Group Title: archeological, ethnohistoric, and biochemical investigation of the Guale aborigines of the Georgia coastal strand
Title: An archeological, ethnohistoric, and biochemical investigation of the Guale aborigines of the Georgia coastal strand
Full Citation
Permanent Link:
 Material Information
Title: An archeological, ethnohistoric, and biochemical investigation of the Guale aborigines of the Georgia coastal strand
Alternate Title: Guale aborigines of the Georgia coastal strand
Physical Description: x, 282 leaves : ill. ; 28 cm.
Language: English
Creator: Wallace, Ronald L
Copyright Date: 1975
Subjects / Keywords: Indians of North America -- Antiquities -- Georgia   ( lcsh )
Antiquities -- Saint Simons Island (Ga.)   ( lcsh )
Genre: bibliography   ( marcgt )
theses   ( marcgt )
non-fiction   ( marcgt )
Statement of Responsibility: by Ronald Lynn Wallace.
Thesis: Thesis--University of Florida.
Bibliography: Includes bibliographical references (leaves 275-281).
General Note: Typescript.
General Note: Vita.
 Record Information
Bibliographic ID: UF00098679
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: alephbibnum - 000177853
oclc - 03101687
notis - AAU4354

Full Text









In so brief a compass, it is scarcely possible to

offer adequate thanks to the individuals who devoted thou-

sands of hours of effort to the research described in this

dissertation. Any merit that the present investigation may

possess is largely due to them, while all shortcomings are,

of course, my own. Initially, it would seem in order to

pay due respect to the field crews of'the spring and summer

academic quarters of 1973 and the summer quarter of 1974.

Whatever it becomes in a theoretical dimension, archeology

will always mean digging, and digging will usually mean

students. The efforts of the over sixty undergraduate

students involved in the above excavations (under conditions

that ranged from hundred-degree heat to springtime floods)

ultimately provided the fundamental data upon which all the

subsequent laboratory and statistical investigations were


In this connection, I would like to acknowledge the

technical advice provided me in the field by Dr. Jerald T.

Milanich. His judicious combination of guidance and flex-

ibility allowed me to have the benefit of his experience

while pursuing my own research designs. For similar rea-

sons, I owe a debt of gratitude to Professor Charles H.

Fairbanks. It was his critical reading of my original

research proposal, as well as his subsequent comments in

the course of the research that aided me in clarifying both

the hypotheses of the field investigation as well as their

experimental implications. As a final note on the field

research, it was the Sea Island corporation of Sea Island,

Georgia which provided food and lodging throughout the

course of the excavation. Mr. William Jones, Coorporation

president, manifested a genuine interest in the research,

and arranged a variety of public presentations of progress

reports on our findings.

Additionally, I am grateful to5 the many individ-

uals who offered me assistance with the portion of the

research conducted at the University of Florida. Dr. William

R. Maples, physical anthropologist of the Department of An-

thropology and Director of Social Sciences at the Florida

State Museum identified the human skeletal material to pro-

vide basic demographic data of sex and approximate age. Dr.

Antoinette 8. Brown, physical anthropologist, provided ad-

vice on strontum analysis and a number of my archeological

statistical designs. Additional assistance with these de-

signs was provided by the personnel of the Statistical and

Calculator Laboratory of the University of Florida, by Dr.

Richard L. Scheaffer of the Department of Statistics, and by

Professor Zoran R. Pop-Stojanovic of the Department of Math-

ematics. For the strontium analyses, I am grateful to


Jerry Bronikowski, student in the Department of Chemistry,

and to his supervisor Dr. James D. Winefordner, for his per-

mission to use their spectrochemical equipment. Dr. Seymour

S. Block, Department of Chemical Engineering, provided ad-

vice with regard to chemical flotation techniques, reviewed

the spectrochemical portion of my research proposal, and

provided cautionary advice concerning the possible effects

of soil chemical composition upon the strontium analysis

results. In a similar vein, How~ard T. Odum, Research Pro-

fessor of Environmental Sciences, called my attention to

the biogeochemical movement of stable strontium to the sea,

and its possible concentration in the 'bones and tissues of

marine organisms. Dr. Daniel B. Ward, Professor of Botany,

conducted the ethnobotanical identification of the seeds

encountered in the excavation, and Dr. Fred G. Thompson,

Associate Curator of Malachology, Florida State Museum,

assisted me with the malacological identifications in the

midden column sample. For the zooarcheological investiga-

tion of that sample, I am grateful to Ms. K. F. Johnson,

who performed her species identifications and dietary re-

constructions under the direction of Dr. Elizabeth S. Wing.

