the broken, basal portion extending into the
excavation from the northwest. Both are visible in Figures 9 and 10. The tufa
formation is of particular interest, for the delicate crystalline structure, rounded
sides, undulating upper surface and flat underside suggest that it formed in place
on top of a lower former surface at the Zone 3 sediment.
The potential antiquity and importance of the human skeletal material
encountered were recognized during excavation. As a consequence, the point
marking the intersection of the vertical survey (buoy) line from the surface,
down which the elevations were measured, with the surface of the ledge, was
permanently identified with a large nail, 15 cm long, driven into the limestone.
The approximate UTM grid coordinates for the survey line and the point on the
ledge identified by the nail were 374,985rn East and 2,993,523.10 North, Zone
17. The south corner of the test lay 2.30 m away from this point, toward the
wall of the spring at 270 magnetic. The aluminum rods marking the corners
of the test were also left in place.

Fig. 9. Completed excavation. The basal portion of a fallen stalactite and another
tufa formation can be seen extending into the excavation immediately in
front of Clausen's lowered head.
Description of Human Skeletal Material
The human skeletal materials from the test excavation in Warm Mineral
Springs, Florida, consist of an almost entire first sacral vertebra and frag
ments of left ilium. The first sacral vertebra (Fig. 11a, 1-4), found in Level 4
of the test, is missing a portion of the cortical bone of the superior anterior
area of the body of the vertebra, and most of the lateral edges of both costal
elements. Likewise, the dorsal portions of the laminae are broken off. All
of these breakages appear to have occurred in antiquity. Bilaterally, the
synchondroses along the body of the vertebra, the roots and transverse processes,
and the costal elements are incompletely fused. Synotosis is more advanced in
the lateral and dorsal regions, leaving a visible gap at the juncture of the edge
of the body and the costal elements. Some very small areas of the epiphyseal
synchondroses of the lateral and inferior margins of the costal elements are
preserved. Unfortunately, there is not quite enough left of the left lateral ari-
cular surface of the costal body preserved to demonstrate articulation with auri
cular surface of the iliac fragment described below.

Fig. 10. Photomosaic profile of 9-to-10-thousand-year-old Zone 3 sediments
exposed in the meter-long northwest side of the test excavation. Note
fragmented gray-green clay in bottom of excavation, tufa above the
sediments beyond the data board, and basal portion of fallen stalac
tite just to the right of the upper part of the scale.

The left ilium (Fig* lib), found 15 cm outside the test to the southeast,
is fragmentary; but three portions are preserved. One of the fragments is
from the lower medial portion of the body of the ilium, comprising the aricular
surface, the posterior portion of the greater sciatic notch, and the area posterior
superior of the aricular surface almost to the iliac tuberosity. The second frag
ment preserves a portion of the iliac and the wing of the ilium from the anterior
superior spine (which is not at all pronounced) superiorally to the approximate
center of the insertion of the obliquus abdominis externus. The third fragment
is a portion of the iliac wing and the iliac crest. It does not join with the other
fragments. All preserved surfaces of the iliac crest show the characteristic
undulating appearance of a synchondrosis, with no evidence of synotosis.
The size of the iliac fragments and the degree of synotosis among the
elements at the sacral vertebra suggest that this individual was approximately
6 years of age at the time of death. Although the two bones were found a short
distance away from one another, approximately 22 cm apart, it seems probable
that both the vertebra and the innominate fragment are from a single individual.
At least there is no osteological evidence to suggest otherwise. There were no
apparent differences between these bones and those of modern children.
Analysis and Conclusions
The two human bones recovered from submerged deposits in Warm Mineral
Springs, Florida, in January 1972 and dated by radiocarbon method as 10,000
years old, would seem to represent the earliest scientifically verifiable evidence
of the presence of man in Florida and the Southeast. The find confirms the I960
claim made by Royal and Clark that the spring contains 10,000-year-old remains.
It is beyond the scope of this report to reveiw previous finds and claims
of early human skeletal material in North America. The majority of these dis
coveries have been reported in detail in the literature, have been questioned
and defended, and have, among interested scholars, their supporters and detrac
tors. An excellent synopsis of the circumstances of these finds, up to 1957, and
an intelligent discussion of the evidence for their antiquity, can be found in a
chapter on early human skeletal remains in H M. Wormington' s Ancient Man
in North America (1957:225-248). More recent finds are reported in the general
literature, and an excellent article on what is generally known of the early in
habitants of North America with comments on recently discovered sites is pro
vided in "The Earliest Americans" by C. Vance Haynes Jr. (1969), in Science.
In view of the problems involved with establishing adequate dating for
other early remains, Warm Mineral Springs man (or child) may prove to be
the oldest, closely dated, adequately documented find of human remains east
of the Mississippi River, or even perhaps so far discovered in North America.

Fig. 11.'Human skeletal material.
a, four views of first sacral vertebra from Level 4 of the test; b, median view
of left ilium from Zone 3 sediments found during preparation of area for exca
vation; c, left ilium of juvenile American Indian from physical anthropogical
range at the University of Texas in Austin included for comparative purposes.

