Title: Remote sensing and soil science applications to understanding Belle Glade cultural adaptations in the Okeechobee Basin
Full Citation
Permanent Link: http://ufdc.ufl.edu/UF00102735/00001
 Material Information
Title: Remote sensing and soil science applications to understanding Belle Glade cultural adaptations in the Okeechobee Basin
Physical Description: Book
Language: English
Creator: Johnson, William Gray
Copyright Date: 1991
 Record Information
Bibliographic ID: UF00102735
Volume ID: VID00001
Source Institution: University of Florida
Holding Location: University of Florida
Rights Management: All rights reserved by the source institution and holding location.
Resource Identifier: oclc - 24915799
ltuf - AHZ0677

Full Text







Copyright 1991


William Gray Johnson


If not for the generous contributions of my friend and

confidant, Bob Ulshafer, this dissertation would not have been

accomplished. His camper provided a roof over my head while

I was in the field and his four-wheel drive vehicle got me out

of some rough country. Furthermore, his assistance in the

field was invaluable. Naming a site after him was a minor

reward for all his help. I only wish he had lived to see the

completion of this dissertation.

Jerald T. Milanich, my committee chairman, deserves

special praise as he was instrumental in focusing my attention

on the archaeology of the Lake Okeechobee Basin. His knowl-

edge of the area and his support for my project were outstand-

ing. The encouragement and support of committee members, Drs.

Mary Collins, Marvin Harris, William Keegan, William

Marquardt, and Michael Moseley, are also greatly appreciated.

Dr. Helen Armstrong, Chief Librarian of the Map and

Imagery Library on the University of Florida campus, was

especially helpful in assisting me with the aerial photograph

collections. Her generous contributions of time and knowledge

are applauded.

Support from Highlands Hammock State Park in the form of

free use of their facilities was greatly valued. Park


managers, Peter Anderson and Valinda Nichols, along with the

rest of the Park's personnel are thanked for their cooperation

and generosity.

Special thanks go to my informants for their time,

patience, and knowledge of the local geography. Jeff and

Debbie Clemens, Larry Luckey, John Pullen, Charlie Wilson, and

George and Peggy Wedgworth are among those who were especially

helpful. In addition, John Fitzpatrick and the staff at the

Archbold Biological Station are acknowledged for their role in

my documentation of the sites on Buck Island Ranch.

Finally, I want to thank the staff at the Florida

Department of State, Division of Historical Resources, for

believing in my project and the Florida Historic Preservation

Advisory Council for funding it. State Historic Preservation

Officer George Percy deserves the highest praise for his part

in securing project research funds.






1 INTRODUCTION . . . . . . . . .


Previous Research . . . . . . . .
Theoretical Perspectives . . . . . .
Theoretical Perspectives on Belle Glade Culture

. 1

S 6

S 13



Environmental Background . . .
Culture History . . . . .
Summary of Cultural Adaptations
Settlements . . . . .
Archaeological Assemblage .
Major Sites . . . . .
Settlement Patterns . . .
Social Organization . . .
Belief Systems . . . .
Subsistence . . . . .


Precolumbian Maize in South Florida
The Antillean Route: A Not-so-Likely
The Soils at Fort Center: Not P
Maize . . . . . . .
Fort Center . . . . .
The Circular Ditches . . .
The Maize Pollen . . .
Evidence of Complexity at Fort
Testing Sears' Model: A
Application to Ar
Interpretation . . .



Center .


Soil Science


. . . . . viii

. . . . . . . . . 2 6

. . . . 28

S. 55

Morphological Descriptions . . . .
Laboratory Analyses . . . . . .
Discussion . . . . .
Ethnohistoric Evidence: Conflicting Accounts .
Discussion . . . . . . . . .
Summary . . . . . . . . .

METHODS . . . . . . . .

Background and Literature Review . . .
Aerial Photographs . . . . . .. .
Pedestrian Survey and Informant Interviews .
Laboratory Analysis . . . . . .

6 RESULTS . . . . . . .

New Sites . . . . . .
Okeechobee County . . .
Highlands County . . .
Glades County . . . .
Hendry County . . . .
Updated Sites . . . . .
Highlands County . . .
Glades County . . . .
Hendry County . . . .
Negative Data . . . . .


Mounds . . . . .
Ditches . . . .
Borrows . . . .
Embankments . . .
Summary . . . .
Mound Groups . .
Circular Earthworks
Circular-Linear Eart
Linear Embankments
Borrows . . .
Chronological Summary

.hworks .

8 SUMMARY . . . . . .

Belle Glade Archaeology . .
Remote Sensing . . . . .
Maize in South Florida . . .
Theoretical Considerations . .
Future Research . . . .

* .
* .
* .
. .
* *
* .


. . . . 87




. . . . 180




REFERENCES . . . . . . . . . . 197

BIOGRAPHICAL SKETCH. . . . . . . .. 210


Abstract of Dissertation Presented to the Graduate School
of the University of Florida in Partial Fulfillment of the
Requirements for the Degree of Doctor of Philosophy



William Gray Johnson

May 1991

Chairperson: Jerald T. Milanich
Major Department: Anthropology

Remote sensing, in the form of early aerial photographs,

was utilized to locate and describe types of aboriginal

earthworks in the Okeechobee Basin. These data, combined with

a review of the relevant literature and the Florida Master

Site Files, provide the basis for a typological classification

of the earthworks. The typology is compared with chronologi-

cal information derived from the Fort Center site to provide

a synthesis of the culture history of the Basin. Data from

analysis of soils at the Fort Center site suggest that maize

played at best a minor role in the development of the Basin's

aboriginal cultures.



The archaeology of the Okeechobee Basin has been de-

scribed as "one of the most unknown and enigmatic in all of

Florida" (Milanich and Fairbanks 1980:181, see Figure 1) .

Although much has been done in the last decade (e.g., Hale

1984, 1989; Sears 1982), the nature of the aboriginal occupa-

tion of the Basin remains debated and, at least somewhat,

enigmatic. This is especially true of the extraordinary

earthworks found there.

This dissertation attempts to synthesize what is known

about the archaeological settlements in the region and provide

a framework for understanding the changes apparent in the

aboriginal earthworks associated with the Belle Glade culture,

the precolumbian and early colonial period peoples who

occupied the Basin. In addition, the question of maize

agriculture as a subsistence base for the Belle Glade people

is addressed through a study of soils at Fort Center.

Previous archaeological research in the Okeechobee Basin

has proved that precolumbian sites can be located by reviewing

aerial photographs for large earthworks, mounds and middens

(Carr 1975). However, this method can yield erroneous data if

not coupled with a pedestrian survey (cf. Clausen 1980).

Figure 1.

The Okeechobee Basin and Key Sites (Source:
Milanich and Fairbanks 1980:168).

. 0.


Thus, these methods were combined with informant interviews to

provide a comprehensive survey of this archaeologically little

known part of Florida.

Early black and white aerial photographs were reviewed in

order to find aboriginal sites within the project area.

Figure 2 provides an example. Photographs are available in

stereo pairs at the Map Library located on the University of

Florida campus. The earliest were taken in 1938 and provide

coverage of the area prior to the changes in vegetation that

resulted from extensive flood control. Large earthwork sites,

such as Fort Center, Tony's Mound, and others are easily

discernible. Mound sites, on the other hand, are not so

easily found, but can be identified with the aid of a stereo-


A ground-truthing pedestrian survey of the sites and

potential sites followed. Unfortunately, time constraints,

access to private properties, and limited manpower inhibited

my ability to ground-truth all of the sites and potential

sites found during the review of aerial photographs. Still,

the project resulted in the recording of 38 previously

unrecorded sites and provided new information for nine

previously recorded sites.

These data, combined with a review of relevant literature

and the Florida Master Site Files, provide the basis for a

discussion of the basic elements that make up the 40 known

earthworks in the Lake Okeechobee Basin, and their change from

Figure 2.

Early Aerial Photograph (BUN-5D-133) of
Tony's Mound (8Hn3).

one form or type to another. This study provides a modifica-

tion of Belle Glade chronology and sheds light on Belle Glade

origins. Also, the application of early aerial photography to

remote sensing archaeological sites is addressed.

The question of maize agriculture in South Florida has

been debated for decades. Maize pollen from the Fort Center

site provides the most convincing evidence to date of its use

among the Belle Glade peoples. However, an examination of the

soils at Fort Center, specifically the purported planting

surface, indicates that maize could not be grown there in any

substantial quantity. Chapter 4 details the maize question

and provides evidence to support this statement.


Previous Research

The foundations for our understanding of the Belle Glade

culture are attributable to John Goggin (1940, 1947, 1952,

n.d.a). His work on the Glades archaeological area incorpo-

rated research within the Okeechobee Basin as a sub-area.

Previously, Matthew Stirling (1936:355) designated the Glades

archaeological area as "the region between the Kissimmee and

Indian rivers and all of the peninsula from Lake Okeechobee to

the Florida keys." Stirling (1935, 1936) believed the entire

area was an archaeological unit that had been home for the

Calusa. Other than the exceptional quality of wood carvings

found by Cushing (1897) at the Key Marco site, he noted the

material culture of the area was "characterized by the use of

an inferior grade of pottery, perforated shell hoes, shell

plummets, antler adze sockets, and bone projectile points"

(Stirling 1936:355).

A.L. Kroeber (1939) considered South Florida to be a

distinct culture area in his study of cultural and natural

areas of North America. He too noted the quality of wood

carvings found by Cushing (1897) but, due to the lack of

additional finds, he characterized the culture as a poorer

phase of the Southeastern type.

Thus, Goggin's work in the area in the 1940s was, for all

practical purposes, a pioneering endeavor. His use of

Stirling's (1935, 1936) "Glades archaeological area" was

modified to include all of tropical Florida--with a northern

boundary stretching from Boca Grande Pass on the west coast to

just below Ft. Pierce on the east coast. Within this area, he

recognized three sub-areas: the Tekesta sub-area, the Calusa

sub-area, and the Okeechobee sub-area and three culture

periods: Glades I, II, and III (Goggin 1947:119-120).

At Fort Center, Goggin realized the distinctiveness of

the Belle Glade earthworks, noting that the site "has few

parallels" (1952:52). His analysis of the materials recovered

from Fort Center and from the nearby Platt site, coupled with

Rita Porter's (1952) analysis of Belle Glade Plain rim sherds

provided a ceramic sequence "remarkably like that from the

Belle Glade Midden" (Goggin 1952:65). In addition, he was the

first archaeologist to recognize the significance of aerial

photographs as a remote sensing technique for locating

archaeological sites in the Basin (Goggin 1952:52).

Goggin (n.d.a) also amassed a great deal of the ethno-

historic documentation on the South Florida natives. Writing

of the people living around Lake Okeechobee, Goggin (n.d.a:56)

states, "The affiliations of the people around Lake Okeechobee

are uncertain but it is possible that they were Calusa."

However, his article with William Sturtevant recognized them

as a distinct tribe called the Mayaimi (Goggin and Sturtevant


Gordon Willey's (1949) report on the excavations at the

Belle Glade type site and Big Mound City earthworks (conducted

by Matthew Stirling under the Works Projects Administration in

the 1930s) provides a cultural chronology specific to the

Basin. His scheme introduced the Belle Glade I and II

periods. Unfortunately, this scheme received little attention

from subsequent researchers.

William Sears' (1966:17-18) early work on culture areas

in South Florida distinguished the Okeechobee Basin from the

Glades archaeological area as a separate cultural region.

Later, Sears (1967) referred to the Okeechobee Area as the

Belle Glade area, a name change that Griffin et al. (1979:28)

found agreeable. Carr and Beriault (1984) prefer the term

Okeechobee Area but basically agree with the objective and are

not so concerned with the semantics. For the purposes of this

dissertation, Belle Glade area, Okeechobee Basin and Okee-

chobee Area are used interchangeably.

Outside the purely culture area designation arguments,

Sears' (1982) book on the excavations at Fort Center and his

articles (Sears 1971, 1977; Sears and Sears 1976) provide a

culture chronology distinct from previous work in the area.

His is a four-period chronology that separates cultural

developments at the Fort Center site. Sears (1982:193-201)

recognizes these developments from their relatively simple

beginnings to the later complex interactions between the Belle

Glade culture and the Calusa. Furthermore, he concludes that

terraforming and the presence of maize pollen at Fort Center

indicate that cultural development within the Basin was based

on maize agriculture.