I was assisted in the ethnohistorical reconstructions by

the courtesy of Professor Eugene A. Hammond who allowed me

access to Stetson Collection materials in the P. K. Yonge

Library of Florida History as well as to limited editions

of the writings of Rene Laudonnibre. A number of questions

regarding these latter documentary materials were resolved

by conversations with Albert B. Smith, Associate Professor

of Romance Languages (who also provided humor when I needed

it). For technical assistance and optimum research condi-

tions, I am grateful to the staff of the Social Science

Department of the Florida State Museum. A special note of

thanks is due to Jerry Evans and Milinda Stafford, extremely

able technicians, who assisted me in the setting up of labo-

ratory facilities for the malacological, ceramic, and chemi-

cal flotation investigations. A feeling for what these in-

vestigations were all about has been clarified (I think)

by over forty drawings and maps, and for this I acknowledge

the excellent work of Ms. Dorothy Harry.

Finally, I would like to take special note of three

individuals whose contributions to me and to my research

have personally meant a great deal. These are: Janet

McPhail, Otto von Mering, and Jean Gearing.

Janet McPhail participated in the excavation of

the summer of 1974. From that time until the spring of the

following year, she assisted me (without pay) in the months

of work that were involved in the ceramic and malacological

classifications. She was an excellent field and laboratory

assistant and a friend during difficult times.

I have known Otto von Mering for four years. In a

time in which specialization is too often a euphemism for

tunnel vision, he has shown me that interdisciplinary

interests are both feasible and intellectually rewarding.

Moreover, I have long benefited from his personal advice,

his sensitivity, and his understanding. I consider him

a wise counsellor---and a personal friend.

My final acknowledgement is extended to Jean Gear-

ing. She assisted me in the 1974 summer excavation, took

part in both the ceramic and malacological classifications,

helped with the final stages of maps and illustrations,

proofread a considerable proportion of the manuscript, and

devoted much of her time to the problems of ceramic statis-

tical design as well as to the final calculations. In many

respects, the research is hers as much `as mine and she gave

me far more than she received. Of all the people here men-

tioned (who have contributed much) I feel that I owe to

her my most extensive personal gratitude and respect.

This study is dedicated to my father.



ACKNOWLEDGEMENTS . . . . . . . . .

ABSTRACT . . . . . . . . . . .

. ii












HYPOTHESES ................,...............


TAYLOR MOUND .. .. .. .. . ..

COUPER FIELD .. .. .. . .. . .

INDIAN FIELD .. . . . . .. ..



SEQUENCE .. . .. . .. . . ..











BIBLIOGRAPHY .. .. .. . . . .. .. ..





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



Ronald Lynn Wallace

August, 1975

Chairman: Jerald T. Milanich
Major Department: Anthropology

Three separate lines of evidence are brought to

bear upon the problem of reconstructing the social organi-

zation and technological adaptation of the late protohistoric

and historic aboriginal societies of the Georgia costal

strand. These lines of evidence include the data from

archeological excavations conducted during the spring and

summer of 1973 and the summer of 1974, the results of ethno-

historical researching of relevant Jesuit and Franciscan

documents, and the results of strontium-88 analysis of

human and zooarchelogical skeletal materials uncovered in

the course of excavation.

The archeological investigations include the exca-

vation of: 1) a ceremonial mound with associated burials;

2) a pavilion-like structure with burials along its southern

walls and a charnel structure, both located within a village


area; and 3) the village middens with associated burials.

The findings of these investigations are compared with

earlier archeological investigations in the coastal strand


Ethnohistorical research sources include the nar-

ratives of French Huguenot Rene Laudonniibre and Spanish

archivist Peter Martyr D'Anghera, as well as the writings

of Jesuit and Franciscan priests. Sociocultural informa-

tion from these documents is compared with relevant exca-

vation data and the results of the biochemical analytical


The biochemical investigation involves the compari-

son of human skeletal remains with associated zooarcheolo-

gical materials to determine absolute and relative amounts

of stable strontium present in human bone. The use of

zooarcheological materials for comparison is designed to

facilitate both the testing of strontium-analysis in a

coastal area, as well as to provide information regarding

the amount of animal and plant procurement by the proto-

historic society of the region. This information is then

related to subsistence data elicited from the ethnohistorical

sources, zooarcheological analysis, and water-screened soil

samples subjected to zinc chloride chemical flotation tech-


These three lines of evidence are then integrated

to provide an outline of the technological adaptation,

the social organization, and the ideological system of

the protohistoric Georgia coastal cultures.