However intriguing the occurrence of human remains in the spring may be,
correct interpretation of the origin of the sediment in which the remains were
encountered is likely to have a more far-reaching impact on archaeology and
geology. Consequently, we shall first advance an explanation of the origin of
the 10,000-to-perhaps-8,500-year-old sediments on the 13 m ledge in the spring
and then follow with possible explanations for the human bones in the deposit.
Interpretation of the Sediments
The preponderance of evidence so far assembled both from the previous
geological and the more recent archaeological investigations indicates that
Zone 3, the sediment in which the human remains were encountered is sub
aqueous in origin, as are Zones 1 and 2, and not a dry or even a moist cave
deposit. Evidence for this assertion follows:
1. The observed interbedding of the leafy material in the sediments on
the ledge was indistinguishable from obviously water-laid similar material in
the central debris cone some 25 to 32 m deeper in the spring.
2. The degree of preservation of the 10,000-year-old organic remains
in this sediment, which in many instances are simply slightly discolored, sug
gests that potent preservational forces were at work. Deposition in extremely
hard water followed by permeation of the sediment by the heavily mineralized
connate waters with low dissolved oxygen levels, which probably began issuing
from the spring within 1,000 to 1,500 years after this sediment formed, could
explain the superb preservation.
3. No disturbance attributable to trampling or disruption of any nature
by man or other large animals, or evidence of animal burrowing, was observed
in the fragile strata penetrated by the test or in any portion of the Zone 3 sedi
ments. Evidence of such disruption would be expected in sediments that were
dry or intermittently emerged. Lack of disruption suggests that the minimum
depth of water over these sediments was probably at least 1 to 1.5 m.
4. Evidence of "drip holes" which form in soft sediments exposed to con
tinuously dripping water in caves and rock shelter situations (Griffin 1974:9)
was nowhere discernible in the Zone 3 sediments.
5. Although there was considerable particulate charcoal, (including some
large fragments several centimeters across) and partially burned branches and
twigs were not uncommon in the deposit, these materials were randomly dis
tributed horizontally through the sediments. No fire-damaged or blackened
rocks or arrangements of rocks or concentrations or arrangements of partly
burned sticks or lenses of charcoal were discernible in the deposit anywhere

on the ledge. Moreover, there was no evidence of in-place destruction of the
organic fraction of the sediment attributable to fire. In fact, the distribution
of charred and partially burned material in the deposit is entirely consonant
with what would be expected from wildfires. Certainly, there was no evidence
of hearths.
6. There was no discernible evidence or rodent gnawing on any of the
bones of the small sample of the remains of larger vertebrates recovered in
the geological and archaeological test; nor was there any evidence of purpose
ful fractures attributable to butchering or consumption, or discoloration attri
butable to heat.
7. The crystalline tufa formation extending into the test described earlier,
if formed in situ, would probably have to have been immersed in water saturated
or even supersaturated with carbonates. This situation existed periodically and
resulted in formation of the strata of lighter color consisting of particulate car
bonates and calcitic mud.
At the present time, organic materials similar to those making up the
organic fraction of Zone 3 are being introduced to the spring, dropping directly
on the water from overhanging or nearby trees and shrubs or conveyed by winds
or surface run-off to the spring where they become water-logged and sink dir
ectly to the bottom debris cone. Deposition of this debris has been prevented over
that portion of the 13 m ledge nearest the wall of the spring shielded by the over
hanging walls and upper ledge. The slope of the exposed outer portion of the ledge,
i. e. that portion lying from 14.3 to 17 m below the present surface, is so steep
that materials settling on it slide off, falling into the central depths of the spring.
No evidence of recent terrestrial plant debris was found in 1972 or in the pre
vious geological study,on that portion of the 13 to 14 m ledge in the northwest
quadrant of the spring.
Sediments intercepted by the test appeared relatively uniform in thickness
from a point near the rear or northwest face of the excavations, nearest the
wall of the spring, to the front toward the center of the spring. To achieve this
uniformity, more-or-less even deposition would be required over at least that
0.5-1.5 m of the width of the ledge lying nearest the mudstone and limestone wall
of the spring. Only when the water level in the spring intersected the walls of
the central cavity at its greatest diameter at this depth, just above the inner
portion of the ledge, can even deposition of the type of sediment encountered in
the test excavation occur. Taking the slope of the wall into consideration (Fig. 2),
the water level in the spring could range from approximately 13 m up to 9.5 m
below its present surface and still permit formation of this sediment. This is an
extremely important point, for if the Zone 3 sediments are subaqueous as we
believe and deposition of the terrestrial plant debris in them could only occur