The identification of precolumbian maize pollen at Fort

Center has led to vigorous debate among researchers interested

in the implications of maize agriculture among the South

Florida natives. Henry Dobyns (1983) uses this information in

combination with other lines of evidence to argue that

agriculture was practiced among the Calusa--a position that

enabled him to bolster his population density estimates for

the Calusa. Jerald Milanich (1987) responds to Dobyns

(principally because Dobyns argued that de Soto's landing was

actually south of Tampa Bay and in Charlotte Harbor) by taking

Dobyns' (1983) evidence to task and refuting each point.

However, Milanich (1987:180) admits that the Fort Center maize

pollen is clearly a major bit of evidence that cannot be

refuted. On the other hand, Milanich and Ruhl (1986) indicate

that maize may not have served as a dietary staple for the

Belle Glade culture.

William Keegan (1987) addresses the issue as it relates

to Sears' (1971, 1982) hypothesized route of movement for the

origins of maize at Fort Center. His research leads him to

believe that maize could not have been introduced to South


Florida via the Antilles until maize was already well estab-

lished in the Eastern Woodlands (circa A.D. 1000). However,

Donald Lathrap (1987) takes exception to Keegan's view, noting

that a preservation bias could exist in the archaeological

record especially if maize were used in its green corn form.

The question of maize agriculture among the Belle Glade

culture continues to generate much debate. In chapter 4 I

summarize current knowledge on this subject and present new

data from soil science applications to provide evidence that

favors Milanich and Ruhl's (1986) view.

Other research that addresses subsistence among the Belle

Glade culture includes Stephen Hale's (1989) dissertation and

a journal article (Hale 1984). Both focus on zooarchaeo-

logical analysis of faunal remains from Fort Center and other

sites within the Okeechobee Basin. Hale's research provides

baseline data on faunal resource utilization and addresses the

issue of maize agriculture. In addition, Hale provides an

analysis of Belle Glade settlement pattern and argues that

earthwork sites are in predictable locations.

Robert Carr's (1975) technical report on a survey of the

lake margin and a later article (Carr 1985) provide insights

on the circular earthworks of the area. Carr (1973, 1975,

1985) was instrumental in demonstrating the effectiveness of

early aerial photographs as a remote-sensing application to

study Belle Glade earthworks (although John Goggin used this

technique in the 1940s, as indicated by his notes on archaeo-

logical site file cards at the Florida Museum of Natural

History). Carr also provided some guidelines for research in

the Martin County portion of the Basin (Carr 1973).

In addition to these archaeologists, a number of re-

searchers working elsewhere have addressed issues related to

the Belle Glade culture area. John Griffin's (Griffin et al.

1982; Griffin 1988) work in Southeast Florida and the Ever-

glades and William Marquardt's (1986, 1988) and Randolph

Widmer's (1988) work in Southwest Florida have discussed the

relationships between these areas and the Belle Glade region.

Griffin et al. (1982) specifically consider the question

of maize agriculture among the Southeast Florida natives

because of Sears' (1982) evidence of maize use among the Belle

Glade peoples. They conclude that no evidence exists to

suggest maize was grown there (Griffin et al. 1982:230). They

do find evidence for trade between the peoples who inhabited

the Granada site and the Belle Glade culture in the form of

Belle Glade Plain pottery, but they cannot specify the items

that were traded from Southeast Florida to the Okeechobee

Basin (Griffin et al. 1982:392).

In his synthesis of the archaeology of Everglades

National Park, Griffin (1988:48-65) provides a comprehensive

overview of the spatial and chronological relationships

between the various culture areas of South Florida. In his

conclusions, he indicates the archaeology shows more affini-

ties with the Southeast coast than any other area (Griffin


Marquardt's (1986, 1988) examination of the Calusa social

formation indicates tributary relationships between the

various native groups of South Florida. Specifically, he

highlights the ethnohistoric documentation indicating that a

bread is made from roots in the interior of Florida and that

roots were brought to the Calusa on the Southwest Florida

coast. While the exact nature of the relationship between the

Calusa and the inhabitants of the Belle Glade area is not

clear, some form of distribution of goods from the Basin to

the coast is certainly suggested.

In The Evolution of the Calusa, Widmer (1988:274-276)

discusses the role of this bread in terms of demographic

distinctions between the Okeechobee Basin inhabitants and the

coastal Calusa population. Based on population estimates

provided by Fontaneda (1945), he determines at least a six-

fold difference in population size between the two groups. He

then argues that the demographic imbalance provided a military

advantage to the coastal Calusa that enabled them to force the

Belle Glade peoples to pay tribute for not being raided

(Widmer 1988:275).

These works provide a wide array of thought on the

archaeology of the Okeechobee Basin. In the next two sec-

tions, I examine theoretical perspectives as they have

developed in the discipline and as they have been applied to

the study of the Belle Glade culture.

Theoretical Perspectives

The archaeologists who have worked in the Basin have

arrived at markedly different conclusions about the cultural

developments that took place. Two schools of thought are

evident: historical particularism and cultural ecology. A

short history of these theoretical orientations within a

framework of cultural evolutionism provides a backdrop for

discussing the conclusions that have been offered and suggest-

ing alternative strategies.

Cultural evolution has been a prominent topic of study

for anthropologists since the beginning of the discipline.

Probably the dominant form is one that has been labeled

cultural evolutionism. Leslie White (1959:vii) states that

the theory of cultural evolution dominated most of cultural

anthropology during the last three or four decades of the

nineteenth century. It was advanced by scientists to replace

the theory of creation as provided by Judeo-Christian theol-

ogy. At its most basic level, cultural development was seen

as a linear progression from simple to complex. Lewis Henry

Morgan's (1877) work provided the stage model for the evolu-

tion of human society. He recognized three distinct stages

that all human groups progress through: savagery, barbarism,

and civilization.

However, during the latter part of the nineteenth century

and the early part of the twentieth century, a reaction

against evolutionism developed in both the United States and

Germany. In the U.S., Franz Boas led the attack and instilled

the ideas of cultural relativity in his students. In Germany,

the nonevolutionist school of thought was led by Fritz

Graebner and his colleagues. Their work promoted diffusion as

an interpretation of similarities between cultures in noncon-

tiguous areas, rather than independent development.

Rathje and Schiffer (1980:302) state that "the attack

against cultural evolutionism was led by Franz Boas, who

helped establish the conceptual scheme we now call historical

particularism. This perspective identified several basic

historical processes that were used to explain specific

instances of change." Among the processes identified were

migration, diffusion, and rarely, invention.

Bower (1986:66) points out that the "historical partic-

ularist view rests on the assumption that cultural traits, or

customs, are 1) independent of one another; and, 2) present in

any given culture as a result of fortuitous contact with donor

cultures and the preferences of the recipients." Unfortunate-

ly, such assumptions require that any explanation fit within

the parameters of individual histories and do not allow for

broad generalizations. Thus, historical particularism has

largely fallen out of favor.


According to White (1959:viii), antievolutionism reached

its peak during the 1920s, but began subsiding and set the

stage for a comeback of cultural evolutionary theory in the

late 1940s. He quotes the News Bulletin of the American

Anthropological Association (1948) as stating that at the

annual meeting of the Association at Albuquerque in December,

1947, "a reawakening of interest in the problem of cultural

evolution . was noticeable." A number of articles

followed on the theory of cultural evolution (White 1959:viii

cites Jacobs 1948; Hoebel 1958; Lesser 1952; Steward 1953,

1955; and Childe 1951).

White (1959:ix) states that he "absorbed the antievo-

lutionist doctrines of the Boas school as a graduate student.

But as he began to teach, he found, first, that he could not

defend this point of view, and later that he could no longer

hold it. In due time he cultivated the theory of evolution in

a course called Evolution of Culture,. . and has attacked

the position of the antievolutionists in a number of arti-

cles." White (1959) then summarized mid-twentieth century

cultural evolutionist theory in The Evolution of Culture.

Since then, cultural evolutionism has provided the basis

for most archaeological and ethnological work on societal

complexity (Gailey and Patterson 1987:3). This model views

societies as entities that develop through a linear progres-

sion of stages. White (1959) believed societal complexity

could be viewed as a sequence of evolutionary stages that


could be documented historically and archaeologically. He

defined each stage in terms of four components--technological,

sociological, ideological, and sentimental or attitudinal--in

which the technological component was the basis and determi-

nant of cultural systems (White 1959:18).

Julian Steward (1951:380), on the other hand, suggested

the term "levels of sociocultural integration" rather than

stages of cultural evolution. He states, "the concept of

levels of integration does not presuppose any evolutionary

sequence." Rather, it is "a methodological tool for dealing

with cultures of different degrees of complexity." He

referred to his approach as a multilinear theory of evolution

that "seeks cross-cultural regularities and explanations but

presupposes no universal schemes" (Steward 1965:733).

According to Hatch (1973:114), "the key to Steward's

ideas about culture is his concept of ecology." Steward

(1955:30) states, "The principal meaning of ecology is

'adaptation to environment.'" Thus, cultural ecology is the

study of the relationship between culture and the natural


Steward's model distinguished between the culture core--

those features most closely related to subsistence and

economic activities--and the secondary features of culture

that are not strongly tied to the core but "are determined to

a greater extent by purely cultural-historical factors--by

random innovations or by diffusion" (1955:37). Thus, it was


the relationship between the core and the environment that was

the most important determinant of cultural systems.

Gailey and Patterson (1987:3) state, "Cultural evolution-

ism differentiated society from the natural world, and culture

. . mediated the relationship between them." Furthermore,

they note that parallel developments in positivist forms of

Marxist social theory began focusing on the use of resources

and technology as determinants of cultural development. Both

schools saw societies developing through a linear progression

of stages, and stages were defined in terms of variations in

the base and corresponding superstructural forms (Gailey and

Patterson 1987:3).

However, problems in the positivist models began to

surface in the late 1950s (Gailey and Patterson 1987:3). Rowe

(1962) challenged the base-superstructure model. Leach (1954)

pointed out that linear models of development could not

account for the internal destruction of state institutions by

their subjects. And, Adams (1956) highlighted the reduction-

ist nature of state-formative models and believed they were

unable to account for the dynamics of historically specific


The results of these critiques led to the creation of two

distinct lines of thought: cultural materialism, as described

by Harris (1968, 1979), and a strand of evolutionist thought

which gave primacy to political dynamics (Fried 1967; Service

1962). The former retained the distinction between nature and


culture and continued to emphasize the primacy of the culture

core--what Harris (1979) calls the infrastructure. The

latter, on the other hand, has challenged the economic

determinism of the earlier evolutionists, but retained the

notion of linear development. The outcome of these lines of

thinking has been a slight revision in the sequence of stages,

i.e., to band, tribe, chiefdom, and state, as developed by

Service (1962).

During this period, anthropological archaeology accepted

evolution as an integral part of its program, largely through

the efforts of Binford (1962, 1965). The so-called "new"

archaeology stressed an emphasis on explaining processual

change scientifically. Explicit goals, the identification of

cultural and natural formation processes, and a move away from

detailed descriptions of artifacts and artifact categories

were its hallmarks.

Systems theory, as promoted by Kent Flannery (1968, 1972)

was an outgrowth of this movement. Borrowing largely from

ecology, systems theorists viewed cultural change as a process

tending towards equilibrium. According to Gailey and Patter-

son (1987:4-5), it retained the idea of determination by the

political level and erased the distinction between nature and

culture. This model abstracted individuals from historical

and social conditions and conceived of them as components of

a system, interchangeable with each other and with those in

other systems. Transformation of these systems is viewed as


a product of factors other than human action--especially human

action as shaped by class, gender, ethnicity, and other

dimensions of social life. Thus, culture, as created by

humans in a social division of labor, is irrelevant to the

analysis (Gailey and Patterson 1987:5).

During the 1970s, a modified form of the systems model,

called catastrophe theory, was adopted (Gailey and Patterson

1987:5). This form was created for dealing with crises.

Colin Renfrew (1978, 1979) provided some of the best examples

of catastrophe theory arguing that crises or catastrophes

occur in the absence of continuous political and economic

expansion. Gailey and Patterson (1987:18) note that the model

appeared in Great Britain following a period of profound

economic crisis.

At the same time, cultural materialism took on a more

important role in the development of anthropological theory.

Harris (1979) produced his seminal work on the subject leading

to vigorous debate on the primacy of the infrastructure in

determining cultural systems.

Interestingly, and almost simultaneously, two new schools

of thought emerged. Both sought to explain cultural develop-

ment from a Darwinian evolutionary perspective. These schools

of thought are sociobiology--as most profoundly stated by

Bateson (1985)--and cultural selectionism--as named by Rindos



Cultural selectionism argues for a Darwinian perspective

on the evolution of culture principally because it serves as

the basis for understanding evolutionary processes in modern

biology. Robert Dunnell (1980:36) states, "Modern evolution-

ary biology is relevant because it is in biology that the

scientific theory of evolution has been developed and because

biologists have argued its applicability to human phenomena

under the label of sociobiology" (e.g., Wilson 1975, 1978)

[italics original]. But perhaps the strongest argument for

supporting a Darwinian perspective on cultural evolution comes

from an approach that Dunnell (1980:36) has aptly named the

evolution of the "human capacity for culture." This approach,

while essentially biological, merges at its extremes with both

cultural evolution and sociobiology. Its basic tenets are to

be found in the assumption that the human capacity for culture

evolved within a biological context and, therefore, was

subjected to the same evolutionary forces that affect all

biological organisms. As such, this approach is necessarily

immersed in the Darwinian perspective.

One of the most recent advocates of the cultural selec-

tionists school of thought comes from Robert Boyd and Peter

Richerson (1985). Although trained as ecologists, they became

interested in the dearth of Darwinian models in the social

sciences. They noted the use of analogies from biology and

concepts such as adaptation, but were surprised by the lack of

any "systematic theoretical argument for cultural behavior

that paralleled the Darwinian theory of biologists" (Boyd and

Richerson 1985:vii). The result of their work, the dual

inheritance model, seeks to provide such a systematically

consistent argument.

This model views culture as the transmission of traits

from one generation to the next via the mechanism of social

learning. In this view, culture is defined "in terms of

people's mental state (the "emic" concept of anthropology or

the "cognitive" of social psychology)" (Boyd and Richerson

1985:36). The authors believe "that it is important to

exclude behavior and the products of behavior from the

definition of culture because behavior is contingent upon both

patterns of thought and feeling and environmental circumstanc-

es" (Boyd and Richerson 1985:36).

The importance of excluding behavior and the products of

behavior, i.e., traditional definitions of culture in materi-

alist paradigms, from their definition of culture is to

provide a focus on social learning as a mechanism that

evolutionary forces can act upon. In their words, "Only by

distinguishing culture from behavior can we see clearly how

social learning interacts with environmental contingencies to

produce behavior" (Boyd and Richerson 1985:36). This then

provides the basis for their dual inheritance model. The

cultural repertoire is analogous to the genotype and social

learning provides the means to produce phenotypes. It is

argued that social learning has the properties of an inheri-


tance system and that random variation, drift, and natural

selection provide the mechanisms of cultural variability.

Cultural materialists likely feel comfortable with the

cultural selectionist school of thought. It, unlike sociobi-

ology, does not invoke genetic control over the development of

human cultural systems but does provide a strong metaphor from

the natural sciences for explaining social science phenomena.

However, unlike the ecology school of thought from which

cultural materialism grew, cultural selectionism argues that

cultural evolution is the result of a natural evolutionary

process (Rindos 1984:38). Where cultural ecology views

culture change as a result of non-equilibrium conditions,

cultural selectionism argues that all human behavior is

determined by the interaction of two inheritance systems--one

genetic and the other cultural.

Rindos (1984:23) points out that there are weaknesses

inherent in the cultural ecology school of thought, especially

in the branch known as systems theory. He finds that even

though cultural ecologists take a basic materialist view of

the development of agriculture the weaknesses in the paradigm

force proponents to argue in terms of intent and teleology.

Intent emerges as the ultimate cause in explaining

origins or evolution of systems because systems theory forces

explanation into a course of action requiring choices. Once

choice is invoked, intent becomes an inherent part of the

explanation. Teleology results when explanations within

systems theory allow us to place emphasis on equilibri-

um/stability, resulting in difficulty explaining any change,

or on change, providing us with a situation in which the

mechanism of change is not inherently specifiable; thus we can

choose one that best fits our prejudices (Rindos 1984:24).

Cultural selectionism, on the other hand, seeks to

identify functional or proximate modes of explanation and

separate them from evolutionary or ultimate modes of explana-

tion. The emphasis is placed on variability, heritability,

and differential fitness in explaining changes in learned


Evolutionary models such as cultural selectionism may

prove beneficial for cultural evolutionists. However, their

application at this time is very limited. Among researchers

of the Belle Glade culture, materialist strategies, especially

cultural ecology, dominate our current understanding.

However, historical particularism is also strong.

Theoretical Perspectives on Belle Glade Culture

Historical particularism and cultural ecology are the

dominant theoretical approaches to the study of the precolum-

bian history of the Lake Okeechobee Basin. These views lead

to markedly different conclusions about cultural developments.

In this section, the various researchers and their approaches

are contrasted to provide a focus for the approach taken in

this dissertation.


Sears (1977:6-7, 1982:191) views the introduction of

maize agriculture and a settled village economy as having come

to South Florida's interior from Venezuela via the Antilles.

This introduction, according to Sears, comes in the form of

migration, the earliest of which was prior to 500 B.C. He

recognizes similar designs and similar manufacturing tech-

niques on the pottery of this time, as well as the terraform-

ing techniques that enabled savannah agriculture. As time

progressed, Sears believes that "the society here [at the Fort

Center site] developed a complex ceremonial center within a

century or two of the beginning of the Christian era"

(1977:7). By A.D. 1000, the ceremonial center was abandoned

and dwellings moved to small, single house mounds, each with

an adjacent linear earthwork up to 1200 ft long and 100 ft

wide. This, he believes, was due to continued contact with

the South American population (Sears 1977:7-9, 1982:200). To

further substantiate his claim of a South American origin for

cultural developments in South Florida, he cites Granberry

(1971; see also Granberry 1989 for a recent update) for the

origin of the Timucuan language (a language group that

dominated north-central Florida prior to European contact and

that is presumed to be similar to the language of South

Florida) from Venezuela.

Thus, for Sears, the migration of a South American

population via the Antillean route explains cultural develop-

ment in South Florida's interior. Such an explanation is


typical of a historical particularist strategy. It provides

a comprehensive explanation for the presence of complexity in

a given area without having to explain the reasons for the

development of that complexity in its original setting.

Bridging the historical particularist approach and the

cultural ecology school of thought is Robert Carr's (1975,

1985) research in the area. In his study of the area's

circular earthworks, he invokes a diffusionist argument for

their origin in Southeast Florida (Carr 1985:300). On the

other hand, he notes that all of the circular earthworks occur

in inundated environments that are located adjacent to

elevated hammocks or other upland environments. Thus, he

reasons that Sears' assertion that the earthworks functioned

for draining garden plots has to be rethought in terms of

other environmental contingencies (Carr 1985:299).

Gordon Willey's ecological approach focuses on a diet

based on wild foods (Willey 1949:71, 129). Moreover, he

believes that the environmental conditions of the area are not

suited for agriculture, especially in aboriginal times (Willey

1949:17). Indeed, he concludes "The area is not now, nor was

it in the past, well-suited for maize" (1949:18).

Stephen Hale (1984:183), who also takes an ecological

approach to the understanding of Belle Glade cultural adapta-

tions, has found that portions of the Belle Glade earthworks

are aligned with the direction of surface water flow and may

represent a method of drainage for agricultural production.

Though his work on the zooarchaeological materials from Fort

Center and other sites in the Basin provides abundant data on

the use of wild foods, he advocates an agricultural subsis-

tence for the Belle Glade peoples prior to A.D. 1545 (Hale

1984:183). Subsequently, Hale (1989:187-191) has retreated

from this position.

The question of maize among the South Florida natives has

been discussed at length. Most of these arguments center

around ethnohistoric documentation and/or archaeological

research. In all cases, the specific arguments can be said to

take a materialist approach [though Sears' (1971, 1982) work

on the subject is certainly couched in a historical particu-

larist strategy].


The conclusions about cultural development are typical of

the approaches represented. Historical particularists will

invariably see diffusion as the source for culture change and

will often invoke migration theories to explain why and how

such change took place. Ecologically-minded researchers will

cite the high productivity of the natural environment and

assume the development of complex social phenomena as a

natural outcome. Such ecological approaches fall neatly in

the cultural materialist school of thought, which is basic to

the research approach reported here. However, the present

study departs from the strictly ecological school in its

approach to adaptation.

Adaptation appears in the title of this dissertation for

two reasons. First, it emphasizes the relationship between

culture and nature, and second, it draws attention to a

materialist approach. In the former, adaptation does not

refer to an equilibrium between cultural and natural systems

as it so often does in ecological models. Rather, it refers

to the juxtaposition of a cultural system in concert with a

natural system. Adaptation in this view becomes a methodolog-

ical tool for a materialist paradigm. The following chapters

are consistent with this position and provide results that

will change interpretations of the Belle Glade culture.


Environmental Background

The Okeechobee Basin geographical area is bounded by the

Western Flatlands (Davis 1943:43-48), the Allapattah Flats and

the Big Cypress Swamp (DeLorme 1987:102,113-114). According

to Brooks (1974:256), the area of the lake is dependent upon

water level and varies from 560 to 730 square miles (1450 -

1890 km2). The total area of the watershed (including the

lake surface) is over 4500 square miles (11,655 km2). Of

this, the Kissimmee River accounts for about 3000 square

miles. The basin itself, excluding the lake, can be said to

encompass between 3940 square miles (10,200 km2) and 3770

square miles (9760 km2) depending on water level of the lake.

If one excludes the Kissimmee River, then the area of the

basin drops to between 940 square miles (2430 km2) and 770

square miles (1990 km2).

However, there is good evidence to include Lake Kissimmee

and the Kissimmee River drainage in this cultural area.

Precolumbian cultural affinities represented in burial mounds,

earthworks, and ceramics are found as far north as Lake

Tohopekaliga (Milanich and Fairbanks 1980:26). Similarly,

both the Kissimmee River drainage and the Basin are character-

ized by relatively poor, sandy soils (composed of quartzitic

sands developed over calcareous marine deposits), relatively

low elevations with little relief, and numerous ponds and

sloughs. Outcrops of silicified limestone, or chert, which

were often exploited by aboriginal peoples as raw material

sources for the manufacture of stone tools, do not occur in

the area (cf. Lane 1980; Lane et al. 1980). The closest known

outcrops lie to the west along the Peace River (Upchurch et

al. 1982). The vegetation cover consists primarily of pine

and palmetto flatlands, wet prairies and occasional hammocks

of live oak usually mixed with cabbage palm. Cypress swamps

are found bordering larger lakes and ponds and a wide assort-

ment of wildlife exists in the area.

Culture History

The region was first inhabited during at least the

Transitional period, c. 1000 B.C. 500 B.C. (Milanich and

Fairbanks 1980:23). Evidence for an earlier Archaic period

population is found nearby in the upper St. Johns "from north

of Palatka to well down opposite the Indian River region"

(Sears 1977:3). However, no sites in the Okeechobee Basin

have been assigned to this period. The earliest occupations

came from the Fort Center site (8G113), where semi-fiber

tempered ceramics have been found in limited numbers (Goggin

1952:58; Griffin 1952:329; Sears and Sears 1976:53; Sears


1982:26). According to Sears (1977:7; 1982:192-193), this

initial occupation is coeval with the construction of the

earliest of the Okeechobee Basin's spectacular earthworks--

i.e., the circular ditches.

Following the Transitional period is the Belle Glade

period, c. 500 B.C. A.D. 1700 (Milanich and Fairbanks

1980:23). During this period, cultural activities in the

Okeechobee Basin increase dramatically suggesting the greatest

cultural complexity in South Florida centered around this

inland territory (Milanich and Fairbanks 1980:181). Several

researchers have attempted to build a chronology for this area

but the lack of comprehensive surveys coupled with very

limited excavations leaves this region of Florida one of the

most poorly known regions of the state.

To date, the most comprehensive chronology for the Belle

Glade culture area was developed by Sears (1971, 1977, 1982)

based on his work at the Fort Center site. A combination of

stratigraphy, seriation, and radiocarbon dates allowed him to

develop a four-period chronology beginning with Period I and

ending with Period IV. These periods are roughly coeval with

the Transitional period as defined by Bullen (1959, 1970), and

the Belle Glade I and II periods as defined by Willey


Sears' (1982:26,185) Period I from c. 800 1000 B.C. to

A.D. 200 most closely fits the Transitional period definition.

At the early end of Period I, Sears (1982:192-193) notes small

populations of perhaps one or two families living along the

riverbank on small house mounds with fish and turtle being

particularly important to the diet. Only semi-fiber tempered

pottery was in use during this time but by the end of this

period most of the pottery was the sand-tempered plain ware.