when the water level in the spring was within certain limits, then the radiocarbon
dates for the onset and ending of the deposition of Zone 3 clearly establish a gen
eral range for the water level in the central cavity some 9 to 10 thousand years
Using the dates for the Zone 3 sediments in the preceding geological and
archaeological sections, we believe that deposition of the Zone 3 sediment on
the 13-to-14-m ledge was initiated a little over 10,000 years ago, perhaps as
early as 10,500 years ago, when the water level in the spring cavity, in general
response to the eustatic rise in sea level attributable to the wasting of Wisconsin
glaciers, first ranged above an elevation some 14 m below present spring level
(12 m below MSL). This phase of deposition continued for 1,000-1,500 years
during which the character of the sediments changed from more to less plant
debris content as the mean water level in the cavity continued to rise, probably
in response to a generally rising sea level. The required lower range for the
water level in the spring during this period (from approximately 10,000-plus to
9,000 years B. P, ) was from approximately 9.5 to 13 m below present spring
surface (7. 5 to 11 m below MSL).
The marked stratification discernible in this zone ( Fig. 7b) is unquestionably
attributable to variation in spring water level and perhaps also to concentrations
of dissolved minerals, particularly carbonates. We assume that the lighter
calcitic strata represent lower than average water levels during this period,
perhaps reflective of prolonged droughts, when cave wall spalling and carbon
ate precipitation occur, and that the dark leafy strata reflect average, or "normal"
water levels. However, under certain circumstances this order might be rever
sed; i. e. if the range of fluctuation were wider, perhaps brought about by varia
tions in the rate of rising sea level, or if for some reason, such as a period of
increased precipitation in north and central Florida (the area of aquifer recharge)
the spring flowed briefly, filling the cavity with connate waters.
Extending this interpretation, we postulate that the sediment in Zone 2 was
formed when the water level in the spring had risen higher. That is to a point
where the carbonate fraction (primarily precipitates) evident in Zone 3 was
still reaching the inner portion of the ledge; while, because the average water
level was somewhat higher than the overhand, the vast majority of the settling
water-logged plant remains, and the larger drowned vertebrates, sank into the
central cavity. The continued, although greatly reduced, presence of the smaller
vertebrates (frogs and mice) in the Zone 2 sediments supports this interpretation.
Frogs of course swim underwater and the small, rapidly swimming rodents
would be expected to penetrate further under the ledge along air-filled irregu
larities and in minimal air space during periods when the water surface was
low e r.

Brooks' radiocarbon dates for the onset and conclusion of the Zone 2
sediment suggest that this deposit was laid down over a relatively brief
interval from as little as 200 years to possibly 900 years. The maximum
age for the onset of the Zone 2 sedimentation, about 9,000 radiocarbon years
ago, and the most recent date for the termination of Zone 3 sedimentation,
about 8,730 radiocarbon years ago, suggest that the water level in the spring
had risen into the range required for deposition of this sediment, some 7-10
m below present spring surface (5-8 m below MSL) by that period.
The continuing cycles of carbonate precipitation visible as discrete
strata in Zone 2 would seem to indicate that the water level in the cenote
continued to fluctuate. Two conspicuous layers composed of a high concen
tration of tufa, one near the bottom and the other approximately a third of
the thickness of the zone from the top, suggest that twice during the deposi
tion of Zone 2 the water level dropped some distance below the constriction
formed by the limestone ledge, 5-8 m below present spring level, and that
sub-aerial spalling from the exposed spring walls under the ledge occurred.
Finally, Warm Mineral Springs ceased to be a passive cenote with
fluctuating water level and started to flow, initiating deposition of the Zone 1
algal sludge. This occurred subsequent to deposition of the upper portion
of Zone 2 dated at 8,500 years B. P.
Supporting evidence for this transformation to a flowing, connate spring
around this date has emerged from yet-unpublished research at Little Salt
Spring, a geologically similar spring with heavily mineralized warm waters,
located just over 3.12 km to the northeast of Warm Mineral Springs. At
that site a pair of pointed wooden pins were discovered which had been driven
into a clay-filled crevice in the limestone at the edge of the drop-off above
the central cavity of the spring at a point approximately 10.66 m below the
present spring surface ( 5.53 m below present MSL) One of the pins was
sacrificed for radiocarbon assay and returned a date of 9,645 160 years B. P.
or 7695 Bo C. (1-6460) Hence Little Salt Spring was also a cenote and the water
level there was at least 10-11 m, probably slightly more, below present sea
level at that time.
Data in an unpublished report detailing the results of examination of
samples of sediment from Little Salt Spring for plant and animal microfossils
prepared by Habib Yezdani and Edward S. Deevey Jr. (1973) suggest that the
lowest of the 7 major strata discernible in the recent sediments in the basin
of that spring, an algal gyttja, was deposited under conditions similar to those
presently in existence there, i. e. a flowing, brackish spring. A radiocarbon
date of 8,455 145 years B. P. or 6505 B. C. (1-6458) was obtained on a sample
of this gyttja.