Distinguishing this period from the Transitional period is the

presence of the circular ditches. Sears (1982:193-194)

believes the ditches functioned as drained fields for maize


Sears' (1982:26-31,186-189) Period II, from A.D. 200 to

about A.D. 600-800, is coeval with the Belle Glade I period,

except that there are no Glades decorated wares at Fort

Center. Griffin (1988:126) notes that sand-tempered plain

pottery is dominant at this time but is accompanied by the

appearance of Belle Glade Plain ware. The latter increases in

its popularity throughout Period II. Furthermore, trade wares

are present representing influence from Deptford, Crystal

River, Cartersville, Pasco and St. Johns areas to the north.

According to Willey (1949:125), evidence for the Belle

Glade I period consists of Belle Glade Plain and Glades Plain

(sand-tempered plain) ceramics with small percentages of

Glades decorated wares (see also Widmer 1988:87-88). There

are no defining dates associated with the Belle Glade I period

but, by inference, Widmer (1988:87) provides a terminal date

at A.D. 900 1000.

The Belle Glade II period is defined by the presence of

Belle Glade Plain and Glades Plain ceramics coupled with a

lack of Glades decorated wares and the appearance of St. Johns

(Biscayne) Check Stamped (Willey 1949:125). Sears'

(1982:27,31,189-190) Periods III and IV most closely resemble

this period. Period III lasts from A.D. 600 800 to A.D.

1200 1400 and is characterized as a period of time which

experienced minimal cultural change (Sears 1982:199). Griffin

(1988:126) indicates that sand-tempered plain pottery contin-

ued to decrease while Belle Glade Plain increased.

Period IV, on the other hand, is described as a time of

excelled craftsmanship, revived focus for authority, and

participation in a larger social system than the one existing

on the site. The introduction of linear embankments occurs in

this period as well as the first appearance of European

objects and artifacts made from European-derived materials.

In addition, Griffin (1988:126) notes that only a small amount

of sand-tempered plain ceramics accompany the large amount of

Belle Glade Plain ware. Furthermore, new rim forms, particu-

larly expanded flat and comma shaped varieties appear (Griffin

1988:126). Sears (1982:200) believes that during this period

Fort Center was part of the sixteenth and seventeenth-century

Calusa domain.

Little is known about the terminus of Belle Glade

culture. Few ethnohistoric accounts exist and only one is

widely publicized. Fontaneda's (1945:13) account of the

sixteenth century South Florida native groups refers to a

people who lived around a large lake known as the Mayaimi.

His narrative provides scanty information on population

density, subsistence, environment, and political affiliations

with the coastal Calusa. Some Belle Glade sites, including

Ortona (8G15), Belle Glade (8PB41), and Fort Center (8G113),

contain European artifacts and European-derived raw materials

such as silver, iron, and gold. Thus, there exists ample

evidence to suggest Belle Glade participation in a larger

economy than existed solely in the Okeechobee Basin.

The demise of the Belle Glade peoples was surely linked

to their participation in this larger economy. The introduc-

tion of European diseases is documented throughout the

southeastern U.S. and is considered a major cause in the death

of the South Florida native groups. It is thought by many

researchers that by the time the Seminoles entered into this

part of Florida, all of the South Florida aborigines were


Summary of Cultural Adaptations


Archaeological sites in the Okeechobee Basin include

earthworks (Allen 1948; Carr 1975:14-36; McGoun 1987, 1988;

Sears 1982:130-183; Willey 1949:73-77), burial mounds (Carr

1973:14; Sears 1982:130-183; Willey 1949:20-23; Williams

1975:15-34), and habitation mounds or middens (Goggin 1952;

Sears 1982:130-183; Willey 1949:19-20). Among the more

spectacular sites are the large earthworks found in the open

savannah, often bordering creeks or major environmental zones

(see Hale 1984:181). These include circular ditches, linear

embankments, and combined mound and embankment complexes.

Sears (1982:185) provides evidence indicating that the

circular ditches are the earliest earthworks at Fort Center,

dating earlier than 450 B.C. His research demonstrates that

some of the habitation middens are coeval with the ditches.

Habitation mounds are believed to have been built during his

Periods I and II (Sears 1982:186,195-199) and again during his

Period IV, but the latter include linear embankments (Sears

1982:199-201). Burials in mounds date to his Period I,

possibly II, and IV (Sears 1982:140-141,200-201). Willey

(1949:125) was unable to ascertain the date of the burial

mound at the Belle Glade site, but believed it was coeval with

the habitation mound. Williams (1975:33-34) obtained a

radiocarbon date of 200 B.C. under one of the burials he

excavated, but he believes the date is erroneous.

The combined mound and embankment complex known as Big

Mound City is thought to have been laid out according to a

master plan, although no conclusions have been made as to

whether it was built all at once (Willey 1949:73-77).

However, a collection from one of the mounds at the site

indicates it is coeval with the late period of occupation at

the Belle Glade site. At least 21 mound and embankment

complexes have been identified and include Barley Barber I

(8Mt19, Carr 1973); the Big Circle Mounds, also known as

Tony's Mound (8Hn3, Allen 1948); the Boynton Mound Complex

(8PB100 Jaffe 1976; McGoun 1987, 1988); Cowbone Prairie,

also known as Hendry Earthworks (8Hn25), L-8 Earthworks;

Lakeport Earthworks (8G126); Little Cypress Slough; Maple

Mound (8Hn5); Nicodemous Earthworks (8G19); Ortona (8G15);

South Mound City, also known as Big Gopher (Hale 1984:181);

Benchmark 24 Earthworks; Pepper Earthworks (Hale 1989); Fort

Center (8G113, Sears 1982); Kissimmee Circle Earthworks

(8G139); South Lake Mounds (8Hn33), and Summer Earthworks

(8Hn26). In addition, Johnson (1990) lists the Kissimmee

Circle (80b31), Lonesome Island (8Hg634), and Palmdale

Earthworks (8G176) as mound and embankment complexes.

Some of these sites have yielded European artifacts, such

as the Ortona Mound in Glades County (Goggin n.d.a:679),

indicating a colonial period occupation. Most, however, have

not been dated.

Archaeological Assemblage

The defining ceramic assemblage for Belle Glade culture

is the presence of Belle Glade Plain and Glades Plain wares.

Lithic tools such as stemmed projectile points are found at

some of the sites, e.g. Fort Center (Steinen 1982:75-76) and

Belle Glade (Willey 1949:34, Plate 6), but not at others, e.g.

Barley Barber sites (8Mtl9, 28, 29, Carr 1973; Williams 1975).

Wooden artifacts have been found at two Belle Glade sites,

Fort Center (Sears 1982:38-58) and Belle Glade (Willey

1949:53-59), and with further research in the area may be

found at others. The currently limited distribution of wooden

artifacts precludes them from being considered part of the

defining archaeological assemblage. Similarly, other artifact

categories have been found to have a limited distribution.

At the Fort Center site, Steinen (1982:68-102) has

identified the nonceramic artifacts. These include eight

morphological categories of sharks' teeth, six types of bone

artifacts, chipped stone points and blades, ground stone

celts, hones, and grinding stones, shell and stone plummets,

shell adzes, celts, axes, and gouges, and chert and quartz


The morphological categories of sharks' teeth represent

two classes of tools. The first is triangular-shaped,

asymmetrical, serrated, and has bilateral wear patterns, and

the second is tapering, usually symmetrical, and lacks cusps.

Similarly, the six categories of bone artifacts represent only

five types of tools. Of the five, only one has been classi-

fied into a functional type: fids.

Comparable nonceramic assemblages are known from only two

other sites in the Okeechobee Basin: the Belle Glade type site

(Willey 1949:34-53) and the Platt site (Goggin 1952:55).

Willey (1949:69) reports socketed and bi-pointed bone projec-

tile points, bone pins, shell picks, adzes, and hammers, shell

celts, conch shell drinking cups, shell and stone plummets,

stone pipes and tubes, and chipped stone projectile points

were recovered from Belle Glade. In addition, fossil shark

teeth, porpoise teeth, and possible bear teeth are in the

Belle Glade site assemblage (Willey 1949:45). At the Platt

site, Goggin (1952:55) found chipped stone projectile points,

shell picks, gouges, and celts, shell and stone plummets, bone

points and pins, and an antler adze socket. However, none

were found in sufficient quantity to indicate significant

temporal differences (Goggin 1952:55).

A recent paper on some of these artifact types, specifi-

cally the bone points and shell and stone plummets, indicate

fishing technology (Walker 1989). Long considered representa-

tive of terrestrial technology and/or ceremonial regalia,

Walker (1989) presents a convincing argument that these

artifact types represent two parts of a composite fish hook.

The importance of her argument for South Florida artifact

assemblages is to focus attention on fishing technology.

Chronological distinctions in Belle Glade ceramics have

been hampered by a lack of decorative motifs on surfaces.

Virtually all Belle Glade wares are plain. However, Sears

(1982:112) provides a seriation based on relative amounts of

plain ceramics. Semi-fiber tempered wares are earliest and

are gradually replaced by sand tempered wares. St. Johns

Plain pottery appears abruptly and relatively abundant at an

early date, but dissipates rapidly with the introduction of

Belle Glade Plain. Minor amounts of St. Johns Plain continue

to be used throughout the occupation at Fort Center. Belle

Glade Plain ceramics are initially a minor component of the

total ceramic assemblage, but become the dominant type at the

expense of sand tempered wares.

In addition to Sears' (1982:112) seriation of ceramic

types, he indicates that some technological distinctions are

related to chronological differences. Specifically, he

provides a seriation based on rim forms that shows flat lip

rims are the dominant form throughout much of the occupation

at Fort Center (Sears 1982:113). These are superseded by

expanded flat rims late in the occupation. This work appar-

ently builds on Porter's (1952) analysis of rim forms from the

Platt site.

The presence of monitor or platform type ceramic pipes at

Fort Center are thought to provide a good chronological

indicator because the pipes are said to be very similar to

those from Marksville sites (Sears 1982:117), known to be a

Hopewellian culture dating from Middle Woodland times. Sears'

(1982:117) associated radiocarbon dates place the pipes in his

Period II.

One other major category of Belle Glade artifacts that

serve as good chronological indicators are the artifacts

derived from European contact. These have been discussed by

Willey (1949), Sears (1982), and Leader (1985). Their

usefulness as indicators of sixteenth and seventeenth century

period activities has not gone unnoticed, but their origin and

form have been the focus of most analyses.

At the Belle Glade site, Willey (1949:59-61) identified

gold, silver, and copper beads, an iron spike and iron

fragments, a lead plummet, four types of glass beads, and

European-made ceramics. Sears' (1982:59-67) work with the

collection from Fort Center categorized the metal artifacts as

symbol badges, disc-shaped ornaments, a cast gold jaguar,

silver and copper plates, and miscellaneous gold and silver

pieces. Sears (1982:59) states, "It is apparent that the

metal specimens were made from metal of Spanish origin,

reworked in most cases into Indian forms." Leader's (1985)

work on the Fort Center collection focused on this statement.

In his quest to discern the origins of these materials he

grouped the artifacts by morphological characteristics into

the following categories: disks, domes, beads, hemisphere and

cones, plaques, and miscellaneous fragments. His conclusions

confirm Sears' (1982:59) statement (but noted that some of the

artifacts were actually made by South and, perhaps, Middle

American natives; the Spaniards were evidently shipping the

artifacts with other booty to Spain when the materials were

salvaged from shipwrecks by the Florida natives.)

Of all the artifacts made from European-derived materials

the symbol badges have drawn the most attention. The Fort

Center symbol badges are made of heavy slabs or ingots of

silver, 1/16 inch thick, cut into tablet forms and etched with

native designs. Sears (1982:60) states they are larger and

heavier than others found in South Florida.

Allerton et al. (1984) have summarized the distribution

of these badges and have found they are concentrated around

Lake Okeechobee. Leader (1985:81) states, "the ethnic origin

of the metal artifacts from the Fort Center metal collection

is American aborigine." Unfortunately, activity areas of

manufacture have not been identified. As such, no one has

been able to ascertain the culture group. However, their

known distribution clustering in the Lake Okeechobee Basin and

the largest and heaviest coming from Fort Center strongly

suggests they are Belle Glade artifacts and should be consid-

ered part of the defining archaeological assemblage.

Major Sites

Major sites in the Belle Glade culture area have been

identified by Carr (1975). His survey is one of the most

comprehensive listings of archaeological and historic sites in

the Lake Okeechobee Basin, but it was limited to areas that

were to be inundated by the Army Corps of Engineers proposal

to raise the regulation range of Lake Okeechobee from 13.5 -

15.5 feet above mean sea level to 15.5 17.5 feet. His

survey lists 16 precolumbian sites including mounds, middens,

and earthworks.

An even more comprehensive listing of precolumbian sites

in the Okeechobee Basin can be found on a map provided by Hale

(1984:181). However, even his figure does not include

earthworks such as Barley Barber I (8Mt19, Carr 1973:9-11),

West Okeechobee Circle (8G157, Carr 1985:295-297), and others.

As such, there is not a complete list of all major sites in

the Belle Glade culture area. To date, eight circular

earthworks, 17 circular-linear earthworks, nine linear

embankment sites, and six miscellaneous earthwork sites are

known to be in the Okeechobee Basin. Of these, major excava-

tions have been limited to the Belle Glade and Big Mound City

sites (both in the 1930s, see Willey 1949) and the Fort Center

site (in the 1960s, see Sears 1971, 1977, 1982).

Settlement Patterns

As for settlement practices, Sears (1982:184-190)

indicates a Belle Glade cultural adaptation to the open

savannah with artificially raised areas occupied during

flooding. The focus of occupation takes place at large sites

(such as Fort Center), though Sears (1982:175) acknowledges

occupation of "the small midden sites on higher land through-

out the sand country." On the other hand, Willey finds Belle

Glade affinities extending from the Okeechobee Basin to all of

South Florida, from the Florida Keys (Willey 1949:26 cites

Goggin n.d.a) to Melbourne (Ferguson 1951:36-39) and the

southwest coast (Willey 1949:124-125). However, he believes

the Belle Glade peoples differ significantly from the east

coast groups and states, "the most likely guess would be that

Belle Glade was a village of the Calusa domain, presumably

occupied by a people related to them" (Willey 1949:127).

Thus, his scenario depicts a wider range of adaptation to a

number of environments than does Sears.

Social Organization

A discussion of social organization provided by Sears

(1982:1919-201) indicates that an egalitarian population

inhabited the Fort Center site during his Period I. This was

followed by a marked degree of social stratification in

subsequent periods. Trade with other Florida groups is

apparent by the number of exotic artifacts he found. Further-

more, the presence of maize pollen and the earthworks suggests

to him that the Fort Center inhabitants had periodic contact

with a South American population (and, indeed, trace their

ancestry to this population). Inclusion of the Fort Center

peoples within the sphere of colonial period Calusa hegemony

is believed to have occurred in the final aboriginal occupa-

tion of the Fort Center site, i.e. his Period IV.

Goggin and Sturtevant (1964:180-182, 195-197) suggest

Calusa hegemony occurs early in the cultural development of

the South Florida native groups. Indeed, they hint at the

possibility of the Belle Glade peoples representing an inland

population of the Calusa culture. However, theirs appears to

be a minority view.

Belief Systems

Only a minimal discussion of beliefs has been presented.

Sears (1982:42,58) suggests that the Fort Center inhabitants

had strong totemic beliefs (as represented by carved wooden

figures of animals and birds). He also argues for a belief in

an afterlife (as represented by burial practices). Such

beliefs would not be surprising given the material remains

found in the Belle Glade culture area.


Based on what is known from the above, the diet of the

Belle Glade peoples has been interpreted to be either one

based on wild foods (Willey 1949:71,129) or one based on maize

agriculture supplemented with an abundance of wild foods,

especially turtle (Sears 1982:184-201). However, Walker's

(1989) recent analyses of fishing technology has alerted the

archaeological community to the potential contribution from

aquatic resources; an area of subsistence activities that

needs further exploration in the Okeechobee Basin. Hale

(1984:183) concurred with Sears (1982) noting that the

planting surfaces he identified are aligned with the surface

water flow facilitating drainage for agricultural production.

However, Hale (1989) has subsequently retreated from this

position. Analysis of the soils of one of the alleged

planting surfaces (reported hence) indicates it could not

support annual harvests of maize. In the following chapter I

explore the question of maize in detail.


Maize was obviously an important cultigen in the develop-

ment of certain New World cultures. Richard MacNeish (1970)

studied the evolution of this cultigen in the Tehuacdn Valley

of Mexico and concluded that settlement changes coincided with

increased reliance on maize. According to MacNeish

(1972:500), the use of more productive hybrids of maize

coupled with increased interactions between developing areas

caused changes in settlements rather than the reverse.

Similarly, Sears (1982) believes maize was an important basis

for cultural change within the Lake Okeechobee Basin of South

Florida. However, the presence of maize in South Florida, has

led to much debate over its use, importance, origins, and


Precolumbian Maize in South Florida

Evidence for precolumbian maize use in South Florida is

limited to maize pollen documented at the Fort Center site by

Sears and Sears (1976; see also E. Sears 1982). At the site,

Sears found numerous pollen grains in a variety of contexts

including at least one excellent precolumbian context dating

between A.D. 100 and 500. Strips of white pigment adhering to


a wooden carving that was found at the bottom of a pond

provided this context (Sears 1982:122; Milanich 1987:177).

Other contexts from which maize pollen was found are soil

samples from earthworks and coprolites. On the basis of size

and exine pattern, Sears and Sears (1976:54) distinguish the

maize pollen grains from other grasses. In addition, they

believe the size range of the maize pollen grains from Fort

Center eliminates the possibility that they could be modern

corn pollen.

In addition to the maize pollen, Sears (1982:171-173)

identified lime deposits at the Fort Center site that were

created by burning shell. He hypothesizes that the creation

of the lime was directly related to the processing of dried

maize into a mush (Sears 1982:173).

Sears (1982) believes the earthworks at Fort Center

served as planting surfaces for maize cultivation. The

origins of the maize and the earthworks are believed to be the

result of a migration of peoples from South America. Sears

(1977:6-7, 1982:191) suggests four possible routes for such

contact, but feels the most likely one was from Venezuela

through the Antilles (Figure 3). He notes similar ceramic

manufacturing techniques and stylistic elements on pottery

between South Florida and Venezuela, as well as similarity

between terraforming techniques (presumably related to

agriculture). His excavations at the Fort Center site

indicate the development of complex mortuary activities within

Figure 3.

Hypothesized routes for maize origins in
South Florida (after Sears 1971).

a century or two of the beginning of the Christian era

(1977:7). By A.D. 1000, the mortuary center was abandoned and

individual houses were placed on small, single house mounds,

each with an adjacent linear earthwork up to 1200 feet long

and 100 feet wide. This innovation, Sears believes, was due

to continued contact with the South American population

(1977:7-9, 1982:200).

Thus, for Sears, the migration of a South American

population with the knowledge of maize cultivation in wet

savannah environments explains cultural development in South

Florida's interior. This interpretation provides a comprehen-

sive explanation for the origin of the maize and the earth-

works but has left a myriad of questions in its wake.

The Antillean Route: A Not-so-Likely Course

Perhaps the most difficult problem with Sears' chronology

and origins of maize in South Florida is its early arrival, as

much as 1000 years prior to its presence in the Eastern

Woodlands. William Keegan (1987) addresses this issue and

concludes that present evidence indicates the introduction of

maize into South Florida via the Antilles could not have

occurred until maize was already well established in the

Eastern Woodlands (circa A.D. 1000).

According to Keegan (1987:331), human groups arrived in

the Greater Antilles by 4000 B.C. These peoples were aceram-

ic, non-agricultural hunter-gatherers who exploited terrestri-

al and marine resources. Their way of life changed with the

arrival of the agricultural Arawakan-speaking population. The

Arawakan-speakers, also known as the Tainos, began their

expansion into the Antilles from eastern Venezuela around 200-

300 B.C. (Haviser 1989). Their arrival into the northernmost

parts of the Antilles, i.e. the Bahamas and Cuba, was not

until A.D. 600 700. As with other parts of the Antilles,

the initial colonization period diet included cultivated roots

and tubers with terrestrial animals and a relatively minor

contribution from marine resources. Shortly after coloniza-

tion, a rapid shift to increased reliance on marine resources

occurred but root crops remained the dietary staple and maize

did not appear until late in precolumbian times (Keegan


Keegan (1987:336) concludes that the available evidence

indicates that the Antilles was not the source of this

cultigen in South Florida. However, he points out that this

conclusion concerns only one of four possible contact avenues

hypothesized by Sears (1977) for the arrival of maize in South

Florida and that little is known of the precolumbian history

of western Cuba. That region may yet be implicated in "a more

direct trans-Caribbean jump from Central or South America to

south Florida" (Keegan 1987:339).

The Soils at Fort Center: Not Productive for Maize

Regardless of Keegan's (1987:339) assertion that alterna-

tive routes of entry remain unexplored (allowing for the

possibility of a direct entrance of maize into South Florida),

a problem closer to the location of the pollen has been

discovered. Analysis of the soils from one of the circular

earthworks at Fort Center has yielded information on their

potential to support annual harvests of maize.

Fort Center

The Fort Center archaeological site is located in Glades

County, Florida (Figure 4). It was first reported by a

professional archaeologist when Goggin (1952) investigated it

and the nearby Platt site. At Fort Center, Goggin recorded

four middens, three small mounds, a large sand mound and its

associated platform, a large circular canal or ditch, a ridge

with refuse on its surface, and a restricted concentration of

nineteenth century artifacts (Goggin 1952:50). The latter

presumably mark the location of Fort Center, a small block

house established at the site during the Seminole wars. Also

recorded were several pairs of earth ridges, two with terminal

mounds, another small mound, and a long narrow ridge in the

prairie to the east. Goggin and Sturtevant (1964:197) suggest

that these earthworks, and similar ones elsewhere in the Lake

Okeechobee Basin, were built by the coastal-dwelling Calusa as

ritual centers.

Figure 4.

Location of Fort
Highway Map 1984).

Center (Source: General

Sears (1971, 1974, 1977, 1982), on the other hand,

suggests that the original inhabitants of the Lake Okeechobee

Basin (the Belle Glade culture) migrated from the northern

part of South America sometime before 500 B.C. According to

Sears (1982:192), continued contact with the mother population

resulted in the adoption of a variety of economic behaviors,

one of which was the practice of draining fields for producing

maize. He states "the economic system supporting the center

. acquired stability through dependable maize crops"

(1982:197). Sears believes this economic system was accom-

plished in the wet savannah environment by the creation of

planting surfaces, formed by the excavation of ditches, for

growing a stable, dependable maize crop. Initially the

ditches served to drain fields but later they were designed to

create raised fields. Hale (1984:183) suggests that this was

a response to rising water levels in the Basin. According to

Sears (1982:178,191-201), such a system sustained the economy

of these people for over a two thousand year period.

His evidence for this belief is twofold. First is the

presence of the circular and linear embankments at the Fort

Center site (Figure 5) and second, the identification of maize

pollen grains from a number of contexts at the site (E. Sears

1982:118-129), including at least one grain from a soil sample

from the Great Circle ditch. The former suggests to Sears the

ability of the inhabitants to have practiced drained and

raised field agriculture, while the latter provides evidence

Figure 5.

Schematic drawing of the Fort Center site
(after Carr 1985:290).

of the presence of maize. Combined with other evidence

regarding the social complexity of the precolumbian inhabit-

ants, these data form the basis for Sears' (1982:178, 191-201)

conclusions that the earthworks were used as agricultural


The Circular Ditches

Among the various earthworks at the Fort Center site are

two remnant circular ditches each of which have a diameter

approximately 140 m across. These are enclosed by a much

larger circular ditch referred to as the Great Circle (Sears

1982:175). Excavations revealed that the remnant ditches were

built prior to the Great Circle (Milanich 1968), but it is not

known how much earlier, nor which of them was dug first.

Radiocarbon dating of midden materials found in the base of

the Great Circle ditch date to 450 B.C. + 105 years (Sears

1982:116). Thus, Sears (1982:178) states all three circular

ditches were completed by then.

Sears (1982:6) believes these ditches functioned as

drainage ditches to enable food production. The basis for

this belief comes from his observation that "ditch bases were

dug below the hardpan so that water would drain across the

hardpan and into the ditch, drying the surrounding area"

(1982:6). He finds additional support for this belief in "the

need for drained garden plots if agriculture was to be

practiced in this environment and by the presence of maize

pollen in soil samples from points inside the ditches and the

[nearby] contemporaneous midden" (Sears 1982:186).

The Maize Pollen

Evidence for maize in the diet consists of maize pollen

grains found in coprolites, white pigment samples applied to

wooden carvings, and soil samples from several of the earth-

works at Fort Center (E. Sears 1982:120). The maize pollen

grains, specifically those from coprolites and the pigment,

lead Milanich (1987:177) to conclude "that maize was being

cultivated at Fort Center sometime during the period ca. A.D.

100-500." However, Milanich and Ruhl (1986:2) state that the

importance of maize within the Belle Glade people's diet is

uncertain. They conjecture that it might have been grown and

used for special purposes rather than as a dietary staple.