Barring fundamental changes in the elevation of the outflow at Little
Salt Spring -- we see no evidence of this -- the net rise in water level in
that spring required to satisfy these conditions would be almost 11 m. This
record so closely parallels what appears to have occurred at Warm Mineral
Springs that it constitutes persuasive evidence for accepting the indicated
transformation of that site from a cenote to a spring at the time indicated
or 8500 Bo P. The simplest explanation for this transformation at both springs
is that post-Wisconsin sea level, around 8500 Bc P. had reached a point
closely approximating present MSL; and that the springs, responding to a
potentiometric water level similar to that existing at present, began to flow.
However, other factors, such as increased amounts of precipitation in the
recharge area in central and north Florida, may have contributed to the
higher potentiometric level in the Floridan artesian system.
Explanation for the Human Remains in the Site
Two theories--the earliest advanced by Royal and Clark in their I960
report that humans inhabited the spring; and the most recent that the human
remains represent burials, directly interred on the ledge (Cockrell 1974)--
are both dependent upon a lower water level in the spring cavity than our re
search indicates existed at the time the remains were introduced. Consequently,
we view these theories, particularly the former, as improbable explanations.
Royal and Clark's (1960:285) theory that Paleo-Indians "lived in limestone
caves" in the spring "when the sea level was considerably lower than at present"
is the least creditable interpretation. Their application of the term "cave" to
any portion of the upper spring is seriously misrepresentative of the morphology
of that feature. The ledge is more like a rock shelter than a real cave.
Moreover, even if the water level in the spring were several meters below
the 13 m ledge, the drawbacks of this small shelf area as a potential habitation
area are overwhelming. At some points, notably along areas of the eastern
wall of the spring, the ledge is little more than a notch in the limestone. Where
the ledge is wider, along the northwestern and northern sides of the spring, the
level usable portion nearest the wall is quite meager, typically only a meter or
two in width. To make matters worse, the balance of the ledge in this area
slopes steeply downward toward the central cavity, and is composed of a soft
mudstone and soft limestone which in the presence of even minor amounts of
moisture becomes extremely slippery. Also, the minimal overhang above the
ledge would provide scant shelter. There would also be the very real problem
of safely reaching the ledge from the overhanging lip of the then-dry basin
above, as well as of having a reliable means of exiting from the cavity. Goggin
(1962:81) pointed out, after one of several reconnaissance dives at Warm
Mineral Springs, that in order for humans to live on this ledge (and to bring in

food) they would have to be "great alpinists. "
Also insupportable is the contention that a human skull and other materials--
"discovered" (or at least recovered) on February 5, 1 973--regrettably during
a staged press conference at the spring--represent a burial or at least a primary
purposeful interment. Our interpretation of the sediments in which the bones are
found as a subaqueous deposit, formed under a water depth from a few centimeters
to severl meters, would make actual burial of an individual in the sediemtns
on the ledge unlikely if not impossible. While artifacts have been found in the
same sediment with the human skeletal material both by Royal and Clark and
reportedly during the most recent investigations (a shall artifact has been sug
gested by Cockrell in a personal communication as possibly being an atlatl hook) ,
the presence of random artifacts near or even with human material in the sedi
ment does not constitute proof of primary interment. Any individual falling into
the spring, then as now, might be expected to have on his person a sample of the
material culture of the period.
Clausen was shown the location of the above bones by Royal when at the
spring in January 1972. These specimens were located in Zone 3 sediments
at a point along the northern wall of the spring where a number of large irregular
blocks of wall material and large stalactites had fallen from above and lodged on
the 13 m ledge. A human mandible lay in full view in an irregular tunnel-like
hold dug by Royal in the leafy deposit underneath one of the large blocks of
limestone. The skull later recovered from this point in early February 1973
(and reported extensively in the popular media) was not in evidence, but portions
of other human bones were clearly visible in the surrounding sediment. (The
mandible and some of the other human bones from this location were brought
to the surface in November 1972) .
Royal, exploring in the spring as he has for years, had apparently dug
back under the rock to his left (or west, when facing the wall of the spring) ,
and encountered the bones. In Clausen1 s opinion, there was some evidence,
in the form of fine particulate carbonate material below the mandible, which
suggested that this bone was no longer in situ, although he did not question the
positions of the other bones. No evidence of articulation or any other evidence
indicating this concentration of bones represented a burial were observed.
Nothing further was done with this find in January 1972 because the location in
the sediment of the concentration of bone, practically at arm's length under a
large rock estimated to weigh 600-1,000 kilos, presented a more difficult area
to excavate than the tested location. It also appeared that the Zone 3 sediments
at this point were irregular or distorted, perhaps due to the presence of the
rock fall, and that more useful information on these strata could be obtained more
easily from the test area.

Warm Mineral Springs is not the only collapsed, water-filled sink or
spring in Florida containing skeletal remains of Paleo-Indian or later Archaic
period inhabitants. It has long been known that the basin area of nearby Little
Salt Spring, 8S0I8 (mentioned earlier), contains large quantities of human bone.
Goggin reported in I960 that the remains of more than 50 individuals had been
recovered from the sloping basin of that spring by Royal and Clark. In the
absence of datable artifacts he theorized that cleaned, unarticulated bones had
been thrown into the spring as a local variant of secondary burial in mounds
(Goggin 1960:352). However, he was puzzled by the absence of pottery which
very likely would have been present if this supposition were correct (Goggin
The preliminary results of recent research at Little Salt Spring (planned
for later publication) suggest that rather than water-deposited secondary burials,
the bones, which are found in a stratum of fresh-water marl (Fig. 12) accompanied
by lithic, shell and wooden artifacts and apparent food remains, reflect some form
of occupation of nearby portions of the now-flooded basin of that spring. This
would have occurred during a period when, as a result of a minor drop in sea
level, the spring had ceased to flow and represented a hard-water pond with a
probably seasonally fluctuating water level 4 to 8 m lower than present (Clausen,
Brooks and others 1974) A radiocarbon date of 5,220 90 years B. P. or 3270
B. C. (Gak-3548) was obtained on the collagen remaining in a 600 gm sample of
human bone obtained from the spring by Clausen. Projectile points found with
the human bone recovered in 1959 are the distinctive stemmed Archaic variety
known as Newman points (Bullen 197 5:31) which are extensively distributed
throughout the northern two-thirds of the Florida peninsula. They were first
isolated at site 8A356 near Gainesville (Clausen 1964a). These points have been
dated at the type site, A-356, at 5,900 130 years B. P. or 3950 B C. (Gak-2031)
and 6,010 150 years B. P. or 4060 B. C. (Gak-2930) (Hoffman and Clausen 1971);
and by R. P. Bullen (1975:31) at approximately 3400 B. C. (averaging samples
M-1 264/5/8/70) at Tick Island on the St. Johns River where they were encountered
with a large number of preceramic burials (A. K. Bullen 1972:166).
Another related sinkhole feature where human bones have been found is
Devil's Den, 8Lv44, in Levy County some 280 km north of Warm Mineral Springs.
Similarly unarticulated human remains were encountered in submerged stratified
sediments now 8-11 m below recent normal water level in and near a cave-like
fissure located on the southeastern side of the central solution cavity. The ele
vation of these sediments is approximately 12-15 m above present MSL.
The upper two strata of this deposit consisted of clay from breakdown of
the limestone walls and contained the remains of at least 60 different species
of vertebrates representing fish, birds, and mammals (including man) in two
generalized bone beds. Although the vast majority of the faunal remains were