Evidence of Complexity at Fort Center

Sears' excavations documented numerous artifacts, long

distance trade, increased population through time, extensive

earthworks, and complex ceremonial activity areas. These

features suggest a fairly high degree of complexity when

compared to the contemporary cultures of the rest of South

Florida (Milanich and Fairbanks 1980:181).

Among the artifacts found at the Fort Center site are a

number of intricately carved wooden objects, including a

series of animal figures which formed an elaborate burial

platform. Similar wooden objects have been documented by

Willey (1949) at the Belle Glade site, also located in the

Lake Okeechobee Basin. Additionally, numerous wooden carvings

were uncovered by Cushing (1895, 1897) on the southwest coast

of Florida, but Sears (1982:58) believes they do not have any

real resemblance or relationship to the wooden objects from

Fort Center [but see Schwehm (1983) for an alternative

explanation Nevertheless, the craftsmanship of all the

carvings appears to reflect a degree of craft specialization.

Evidence for long distance trade includes the presence of

non-local artifacts including chipped stone tools, Deptford,

Crystal River, Cartersville, Pasco and St. Johns ceramic

types, marine shell tools, sharks' teeth and European-derived

items. All indicate participation in a wider economy than

just the Lake Okeechobee Basin. Most of these materials

indicate trade with coastal dwelling populations. Sears

(1982:32) suggests "at least fringe participation in the

Hopewell phenomenon" because of the presence of ceramic types

which co-occur with the strictly ceremonial Yent complex.

Sears (1982:32) also lists Deptford series sherds, Deptford

copies on chalky paste, and Crystal River series specimens as

evidence of the Yent complex at Fort Center. Milanich and

Fairbanks (1980:84) agree that similarities with materials of

the Yent complex suggests trade between the Woodland peoples

to the north (e.g., Adena, Hopewell, Cartersville, Copena,

early Weeden Island) and the South Florida peoples. However,

one must keep in mind that Sears (1962) defined the Yent

complex and may be biased in his interpretations.

Increases in population through time are also apparent at

the site, according to Sears (1982:193). He believes "that

there were not more than one or two families on the site at

any one time" during his Period I (approximately 1000 B.C. to

A.D. 200) but that number increases up to three-fold during

his Period II (Sears 1982:193-195). Moreover, Sears

(1982:197) suggests that this increase coincides with the

participation of additional peoples in the economic activities

at Fort Center. Numerous small sites throughout the savannah

region were occupied by peoples who, according to Sears

(1982:197), were supported by the economic system at the Fort

Center site. Finally, population density at the site drops

back to one or two families during the last period of occupa-

tion, Period IV (Sears 1982:200). However, Sears

(1982:200-201) believes that the inhabitants at that time were

part of the sixteenth- and seventeenth-century Calusa domain.

Thus, a considerable increase in the population which partici-

pated in the Fort Center economy can be discerned through


While most population estimates were derived by midden

deposits on house mounds, the burial population of the Mounds

A and B and charnel pond complex provided reinforcement for

Sears' estimates (Sears 1982:175). This mortuary area was

treated as a single unit by Sears because "one could not be

investigated without disturbing at least one other" (Sears

1982:147). Furthermore, an earthen embankment surrounding

Mound B "continues on at both ends to Mound A, thereby tying

both mounds and the pond into a unit" (Sears 1982:148). Thus,

Sears (1982:147-148) believes it is important to treat the

area as a single unit.

Mound A, the shorter of the two, yielded midden materi-

als, indications of two to four dwelling units, ceremonial

artifacts, and lime deposits. The midden materials contained

the usual assortment of faunal remains [although Hale

(1984:177) notes a larger amount of deer bone than found at a

similar deposit elsewhere at the site] and ceramic sherds but

also included shell tools, pipes, some stone grinding tools,

and a few other items with no "meaningful clusters or associa-

tions" (Sears 1982:171). The lime deposits, created from

burned shells, and the pipes, limited to this one mound (save

for the few fragments found at the UF Mound and Mound 3, Sears

1982:34), led Sears (1982:175) to believe that the people

living on Mound A were ceremonial specialists. Their services

"providing lime for treatment of their corn and probably pipes

and tobacco," are thought to have been performed for the

people who "lived elsewhere, in the small middens sites on

higher land throughout the sand country" (Sears 1982:175).

Mound B and the charnel pond, on the other hand, served

principally for mortuary activities. Approximately 150

burials were recovered from the pond along with numerous

carved wood figures, midden materials, and coprolites. The

wood figures are thought to be the remains of a charnel

platform that collapsed after a fire destroyed its structural

integrity (Sears 1982:167). Another 150 burials, many with

grave goods, and mud-streaked sand were documented in Mound B.

Sears (1982:157) believes the mud-streaked sand occurred when

skeletal remains were retrieved from the pond, after the

charnel platform collapsed, and placed in Mound B.

The extensive earthworks and complex ceremonial activity

areas documented at the site attest to a highly organized

population. The circular and linear ditches accompanied by

mounds, middens, and a charnel pond provide ample evidence of

large-scale work projects that could only be accomplished by

an organized labor force. Thus, it is not surprising that

Sears' discovery of maize pollen at the site led him to

believe that the inhabitants could only have developed such a

highly complex society on the stability of maize cultivation.

However, Feldman (1980:4) cautions that "once a trait has been

found in the archaeological record, any number of additional

features are often assumed to have been present from that date

forward. For example, the presence of maize at a site--

however small the amount recovered--is often assumed to

indicate that the subsistence of the site was based on

efficient maize farming."


Testing Sears' Model: A Soil Science Application to Archaeo-
logical Interpretation

Soil science applications to archaeological interpreta-

tion have concentrated on comparisons between natural soils

and human influenced soils (e.g., Collins and Shapiro 1987).

Such applications have proven useful in avoiding misidentifi-

cation of natural formations as cultural. In this test of

Sears' model of maize-based complexity, soil science applica-

tions are utilized to identify the limitations of the archae-

ologist's interpretations.

Soil samples were taken from one of the remnant circular

ditch formations in order to examine the potential of the

soils associated with the earthwork to support long-term maize

agriculture (Figure 5). Three column soil samples were

removed with a bucket auger. These bisected the ditch with

column #1 removed from the berm of the earthwork, #2 taken

from the bottom of the ditch, and #3 from outside of the

ditch. Each was excavated to 188 cm below the surface, the

maximum depth that the bucket auger could reach. These

resulted in a profile view of the ditch and its associated

berm (Figure 6).

Morphological Descriptions

Morphological descriptions of each field specimen were

conducted prior to preparing the samples for laboratory

analyses. These included horizon designation, moist Munsell





Figure 6.


Schematic profile view of Columns 1-3 (not
to scale).

-H Bw


color, structure (Soil Survey Staff 1981), and any other

observable characteristics (Table 1). Unfortunately, removal

of the samples with a bucket auger and immediate storage in

specimen bags resulted in some disturbance to them. Thus,

structure was reported only as to type. However, careful

storage and handling of the specimens prevented mixing and,

therefore, structure disturbance should not be considered a

threat to the reliability of the data.

Laboratory Analyses

Four soil analyses were performed on the samples: organic

carbon content; pH in water, and in potassium chloride and in

calcium chloride solutions; particle-size distribution (Soil

Survey Staff 1984); and aluminum content (Rhue and Kidder

1984). These data, along with data obtained from the morpho-

logical descriptions, provide a general basis for classifying

the soils and making some interpretations about their forma-



Based on the available data and recent mapping of the

county (Soil Survey Staff 1989), the soils associated with the

ditch belong to the order Spodosols. Spodosols are character-

ized as naturally infertile (Soil Survey Staff 1975:333).

This is due, at least in part, to their high acidity. Figure

7 provides evidence of the highly acid nature of the soils

Table 1. Morphological characteristics of soils.

Column #1
Depth Horizon Moist Colors Structure Roots
18 Apl 10YR 3/1 subangular blocky abundant
31 Ap2 moderate
46 Ap3 granular; few
61 Ap4 ; *
71 Ap5 ..
81 Ap6 7.5YR 3/0 single-grain none
89 Abl
102 Ab2
112 Ab3 10YR 3/1 some
122 Ebl 10YR 5/1
130 Eb2 10YR 4/1 none
147 Bhb 10YR 2/1 subangular blocky; **
155 Bwbl 10YR 3/2 single-grain
170 Bwb2 10YR 4/4 some
183 Bwb3 none
188 Bwb4

Column #2
18 Ap 10YR 3/1 subangular blocky abundant
31 El 10YR 4/1 some
46 E2 none
61 E3 10YR 3/4
74 E4 "single-grain
86 E5 10YR 4/4
102 Bhl 10YR 5/4 subangular blocky; **
119 Bh2 10YR 4/3 **
135 Bwl 10YR 6/3 ; **
152 Bw2 **
163 Cl 10YR 6/2 single-grain
178 C2 10YR 7/2
188 C3 10YR 7/1

Column #3
18 Ap 10YR 3/1 subangular blocky abundant
31 El 10YR 4/1 none
48 E2 10YR 5/1
61 Bhl 10YR 5/1 ; **
w/ 10YR 2/1
74 Bh2 10YR 2/1 "" **
86 Bh/E 10YR 2/1 ; **
w/ 10YR 7/2
104 B'hl 10YR 2/2 ; **
119 B'h2 10YR 3/2 ; **
137 Bh3 10YR 3/2 ; **
150 Bw 10YR 4/3 ; **
163 Cl 10YR 6/3 single-grain
175 C2
188 C3 10YR 7/2
fragmented cemented spodic horizon found in matrix.
** occurs in single-grain matrix.










0 1 2 3 4 5 6 7 8 9 10 11 12 13 14

As measured in
o- H20
o- KCI
a- CaCl2

0oo 0

0 0

0 0
o 0

o o




0 1 2 3 4 5 6 7 8 9 10 11 12 13 14

As measured in
o- H20
o- KCI
A- CaC12

2 3 4 5 6 7 8 9 10 11 12 13 14

As measured in
o- H20
o- KCI




Figure 7.

pH distribution of soils in water, potassium
chloride, and calcium chloride solutions.












9 0


0 a

1 \










0 0

I \


0 AO

0 0
o s,

I ,


tested. Nowhere in the soil columns is the pH value higher

than 4.9. Darrel Metcalfe and Donald Elkins (1980:340) state

that maize will not grow adequately below pH 5.5. Further-

more, Clyde Evans and E.J. Kamprath (1970) indicate that

acidic mineral soils, such as these, combined with the

presence of exchangeable aluminum, yield poor maize growth.

Figure 8 provides evidence of the very high concentrations of

aluminum in these soils. While not providing the amount of

exchangeable aluminum, this figure does suggest that aluminum

toxicity could be a problem for maize. As such, the likeli-

hood that the earthworks could support maize cultivation for

the long time span suggested by Sears (1982:191-201) is not

very high without considerable amendments.

Evans and Kamprath (1970:894) indicate that maize growth

can be improved on acidic mineral soils, such as these, by

liming the soil. This practice, along with fertilizing, would

have resulted in good maize growth. Sears believes the

residue in the ditches could have served as a source of

fertilizer. He states, "The muck that accumulates in the

ditches, particularly after a flood, is practically protoplas-

mic, wiggling with great quantities of fish and amphibian

small-fry and eggs. The nitrogen content at least should be

most useful in maize growing, and the tiny bones might provide

some phosphorus. Certainly the ditches had to be cleaned

occasionally if they were to function" (1982:189).


Aluminum (mg/kg)
0 100 200 300 400 500 600 700 800 900 1000
0 I I I Horizon
0 Apl
S -25-- Ap2
-50-- Ap4
r Ap4
o" mAp5
D -75-- Ap5
o Ap6
S-100- Ab2
-125-" Eb1
M Eb2
S-150-- Bhb
3 Bwbl
-175- Bwb2
-200- Bwb4

Aluminum (mg/kg)
0 100 200 300 400 500 600 700 800 900 1000
0 I Horizon

S -2 El
S-50-- E2
r -5 / E3
S-75-- E4
E E5
-100- Bh
o -125- Bh2
U *- 9wl
C -150- Bw2

-175 /.- C1
-200-- C3

Aluminum (mg/kg)
0 100 200 300 400 500 600 700 800 900 1000
0| I 1 Horizon

-25j/U Apl
- -50- E2
I Bhi
-75- h2
-100-- 'hl
I -125-
i In Bh3
- -150- R Bw

-175 C2
-200 C3

Figure 8. Al distribution in columns 1-3.