8 So 18 TEST 2 2.0
Fig. 12. Photomosaic of profile of northeast face of Test 2 at Little Salt Spring.
Human skeletal remains visible in situ on pedestals in the excavation
were found in a fresh-water marl zone and date from approximately
5,500 to 5,000 years ago.

modern varieties, several extinct species, notably giant ground sloth, a species
of giant tortoise, dire wolf, and two varieties of short-faced cave bear were
present with the modern fauna in the lower bone bed. Artifacts including an
assemblage of bone pins and several projectile points were encountered in the
sediments, both on exposed ledges within the cave and in and on the talus at the
mouth of the cave (Clausen 1964b).
These upper two strata were interpreted by Brooks as originating during
a period when a fluctuating local ground water level had risen to a point where
the roof and walls of the cave, and evidently at times the deposit itself, were
alternately exposed and inundated. Rodent teeth marks were a common feature
on many of the bones; however, one character of the matrix indicated subaqueous
breakdown of the cavern wall rock. The lowest stratum, a red clay, contained
no faunal remains and was typical of material washed into and deposited in dry
caverns in the area (Brooks 1961:80-85). Radiocarbon dates on the carbonate
fraction of samples of deer and bear bones selected by Brooks from the upper
and lower zones of bone concentration in the two highest strata were 6,975 180
years B.P. or 5025 B. C. (Gx-0637) and 7,045 1 85 years B P. or 5095 B. C.
(Gx-0638) respectively.
Besides the similar geological origin shared by the two sites described
above and Warm Mineral Springs, the single factor these three sites have in
common is that the sedimentary and archaeological evidence in them indicates
that humans were associated with all of them when both sea level and/or local
ground water levels were markedly lower than present. Moreover, all three
of the features penetrate the terrain to an extent which insures that if ground
water is at all available it can be reached in their depths, albeit not without
some difficulty and danger.
Brooks has amassed a considerable body of information on the recent
geology and climate of the Florida peninsula from investigations of swamp,
lake, spring, and sinkhole deposits. These data indicate that during periods
of a colder climate in continental temperate regions of the world (associated
with increased glaciation and lower sea level) the peninsula, particulary the
southern coastal portions, experienced severe aridity. Conversely, during
warmer periods, for example the Two Creeks Interstade, when sea level was
higher, the evidence suggests that relative rainfall on the peninsula increased
(Brooks 1973:558-559; Gleason, Cohen and Brooks 1974:309-311).
Today it is possible to obtain potable water in virtually any portion of
Sarasota County simply by digging a shallow pit. However, with a much
drier climate than now exists there probably would have been little or no
perched water, including stream run-off, available. The 10-13 m lower
water level we postualte for this area 9 to 10 thousand years ago would have