But, if the ditches were cleaned occasionally and if this

cleaning resulted in a preference towards dumping the debris

from the ditch into the interior of the earthwork (as would be

the case if the residue caught in the ditches was being used

as fertilizer), then a pattern in the particle-size distribu-

tion of the upper levels of the column located on the berm

(i.e. column #1) would reflect the particle-size distribution

of the lower levels of the column located inside the ditch

(i.e. column #2). That is to say, if the ditch were cleaned

periodically and excavated below the Bh (spodic horizon) as

suggested by Sears (1982:6), then the particle-size distribu-

tion found in the lower levels of column #2 should be the same

as that found in the upper levels of column #1. However,

instead of periodic cleaning, the particle-size distribution

of column #1 reflects a single episode of activity which built

the ditch. This can be viewed graphically in Figure 9.

Figure 9C, column #3, the column taken outside the ditch,

provides a model for what the others looked like prior to

excavation of the ditch. Its particle-size distribution shows

a trend towards an increase in the percentage of the larger

particles (i.e. vcs + cs) and a trend toward a decrease in the

smallest particles (i.e. fs + vfs and silt + clay) with

increasing depth. This pattern is even more pronounced in the

lower levels of column #2 (Figure 9B). Thus, if periodic

cleaning of the ditch resulted in excavation below the Bh,

then a substantial increase in the percentage of the larger











Particle-size (%)

) 10 20 30 40 50 60 70
I I :


/ I
** A

** A





*- cloy+silt
A VCs+cs
* ms
*- fs+vfs



Particle-size (%)

0 10 20 30 40 50 60 70



7 -50-





A *

I *
A 0

\ I
A *
A 0
A *
A \
\ \

A 0

*- clay+silt
*- ms
U- fs+vfs



Particle-size (%)

0 10 20 30 40 50 60 70

0 --

-25-/ \
I / /
-50o *

\ / /
-125 --

-150 \

SA A *


* clay+silt
-*- ms
a- fs+vfs








Figure 9.

Particle size distribution of columns 1-3
(where vcs = very coarse sand, cs = coarse
sand, ms = medium sand, fs = fine sand, vfs
= very fine sand).

I i : :

particles should be evident in the top layers of column #1

(Figure 9A). However, the particle-size distribution for

column #1 reveals an almost mirror image of particle-size

distribution between the Ab and the Ap horizons. That is to

say, the Abl and the Ap6 are almost identical. Likewise, the

Ab2 and the Ap5 are very similar. This "mirror image"

correlation continues with the Ab3 and Ap4, the El and Ap3,

the E2 and Ap2, and the Bh and Apl layers. Nowhere in this

profile are there substantial increases in the larger parti-

cles as would be expected if periodic cleaning of the ditches

had occurred.

Supporting evidence for this model is provided by the

graphic illustration of the distribution of organic carbon

content for each column. Figure 10 provides a comparison of

the distribution of organic carbon content among the three

columns and indicates the highest percentage is located

between 75 and 100 cm below the surface in column 1. Although

it is evident that column 3 also shows a high percentage of

organic carbon deep in the soil profile, note that it coin-

cides with the Bh horizon. This is because the Bh horizon is

defined as a horizon with an accumulation of organic matter

(Soil Survey Staff 1975). However, the high percentage of

organic carbon in column 1 coincides with the Ap6. (Note that

this horizon has the lowest concentration of aluminum content

within the Ap and Ab horizons of column 1 [Figure 8] and,

therefore removes the possibility that a Bh horizon was

Organic Carbon (%)
0.25 0.50 0.75 1.00



*---- -e

5 *

Organic Carbon (%)

0.25 0.50 0.75 1.00 1.25



Organic Carbon (%)

0.50 0.75 1.00




.25 1.50


Organic carbon distribution of columns 1-3.





















- Horizon

- Horizon



0.00 0.25








Figure 10.


mistakenly identified as an Ap horizon.) Thus, I argue that

the ditch is the result of an excavation that began with

removal of the A horizon and subsequent discard on top of the

old ground surface (effectively beginning the berm). This was

followed by excavation of the E and Bh horizons, which were

heaped on top of the A horizon. Subsequent pedogenetic

processes resulted in the accumulation of organic carbon,

creating an artificially thickened Ap horizon as seen in

column 1. Further proof of this scenario is provided by the

traces of the Bh (as seen in Table 10) in the Ap3 and Ap4

horizons. This evidence, in conjunction with the particle-

size distribution data, supports the model of a single episode

of activity rather than periodic cleaning of the ditch. An

important archaeological implication of this is the amount of

labor involved in construction of the ditch. While it is not

possible to determine the exact number of individuals nor the

amount of time it took to construct, the model suggests an

organized labor force over a relatively short period.

As for the liming of the fields, Sears (1982:171-174,178)

identified lime deposits on the site that are contemporaneous

with the use of the Great Circle. These deposits were found

associated with the Mound A midden deposits. Conceivably,

they could represent lime processing activity areas for

artificially raising the pH of the soils within the Great

Circle, but liming without fertilizing would still result in

poor maize growth. On the other hand, liming with fertilizing

should have left some evidence of the fertilizer--e.g., the

large accumulations of midden materials at the site. Since

these could easily have served as a source of fertilizer, one

could expect that some would have been found within the

alleged planting surfaces. However, Sears (1982:176-177)

indicates that repeated attempts to find materials within the

Great Circle were made by surface inspection and excavation

but nothing was ever found. Thus, the evidence indicates that

neither lime nor fertilizer was used within the Great Circle

to prepare the soils for maize cultivation.

Finally, Sears (1982:186) states that the ditches were

dug through an impermeable hardpan (i.e. the term that Sears

uses for the spodic horizon) in order to drain the planting

surface. However, this analysis of the soils reveals that

doing so results in additional aluminum in the upper horizons

of the soil (Figure 8) well within the rooting zone. As such,

Sears' explanation for the ditching would have a net result of

adding toxins to the surface prior to planting maize. Thus,

soil science data indicate the circular earthworks were not

planting surfaces for "a stable, dependable maize crop," but

must have served some other purposess.

Ethnohistoric Evidence: Conflicting Accounts

Perhaps the strongest evidence challenging Sears' (1982)

belief that maize served as the dietary mainstay comes from an

eyewitness account of life in sixteenth century South Florida.

Do. d'Escalante Fontaneda, a shipwreck survivor, lived among

South Florida's natives for 17 years from 1548 to 1565. His

memoir, focusing on the Calusa of southwest Florida, provides

some details on subsistence activities within the Okeechobee

Basin. He refers to a bread made from roots as the common

food most of the time except when the lake rises in some

seasons so high that the roots cannot be reached and they are

"for some time without eating this bread" (Fontaneda 1945:13).

He does not mention maize at all. Sears (1982:201) acknowl-

edges this problem, but questions whether Fontaneda would have

recognized maize if he had seen it.

Of course, there were probably many things Fontaneda did

not recognize and could not describe, but one must wonder why

it is that both the Narvaez and de Soto entradas described

maize in Florida years before Fontaneda lived among the

Calusa. Milanich (1987, 1989) believes this is because both

of these Spanish explorations landed in present-day Tampa Bay,

well north of the South Florida aboriginal groups that

Fontaneda described. On the other hand, Dobyns (1983)

believes that Narvaez and de Soto landed in Charlotte Harbor

and described maize use among the Calusa. In either case,

Fontaneda surely would have known about maize by the time he

had written his memoir. If maize had been as important in the

Calusa diet as Dobyns (1983) contends, how could Fontaneda

have omitted it?


Such an argument might go on indefinitely except for the

fact that of all the archaeological research conducted in

South Florida only the Fort Center site has produced any

evidence for maize (save that which has been found in Seminole

Period contexts). On the southwest coast, the few systematic

attempts to recover plant remains have produced no evidence of

maize (Marquardt et al. 1985). Similarly, Margaret Scarry

(1982) found none on the southeast coast. Indeed, John

Griffin (1988:298) remarks that there is no evidence for maize

cultivation in the Caloosahatchee or Everglades archaeological

areas. Based on the maize pollen at Fort Center, Milanich

(1987:177) accepts that it was present at the site sometime

during the period ca. A.D. 100-500. But, whether it was

present before or after that he feels is speculative.

Milanich and Ruhl (1986:2) point out that the importance of

maize within the Belle Glade people's diet is uncertain.

Johnson and Collins (1989) concur with Milanich (1987)

that the maize pollen found at Fort Center indicates maize was

grown at the site but believe that the circular earthworks

were not dug to create planting surfaces for a stable,

dependable maize crop. Following Milanich and Ruhl (1986:2),

they believe the maize grown at Fort Center must have served

some other purposes) other than a dietary staple.

In addition, Carr (1985:299) notes that all of the

circular earthworks are located near elevated hammocks or

other upland environments that could have naturally offered

the same drainage characteristics that Sears (1982) believes

the ditches offered. Indeed, the only advantage that the

ditches offered was the accumulation of organic sediments that

could serve as a source of nutrients for Sears' hypothesized

planting surfaces (Carr 1985:299) and that advantage apparent-

ly was not realized.

On the other hand, Donald Lathrap's (1987) study of the

introduction of maize into precolumbian eastern North America

questions current wisdom and warns us "not to be overly

skeptical of the presence of maize in situations where its

preservation is a matter of chance, such as in 'Green Corn'

utilization." Such skepticism may well be pertinent since

some ethnohistoric evidence for "Green Corn" use exists. For

instance, Oviedo (1959:2) is apparently speaking of green corn

use among the Taino of the Greater Antilles when he says "when

the ears are tender they are eaten almost like milk."

Furthermore, Lathrap (1987) outlines a plausible early

introduction of maize which conforms to Sears' (1982) hypothe-

sis of its introduction into Florida and takes exception to

Keegan's (1987) view that maize could not have passed through

the Antilles prior to the arrival of the Taino. Lathrap

(1987:347) notes that navigation in the Circum-Caribbean is

well documented as early as 3100 B.C. and that the early maize

in the Southeast discussed by Sears (1982) had a totally

different origin from the maize that was utilized by Missis-


sippian societies. Furthermore, he believes that a

two-pronged entry of maize agriculture into the eastern United

States may well explain the heterosis and success of Midwest

Hybrid Maize (1987:347).

Mainstream opinion on this topic, however, has been at

odds with Lathrap. Walton Galinat (1985) has provided a

recent summary of the domestication and diffusion of maize

into North America (north of Mexico) and while he readily

admits that isolation of races of maize and subsequent

heterosis from reconvergence of divergent populations accounts

for the Corn Belt Dent variety (what Lathrap is calling the

Midwest Hybrid Maize), the origins of the heterosis are to be

found in the introduction by the Spanish explorers of white

dent corn from Mexico blending with the Northeastern Flint

variety (originating as Maize de Ocho in the Southwest around

A.D. 200 700).


The soil data presented here strongly support a view that

is contradictory to Sears' model. Periodic cleaning of the

ditches, fertilizing the planting surface, and the effects of

digging below the hardpan are the main elements of Sears'

model that have been addressed. Soil science applications to

each element indicate annual harvests of maize were not

possible from the planting surface identified by Sears (1982).

These, however, do not refute the presence of maize pollen at

the site nor do they suggest that maize was never grown at the

site. Rather, the soil science data modify Sears' model in

two fundamental ways. First, they indicate the role of maize

in the Belle Glade peoples diet was probably more akin to

Milanich and Ruhl's (1986) view that maize was not a dietary

staple and second, they suggest that Sears' purely functional

interpretation of the earthworks is incorrect.

In the case of the latter, Goggin and Sturtevant's

(1964:194-197) view that the earthworks were ceremonial is

certainly the most reasonable interpretation at this time. A

cursory inspection of the literature on drained and raised

fields in South America (Denevan 1970; Denevan et al. 1987;

Farrington 1985; Smith et al. 1968) reveals morphological

similarity throughout the areas studied. In no case do these

agricultural fields resemble the Belle Glade earthworks.

Indeed, the only earthworks that are morphologically similar

are those in the midwestern United States. Current research

on them supports a ceremonial interpretation (e.g., Brose and

Greber 1979).

As to the former, Sears' model lacks corroborative

evidence to support his arguments. If Lathrap (1987) is

correct in his assertion that a preservation bias exists

because of "Green Corn" utilization, then we must look for

corroborative evidence in the pollen record. If possible,

phytolith analysis should be undertaken as well. As long as

the Fort Center maize pollen stands alone as evidence, the

ability to interpret the role of maize in the Belle Glade

culture is compromised.