put the sub-surface water supply beyond the reach of even the impressive
well-digging capacities demonstrated by the Orange or fiber-tempered ceramic
period inhabitants of the South Indian Fields site, 8Br23, near the headwaters
of the St. Johns River (Rouse 1951:100-103), Under such conditions, the huge
rock-lined, essentially inexhaustible, sources of very hard, although potable,
water that Warm Mineral Springs and Little Salt Spring represented would have
been very attractive to humans. These natural wells or cenotes might have been
the only dependable sources of water for tens of kilometers in any direction.
With prolonged drought conditions, the water in Warm Mineral Springs and Little
Salt Spring would have become the enticing bait in two huge funneled traps, with
insloping slippery walls, into which man and other animals could have fallen
with little chance of escape.
The pins found above the drop-off at Little Salt Spring are proof enough
that early man visited these sites, probably to obtain water or to remove game
that might have been trapped in thm. These pins were probably installed as an
anchor point for some sort of braided skin or twisted fiber line attached to a wooden
or skin container to prevent its accidental loss while "bucketing up" water from
the surface of the cenote. The pins might have secured a heavier line on which
a human might have been able to descend to the water surface (if the disparity
in elevations were not too great) to retrieve drowned or subdued animals.
It could be argued by those who support the theory that man lived in these
features that the pins constitute evidence for man entering the cavity to reach
the ledges which also exist at Little Salt Spring in order to set up housekeeping.
This proposition is unacceptable, however, for two reasons. First, a line hanging
vertically down from the lip of the drop-off where the pins were located does not
come near any ledges. In fact, the line misses both lower ledges in that spring
by many meters horizontally. Secondly, if our water level projections for this
area at the time the pins were emplaced are correct, the two principal ledges
of Little Salt Spring at depths of 21 and 27 m below the water surface would have
been 5-8 m underwater in the case of the shallower ledge and 11-14 m below the
surface in the case of the deeper ledge. The latter is the only one which would
even support a man (in one area it is 7-8 m wide). Admittedly, if the water were
at or just below the lower ledge, which it undoubtedly was during at least one
glacial period (judging by the extinct vertebrate remains found there by Clausen:
1972a; 1972b), perhaps at the height of the Wisconsin, one could let go of the line
and swim to the side, although getting back up the approximately 1 6 m of hanging
line to the edge of the drop-off, unassisted, would be no small accomplishment.
Although we believe that Paleo-Indian and later Archaic peoples' principal
use of these features was as a source of water, the sites could also have served
as a source of food. During investigations at Devil's Den, the remains of an
extremely wide range of both extinct and extant vertebrate species were encountered.

It was clear that for the last 7 to 8 thousand years that site had functioned as
a passive, though extremely effective, trap, collecting a relatively complete
and substantial sample of the indigenous fauna. The vertebrate remains found
at Warm Mineral Springs indicate that this site also collected animals, probably
in substantial numbers in the case of the smaller vertebrates.
It should also be pointed out that the effectiveness of such sites as poten
tial food sources, particularly in securing deer, could have been considerably
enhanced by judicious clearing of approahces to the cavity through the vegeta
tion which the research at Little Salt Spring indicates grew in the basins of these
sites (Clausen, Brooks and others 1974), camouflaging the edge of the drop-off
to drive animals into the cenote. It is also probably that as Warm Mineral
Springs functioned as a natural trap for other animals it could be expected to have
taken a toll of humans over the thousand or so years that the Zone 3 sediments
were forming.
The presence of human skeletal material in Warm Mineral Springs is best
attributable to accidents--slips, as it were--befalling early inhabitants either
when frequenting the area in search of water or trapped game or for some other
purpose, or possibly even to humans who like many of the other animals simply
blundered up to the edge, discovering the site when they fell in. Once in, whether
by accident or possibly due to failure of an entry system, without assistance from
other humans their demise was almost assured. They would drown or, if the
water level were low enough, perhaps subsist for a time clinging to or perched
on rocks extending above the water surface, until death from exposure or star
vation relieved them of their plight. Their remains would settle along with the
other organic material to the bottom or into rock crevices, ofttimes to be dis
articulated by the scavenging activities of fish or other trapped animals, and
finally would be incorporated in the sediments.
Another possible explanation which merits discussion is that the remains
do in fact represent burials, although not primary interments, in dry or moist
sediments or in crevices on the ledge (as some Southwestern Indians were known
to do). The cenote may have represented a simple, expedient means of disposing
of the dead or its dark waters may have had a place in the mythology of these
early people. Relatively large quantities of human remains from later periods
have been found in dry rock chimneys and solution pipes in both Texas (Curtis
Tunnell, Texas State Archeaologist, personal communication) and Alabama (L.
Ross Morrell, Florida State Archaeologist, personal communication).
Because the overhang at Warm Mineral Springs shields most of the level
area of the 13 m ledge from direct deposition, it is likely that if burial is the
explanation for the bones in the spring that intact bodies were thrown into the
water. If bundles of bones or single bones were introduced into the water they

would tend to sink directly downward, settling on the lower sloping portion of
the ledge and sliding off into the depths of the spring, missing the area where
the sediment was accumulating, Bodies, on the other hand, depending on the
length of time that had passed between death and entry into the spring, might
initially float, or if they sank, might return to the surface buoyed up by the
formation of bacteria-generated gases in organs and body cavities. In either
case, a large percentage of these bodies would find their way to the wall of the
spring below the overhand to decompose and sink. The same situation would
prevail in the case of accidental drownings in the spring. On the other hand,
bundles of bones (if dry and wrapped in skins or perhaps buoyed up by attached
wooden artifacts) might float for awhile, finding their way under the overhand
to settle on the ledge.
Among the more remote explanations for the presence of the human skele
tal material in the spring which should be mentioned are infanticide, or some
form of sacrifice. Systematic infanticide is assumed to have characterized
human populations throughout the Pleistocene (Birdsell 1968:236, 237, 239). The
fact that 6 of the 7 individuals represented by the various skeletal elements re
covered in 1959 were adults (Royal and Clark 1960:285), and the indicated chron
ological age of approximately 6 years for the two bones of the individual recovered
in our 1972 excavation, argue strongly against infanticide. We can only speculate
on the possibility that some form of sacrifice may have been involved. The ex
treme environmental pressures to which these people may have been exposed
could have been expected to engender a wide range of cultural responses of which
sacrifice might have been a part. We need only look to Yucatan for an example
of later period North American aboriginal inhabitants, the Mayans, throwing
people into geologically related semi-water-filled wells, the cenotes of Chichen-
Itza (Tozzer 1957).
Until a sufficient and unbiased sample of the skeletal material on the 13
m ledge has been collected in an orderly manner and subjected to examination
by competent authorities (a task made difficult or possibly even impossible due
to the rampant destruction of the sediments on the ledge over the past 16 years),
we will not be in a position to determine the true explanation for the bones in the
Analysis of the skeletal material and artifacts already recovered, first
by Royal and Clark and later by Royal who worked for years alone in the spring,
will be of minimal value in searching for an explanation (Fig. 13) In addition
to the rudimentary recovery techniques, the recoveries were not cataloged. For
10 to 15 years the human remains and artifacts recovered from both Warm Min
eral Springs and Little Salt Spring were stored in the same room at Warm Mineral
Srpings in open boxes or occasionally laid out together on tables for casual exam
ination by visitors to the spring. This practically guarantees that these collections
separated in time of origin by almost 5,000 years, are now mixed.

Fig. 13. One of 7 human skulls found by William Royal while digging through
the 10,000-year old sediments on the 13 m ledge at Warm Mineral
Springs. Unfortunatelym no prevenience exists for these skulls
or for the unarticulated remains of more than 20 individuals found
and brought to the surface before 1971,

The ease and success with which adequate excavational controls were
established for Clauson's test at Warm Mineral Springs justify the confidence
of Jewell (1961) that archaeology in submerged fresh-water sites is feasible.
It should give pause to those who have maintained that the untold thousands of
archaeological sites inundated in the many hundreds of reservoirs created in
this century constitute a cultural resource lost forever to man. We should
consider the possibility publicly stated in 1972 (Clausen 1972b) and more re
cently by Cummings (1975; and personal communications, 1972 and 1973) that
these drowned sites, rather than being lost, may figuratively speaking be in an
archaeological "bank, 11 in effect safely locked away from the destructive effects
of continuing human expansion and growing disruption by collectors which have
characterized the last few decades.
Of even perhaps greater potential as a repository of information relative
to our understanding of the climatic record in this area and the reason for the
human remains in the spring, is the mass of unconsolidated, but stratified,
largely organic sediments comprising at least the upper portions of the central
debris cone in the depths of the spring. Material has been continually settling
to the bottom of Warm Mineral Springs from the time the cavity first opened to
the surface to the present. Human skeletal remains are known to exist in this
deposit, which at the bottom undoubtedly antedates the approximately 9-to-10
thousand-year-old span of the Zone 3 sediments on the 13 m ledge by thousands,
perhaps tens of thousands, of years.
Unfortunately, the depths of 38 to 50 m present a formidable obstacle to
efficient, safe excavation by archaeologists operating from the surface. Mean
ingful, attentive, technical work at depths--utilizing repetitive dives on com
pressed air from the surface--is difficult due to the impairment of faculties
experienced because of the effects of the nitrogen and even possibly the oxygen
fraction of air breathed under pressure. This impairment often goes unrecognized,
even by the divers themselves, but tests conducted in hyperbaric chambers have
shown significant drops in mental acuity under simulated dives to these depths
(Denny Bowman, Marine Biomedical Institute, University of Texas Medical Branch,
personal communication). In the opinion of the writers, effective excavation of
these deeper deposits can with safety and acceptable results only be undertaken
using saturation diving techniques.
We are indebted to the Warm Mineral Springs Corporation for expending
the funds necessary to obtain the radiocarbon assays on the samples collected in
1972; to the General Development Corporation for providing substantial funds with
which to carry on investigations at Little Salt Spring, a part of the results of

which were included in this report; and to Bates Littlehales, underwater photo
grapher National Geographic Society, who kindly granted permission for use
of the photographs in Figs. 8-9. Dr. Fred Thompson, malacologist, Florida
State Museum, assisted in the identification of the snail shells and Dr. Albert
Laessle, Department of Biology, University of Florida, helped with the botanical
References Cited
Birdsell, Joseph B.
1968 Some predictions for the Pleistocene based on equilibrium
systems among recent hunter-gatherers. In Man the hunter,
edited by Richard B. Lee and Irven DeVore, pp. 229-240.
Aldine, Chicago.
Breder, C.
A note on preliminary stages in the fossil ization of fishes,
Copeia, No. 2, pp. 132-135. Baltimore.
Brooks, Harold K.
1961 Geology in Devil's Den. Part of Devil1 s Den, an early
recent underwater cave, a multi-disciplinary presentation
before the 1961 Southeastern Archaeological Conference,
Gainesville, Fla.
1973 Holocene climatic changes in peninsular Florida. In Abstracts,
1973, Annual Meetings of the Geological Society of America,
Vol. 5, No. 7, pp. 558-559. Dallas.
Bullen, Adelaide K.
1972 Paleoepidemiology and distribution of prehistoric treponemiasis
(syphilis) in Florida. Florida Anthropologist, Yol. 25, No. 4,
pp. 133-174. Gainesville.
Bullen, Ripley P.
1975 A guide to the identification of Florida projectile points, re
vised edition. Kendall Books. Gainesville.
Clausen, Carl J.
1964a The A-356 site and the Florida Archaic. Unpublished M.A.
thesis. Department of Anthropology, University of Florida,
Devil's Den. Abstracts, p. 10, 1964 Meeting of the Society
for American Archaeology, edited by Joffre L. Coe. Chapel