The methods used for obtaining data on the Belle Glade

culture area involved a background and literature review, a

review of aerial photographs, a pedestrian survey for ground-

truthing sites found on the aerial photographs and informant

interviews. The background and literature review amassed data

from all over the Basin, but the review of aerial photographs,

pedestrian survey, and informant interviews were limited to an

area on the west side of the lake as part of the West Okeecho-

bee Basin Project (Figure 11). These methods, as well as the

methods used for obtaining data from the field and for

laboratory analysis of the collected data, are explicated


Background and Literature Review

The background and literature review provided a frame of

reference for evaluating archaeological sites. Pertinent

references on the precolumbian archaeology of the area were

consulted. Furthermore, a review of the Florida Master Site

File provided information on all known archaeological sites in

the project area.


I/ /4

/ _

0 28

Figure 11. West Okeechobee Basin Project Area.

Unfortunately, many of the known sites in the project

area were recorded on the basis of scanty information and

recorded locations, at times, are vague. Thus, a conscious

effort was made to obtain exacting information on recorded

sites that were lacking locational data.

Early maps of the project area were checked for informa-

tion on aboriginal site locations, early homesteads, and

military roads. The original survey maps (Florida State

Archives 1837) were considered the best early maps for this

information. One aboriginal site, the Old Venus Mound (8Hgl),

two historic homesteads, and numerous military roads were

discovered by this method. However, only the Old Venus Mound

was surveyed while in the field and, as such, the homesteads

and military roads were not given site numbers and do not

appear in the results section.

Aerial Photographs

The review of aerial photographs proved most enlightening

as not only were 10 previously unrecorded sites found but

significant additional information on the architecture of six

previously recorded sites was obtained. As stated in the

introduction, early aerial photographs are available in

stereopairs at the Map Library on the University of Florida

campus. Reviewing these involved locating the project area on

an index, retrieving the pertinent photographs, and viewing


Earthwork sites were easily discerned and plotted on USGS

quadrangle maps. Stereoscopes were employed at this stage to

provide detail on the sites. Whenever a possible mound was

found the stereoscopes were used to aid in deciding whether or

not it should be plotted.

Once a site was identified and plotted, later aerial

photographs of the site were viewed for additional detail.

These provided remarkably more information than could any one

photograph because portions of a site not visible in one were

often visible in another. Many factors are responsible for

this variation and include, but are not limited to, the time

of day the photograph was taken, season the photograph was

taken, angle of the camera, and moisture conditions at the

site. Thus, the best image of any one site is a composite

from many photographs.

Overall, this remote sensing application is most useful

in relatively flat, savannah areas of the Basin. Forest

cover, regardless of habitat, obscures manifestations of

cultural activity. Thus, in areas of forest cover, this

method is useful only in cases of known site locations.

Pedestrian Survey and Informant Interviews

The fieldwork was conducted via a pedestrian survey

coupled with informant interviews. The pedestrian survey

included the employment of judgmentally placed shovel tests.

Informant interviews included contacting landowners and other


knowledgeable persons during the fieldwork portion of the


Subsurface tests included shovel tests and bucket auger

tests. The shovel tests were approximately 50 cm in diameter

and were dug in 20 cm levels. The matrix from each was

screened through 1/4-inch hardware mesh and, for some levels,

through 1/16-inch window screen. (The addition of the

1/16-inch fraction was requested by the zooarchaeology

department personnel at the Florida Museum of Natural History,

the repository for the faunal remains.) However, artifact

analysis is based solely on the 1/4-inch fraction.

Soil from each shovel test was described in the field

notebook. The color of contrasting zones and their thickness-

es were noted. All were dug to culturally sterile soil unless

groundwater seepage or soil conditions prevented it. The

tests were placed judgmentally unless more than one was dug at

a site. In those cases, additional shovel tests were placed

on a grid pattern with spacing of the subsurface tests

depending on the local conditions.

In the event that cultural materials were recovered in a

shovel test no attempt was made to delineate the subsurface

boundaries of the site. Rather, the tests served as a sample

of the materials from the site for comparative purposes as

well as document information on the site's integrity and

possible function.

The bucket auger tests, on the other hand, were generally

used to obtain soil samples from sites. However, there were

cases in which the bucket auger was used to test whether or

not cultural materials were present at a particular locale.

The bucket auger is an 8 cm in diameter coring device that

allows quick removal of a soil core. The soil was not

screened when using this method. Rather, the soil is poured

from the coring device onto a tarp and visually examined for

the presence or absence of cultural materials. If any

cultural materials were found, they were noted and a site

number was assigned. On the other hand, if no cultural

materials were found, the location was not considered a site.

Thus, the accuracy of this method is limited by the variabili-

ty in cultural material content at a site. A site that is

characterized by a relatively high density of cultural

materials is easily discernible with the use of the bucket

auger. However, sites that are characterized by a low density

of cultural materials (i.e., no cultural materials found in

the bucket auger sample) were not identified as sites.

Therefore, caution must be observed when determining whether

or not a locale is a site with this method.

When the auger tests were used in this manner they were

placed judgmentally on the site. Once cultural materials were

encountered (usually within the first 5 cm), the presence of

materials was noted, the site named, and the matrix poured

back into the auger test hole. If no cultural materials were

encountered, the test was dug below the dark colored soils to

what appeared to be culturally sterile soils and then back-

filled. Generally, only one such test was excavated at each


Laboratory Analysis

Laboratory analysis of the materials collected from this

survey involved separating the specimens from each provenience

into four gross categories. These categories are lithics,

ceramics, animal bone, and soil samples.

The lithics were measured across their longest axis and

grouped into the following categories: very small (<1 cm),

small (1-2 cm), medium (2-4 cm), large (4-6 cm), very large

(6-8 cm), and extra large (8-10 cm). They were then counted

and, based on the presence or absence of cortex, were classi-

fied into primary flakes (those flakes with cortex completely

covering the dorsal surface), secondary flakes (those flakes

with little cortex on the dorsal surface), and tertiary flakes

(those flakes with no cortex on the dorsal surface).

The ceramics were compared with typology collections at

the Florida Museum of Natural History and subjected to

microscopic analysis (when warranted) in order to determine

their typological classification. Each sherd was measured

across its longest axis and grouped into the following

categories: very small (<1 cm), small (1-2 cm), medium (2-4

cm), large (4-6 cm), very large (6-8 cm), and extra large


(8-10 cm). They were then counted and are reported in the

results section as total number of sherds/number of rim sherds

in that total.

The animal bones were examined for obvious diagnostic

elements (e.g., Lepisosteus spp. vertebra and scales, Odocoil-

eus virginianus astragali, etc.) and species observed in the

assemblages are noted. However, no attempt was made to

provide a full analysis of the materials. Instead, they were

weighed by provenience (note that some weights are skewed by

the presence of impurities, usually concretions) for compara-

tive information on relative density of deposits.

The soil samples were subjected to either morphological

analysis or pH analysis. Morphological analysis was reserved

for auger samples from two of the earthworks. It is used to

describe the soil horizons, thereby providing insight on their

cultural manipulation. Analysis of pH, on the other hand, was

utilized at a number of the sites to provide a means of

checking for possible preservation bias. The readings are

based on an E.M. System Soil Tester.


A total of 38 previously unrecorded archaeological sites

were discovered in this survey. In addition, the review of

aerial photographs revealed significant new data on nine

previously recorded sites. These are described in this


A discussion of the previously unrecorded sites is

included in the section New Sites and the previously recorded

sites are listed under the heading Updated Sites. These are

followed by the section Negative Data, which provides informa-

tion on the field inspection of two localities that were

thought to have mound features based on inspection of aerial

photographs, but the pedestrian survey revealed they did not.

New Sites

Okeechobee County

80b27 Wedqworth Mound. The Wedgworth Mound is located in

Township 33S, Range 35E, Section 25 (Figure 12). The site was

located via informant interview and consists of a large sand

mound measuring 48 m north-south by 50 m east-west and is at

least 2 m high. It was reported to the author by the owner,

Mr. George Wedgworth, via Dr. William H. Marquardt of the



(b" + '- . 2j:C SA'; 3;E LOCATION
2 214 J. II
.4%.. ) I '*",M.C

S-, / 1

*! a 1. 214" -

A;u s0 .

\ -6 ..... -_ : .- -_ -,- -. .- - - -----
.3"-, .:0b2 .7. :.

..5 0 1 MILE -

Figure 12. Wedgworth (80b27) and Tennis Court (80b34)
.5 0 1- KILOMETE
Figur 2. edwrh(02)adTnis or 8b4

Florida Museum of Natural History. Inspection of aerial

photographs before and after fieldwork revealed no earthworks

in the vicinity.

One of two shovel tests yielded a few highly deteriorated

bone fragments and a single sherd. The sherd is the type

Dunn's Creek Red, suggesting St. Johns period cultural

affiliations. The bone fragments are not identifiable and

their condition precludes curation. Since it has been

documented that soil acidity can affect bone preservation (see

Gordon and Buikstra 1981), the soil pH was obtained. It

revealed that soil acidity is probably not responsible for the

poor condition of the bone as the pH registered neutral (at


The mound probably did not serve as a habitation site.

Its large size coupled with its lack of preserved faunal

remains suggests alternative functions.

It is thought that additional testing may reveal other

cultural affiliations however, based on the limited data

available, the site's contents and its presence near the Fort

Drum Creek drainage suggests a cultural boundary between the

Belle Glade and St. Johns regions.

80b34 Tennis Court Mound. The Tennis Court Mound is

located in Township 33S, Range 35E, Section 24 (Figure 12).

The site was located via informant interview and consists of

a small oval mound measuring approximately 15 m in diameter

and is at most 0.5 m high. It, like the Wedgworth Mound, was

reported to the author by Mr. George Wedgworth. Inspection of

aerial photographs was done after fieldwork and revealed no

earthworks in the vicinity.

One of two shovel tests produced one flake at a depth of

60 cm below the surface (cmbs). Furthermore, these tests

revealed what appeared to be a buried midden stain in the soil

profile between 45 and 90 cmbs. No animal bone was discovered

nor does the soil pH (6.9) suggest that soil acidity has

created a preservation bias. As such, it is difficult to

determine the functions) of this mound.

At this time, no specific cultural affinities can be

discerned but the presence of the flake and the buried midden

stain strongly suggest that the mound is precolumbian.

80b28 Fort Kissimmee Earthworks. The Fort Kissimmee

Earthworks are located in Township 34S, Range 31E, Section 11

(Figure 13). The site was located via remote sensing (aerial

CYW-4H-178) and consisted of a large rectangular shaped ditch

with rounded corners measuring approximately 500 m north-south

by 265 m east-west. The northwest corner of the rectangle

opened to a bend in the Kissimmee River such that water could

flow from the River into the ditch. Two embankments measuring

approximately 110 m each were located within the interior of

the rectangular earthwork and may have terminated at mounds.

Unfortunately, this site was not visited during the field

survey. Access is restricted by the Kissimmee River on one

side and the Kissimmee Canal on the other. Aerial photographs

I -


_.._ -... . .. . ---.- /.

\N ._/'- ._ : ., -1; -- ,, / *"^,* ..--

*-- ,:': N' .


80 b2
I -
,/i_-. -- .j.,. ,

^wj(r ~^ H "-_' ,. -

/ / V-u~v/ -) *-
it 49 X I t

1 .5 0 1 MILE

N-- ;

Fort Kissimmee Earthworks (80b28).

Figure 13.

reveal that construction of the Kissimmee Canal resulted in

destruction of about 2/3 of the site leaving only the north-

west corner of the earthwork intact (aerial 12055-174-126).

(Note that figure 13 suggests the northwest corner of the site

was destroyed by the canal; this is not the case. Instead, it

is the only portion of the site that is intact. The problem

with the figure is due to an error in placement of the

Kissimmee Canal by the U.S.G.S.) The embankments and most of

the ditch were dug through during excavation of the canal (see

Figure 14).

The limited data on this site do not allow for definitive

statements of its origin, much less its function. However,

its overall morphology, especially in light of the data on the

next two sites, suggests it originates with Belle Glade

cultural activities.

80b29 Fulford Earthworks. The Fulford Earthworks are

located in Township 34S, Range 32E, Section 31 (Figure 15).

The site was located via remote sensing (aerial CYW-4H-117)

and consists of a large rectangular shaped ditch with rounded

corners measuring approximately 230 m north-south by 170 m

east-west. In addition, a linear ditch extends from the

southwest corner of the earthwork approximately 200 m west

connecting the earthwork to the Kissimmee River. Similar to

the Fort Kissimmee Earthworks, an embankment measuring

approximately 100 m is located within the interior of the

earthwork and may terminate at a mound.

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

Acceptable Use, Copyright, and Disclaimer Statement
Last updated October 10, 2010 - Version 2.9.7 - mvs