1972a Little Salt Springs yields clues to the past. Florida Division
of Archives and History Newsletter, Vol. 3, No. 2. Tallahassee.
1972b Underwater excavations at Little Salt Spring and Warm Mineral
Springs, Florida. Research report before the 1972 Meeting of
the Society for American Archaeology, Bal Harbour, Florida.
Clausen, Carl J. Harold K. Brooks, and others
1974 Investigations at Little Salt Spring, 8S0I8, Florida. Manuscript
in preparation, multidisciplinary report of research carried out
1971-72, in possession of the authors.
Cockrell, W. A.
1974 Current status of early man investigations at Warm Mineral
Springs, Florida ( 8Sol 9) Paper presented at 1974 Conference
for Historical Archeology and the Conference on Underwater
Archeology, Berkeley.
Cummings, Calvin R.
1975 Summary of National Park Service underwater archaeological
activity. Paper presented at the Underwater Symposium, 1975
meeting of the Society for American Archeology, Dallas.
Fellows, Robert
1962 Warm Mineral Springs, Bayplay Magazine, January 1962. Tampa.
Gleason, P. J. A. D. Cohen and H. K. Brooks
1974 The environmental significance of Holocene sediments from the
Everglades and saline tidal plain. In Environments of South
Florida: Present and Past, Memoir 2, edited by Patrick J.
Gleason. Miami Geological Society. Miami.
Goggin, John M.
I960 Underwater archeology: its nature and limitations. American
Antiquity, Vol. 25, No. 3, pp. 348-354.
1962 Recent developments in underwater archeology. Newsletter of
the Southeastern Archeological Conference Vol. 8, pp. 77-88.
Griffin, John W.
1974 Investigations in Russell Cave. Publications in Archeology No. 13
National Park Service, U. S. Department of the Interior. Washington.
Haynes, C. Vance
1969 The earliest Americans. Science, Vol. 166, pp. 709-715.

Hoffman, C. and Carl J. Clausen
1971The A-356 site and the Middle Florida Archaic. Manuscript
in preparation, in possession of C. Hoffman, University of
Northern Arizona, Flagstaff.
Jewell, Donald P.
1961 Fresh-water archaeology. American Antiquity, Yol. 26, No. 3,
pp. 414-416.
Rouse, Irving
1951 A survey of Indian River archaeology, Florida. Yale University
Publications in Anthropology No. 44. New Haven.
Royal, William, and Eugenie Clark
I960 Natural preservation of human brain, Warm Mineral Springs,
Florida. American Antiquity, Vol. 26, No. 2, pp. 285-287.
Tozzer, A. M.
1957 Chichen Itza and its cenote of sa.crifice. Memoirs of the Peabody
Museum, Harvard University, Vols. 11, 12. Cambridge.
U. S. Department of the Army Corps of Engineers
1944 Myakka River quadrangle 7. 5 minute topographic map.
U. S. Geological Survey
1972Report: Warm Mineral Springs, Florida: water chemistry,
Ocala, Florida.
Wormington, H. M.
1957 Ancient man in North America. Denver Museum of Natural
History, Denver Colorado.
Yezdani, G. Habib, and E. S. Deevey Jr.
1973A report on the microfossils of Fittle Salt Spring. Manuscript
in possession of the authors (Clausen et al)
July 5, 1975
American Institute of Nautical Archaeology, Philadelphia
Department of Geology, University of Florida, Gainesville
Christ College, University of Cambridge, England

PUBLICATIONS of the Society appear at irregular intervals as funds permit. These are
provided by Sustaining Membership dues, special gifts, or subsidies provided by the au
thor. PUBLICATIONS are distributed to all members of the Society on issuance. Addi
tional copies may be secured as follows:
1. Two Archeological Sites in Brevard County, Florida, by Hale G. Smith,
31 pp. 2 figs. 4 pi. 1949. Out of print.
2. The Safety Harbor Site, Pinellas County, Florida, by John W. Griffin and
Ripley P. Bullen, 42 pp. 2 figs. 4 pis. I960. Out of print.
3. The Terra Ceia Site, Manatee County, Florida, by Ripley P. Bullen,
48 pp. 6 figs. 7 pis. 1951. Out of print.
4. The European and the Indian by Hale G. Smith, 150 pp. 6 maps,
1 plate, 1956. Out of print.
5. Florida Anthropology edited by Charles H. Fairbanks, 81 pp. ,1 chart,
4 summary articles, 18 pp. bibliography, 1958. Out of print.
6. Fiber-tempered Pottery in Southeastern United States and Northern Co
lombia: Its Origins, Context, and Significance edited by Ripley P. Bullen
and James B. Stoltman, 72 pp. 3 figs. 16 pis. 1972 $2. 00
7. Florida Spring Confirmed as 10,000 year old Early Man Site by
Carl J. Clausen, H. K. Brooks, and A. B. Wesolowsky $2.00
PUBLICATIONS 1-5 are out of print. Numbers 67 may be secured
from the Treasurer (see inside front cover).