The major objective of the present research was the

utilization of multiple investigatory techniques, particu-

7arly those of archeology, zooarcheology, ethnohistory,

and biochemistry, in order to provide a more detailed un-

derstanding of the social organization and environmental

adaptation of the protohistoric Guale Indians of the Geor-

gia coastal strand. In the chapters that follow, the

nature and findings of those investigations as well as

the specific locations and the regional ecology where the

excavations were conducted will be discussed. In the pres-

ent chapter, we will first present a brief summary of

earlier archeological research conducted on the Georgia

coastal strand and then summarize the research hypotheses

that guided the present investigation.

Previous Georgia Coastal Research on
Late Prehistoric Aborginal Sites

While it would be very incorrect to state that the

Georgia coast has been archeologically ignored, it never-

theless seems justified to remark that, excepting the

Figure 1. St. Simon's Island

Irene site near Savannah, few extensive investigations of

the area have ever been conducted and adequately reported.

Research began at a comparatively early date. In the

latter part of the nineteenth century, Clarence B. Moore

travelled the length of the Georgia coast searching out

low sand burial mounds for excavation. By present stan-

dards, Moore's work was unsystematic and much information

was lost by the "demolishing" (Moore's word) of the mounds.

When seen, however, in the context of his times, he be-

comes less deserving of strong professional criticism.

Systematized techniques of archeological investigation

were still three decades away, and American cultural an-

thropology, which might have provided a body of compara-

tive data for the better understanding of C. B. Moore's

results, was itself still a miniscule organization strongly

beset with racist interpretations and fanciful cultural

evolutionary schemes (Harris 1968:251, 292, 297-8). Be-

yond this, it is noteworthy that Moore provided highly

detailed drawings of aboriginal burials and their associ-

ated grave goods, published the results of his investiga-

tions, and routinely delivered burials, grave goods and

occupational debris to the trained specialists of the

northern museums (Martinez 1975:9-10). Moore's Georgia

findings were eventually published in 1897 by the Phila-

del hia Academy of Natural Sciences under the title Cer-

tain Aboriginal Mounds of the Georgia Coast.

There was no archeological research of any sig-

nificance conducted on the Georgia coast from C. B. Moore's

time until the Works Project Administration's excavations

of the 1930's. At that time archeological excavations

were characterized by a more systematized methodology (Fay

Cooper-Cole had introduced the University of Chicago grid

system and the standard-block method of excavations in the

late 1920's), with the consequence that the reports of

these excavations, when available, are considerably more

useful to present-day researchers.

The work of the 1930's presents an unfortunate

irony, however, to the present research; i.e., the best

reported excavations are those farthest from St. Simon's

Island, while the excavations conducted on St. Simon's Is-

land, the region of this report, were scarcelyreported at

all. These latter excavations were conducted by Preston

Holder in 1936 on St. Simon's Island. They had the ob-

jective of determining by rapid site surveys and test-pit

excavations, a general outline of the cultural (read ce-

ramic) sequence of the island. Holder's results were

written down only in his letters to his supervisor, A. R.

Kelly (these have recently been made available for study)

and in a brief published account several pages in length.

The contrastingly remote (but well-reported) excavation

was the Irene Mound site of Chatham County, Georgia, re-

ported by Joseph R. Caldwell and Catherine McCann, with

physical anthropological analyses conducted by Frederick

Hulse (1941). Both Preston Holder's and Caldwell and

McCann's excavations will be discussed in some detail,

since their results were instrumental in generating the

research hypotheses of the present investigation.

Preston Holder's barrier island excavations in-

cluded the Airport Site and the Charlie Kind Mound at thle

southern part of St. Simon's Island, Gascoigne Bluff on

the western part of the island, five stratigraphic test

pits on Cannon's point, where the present research was

conducted, and the Sea Island burial mound on the northern

end of Sea Island. Of these, the Airport Site is of par-

ticular interest as being possibly close in time to the

sites of our own excavation. A summary of Holder's work

at that site, prepared for the Professional Projects

Supervisor of the Works Project Administration in Savannah,

Georgia for presentation at the Proceedings of the Society

for Georgia Archeology, provides the following data:

Work at the Airport Site revealed that the
characteristic burial types among the two
hundred individuals recovered were the group
secondary (that is; the burial of disarticulated
fleshless bones) and the prone, full-extended
primary (burial immediately after death). The
21,000 sherds recovered established a grit-
tempered, complicated-stamp ware as the typi-
cal decorated pottery. Noteworthy, also, was
the high percentage of bone awls and gravers
from this site (50 from a total of 200 arti-
facts), perhaps indicating an extensive use of
skin-clothing by these people. Unfortunately,
it was impossible to determine a house-type
from the more than 3,000 [sic] post-molds re-
corded. (Holder, 1938:8)

It is of interest, in the above account, to note

that two types of burial are reported. That one of these

is secondary suggests possible removal from an ossuary, a

possibility that, unfortunately, is best corroborated by
the evidence of its architectural features. We will re-

turn to the matter of ceremonial mound burial and ossuary

burial (with subsequent removal) in chapters that follow,

and will take note as well of the possible sociocultural

significance of these practices as diagnostic markers of

differential group prestige.

As a final note on Holder, we will point out that

the similarity of Airport Site ceramics to those of the

Irene Site, as well as the probable (ceramic-derived)

protchistoric date for both sites, suggest that Cald-

well, McCann, and Hulse's excavation should be considered,

along with the work of Holder, as having a possible bear-

ing upon the present research.

The Irene site was excavated as a W.P.A. project

in the late 1930's, and the final report, The Irene Mound

Site: Chatham County Georgia, appeared in 1941. The site

was located on the west bluff on the Savannah River, imme-

diately south of its juncture with Pipemaker's Creek, some

five miles above the city of Savannah. The excavation

determined that the site was probably occupied during two

different archeological phases (defined primarily in terms

of ceramics), the Savannah (1200-1400 A.D.) and the Irene

(1400 A.D.-1600 A.D.). It functioned as a burial and

ceremonial center, and contained a total of 265 burials.

The majority of these were distributed as follows:

1) a small mound (2-1/2 feet high, 55 feet in

diameter) spanning both Savannah and Irene

times. 106 burials.

2) the eighth (and final) stratum of a large

mound. 6 burials.

3) a palisaded mortuary, associated exclusively

with Irene ceramics.

Additionally present was a rotunda structure composed of

six concentric circles of wall trenches and postmolds. No

midden accumulation was found in this structure, and many

of the ceramics (entirely Irene) were concentrated in a

pottery cache outside the rotunda. The similarity of the

structure to the Creek council houses described by natural-

ist William Bartram, the lack of midden, and the presence

of a pottery cache led Caldwell and McCann to conclude

that the rotunda likely functioned as a ceremonial struc-

ture, conceivably involving the ritual of cassina drinking

and subsequent vessel destruction (Caldwell and McCann


Of further interest are the investigations of

physical anthropologist Federick Hulse. Sixty-two of the

265 burials were too fragmentary for analysis, but the re-

maining 203, including 74 adult males, 75 adult females,

16 adolescents, and 38 children or infants, were subjected

to morphological and metrical analyses. Hulse arrived at

conclusions regarding sexual differences and similarities

as well as aboriginal dietary patterns that are both of

interest to the present research. He noted that

the teeth of the adult inhabitants of the Irene
site are badly worn down. ... There is no
case of a really mature skeleton which does
not show tooth wear ranging from excessive to
extreme, at least for the incisors and canines
and usually for all the molars as well.. .
In contrast to this enormous amount of tooth
wear, caries is rather unusual, and when it
occurs it usually seems to be mild rather than
extreme. .. The premolars and the first
molars are the teeth which were more frequently
lost. (Hulse in Caldwell and McCann 1941:60)

Hulse attributed the combination of infrequent caries and

excessive wear to a diet that was low in carbohydrate and

high in grit:

[It is] likely that these people lived on a
diet composed largely of shellfish such as
oysters, and other sea food. (Caldwell and
McCann 1941:60)

Of shellfish as a significant staple, more later.

At the moment, it is of interest to note Hulse's further

conclusions regarding similarities and differences of the

Irene skeletal material. He observes that

with respect to head length .. the females
from the burial mound [n=12] are less variable,
and therefore more homogeneous, than the fe-
males from the mortuary structure [n=17]. The
females from the burial mound are also less
variable in head length and cranial index than
the males from the burial mound [n=20]. .
on the whole, the females at Irene appear to
be less variable that the males. (Caldwell
and McCann 1941:67)

An interesting sociocultural inference related to these

data is drawn by Caldwell and McCann:

Since as many burials were found in and about
the mortuary structure as in the burial mound,
which served for a longer time, it is logical
to assume a lesser population during each gen-
eration of the earlier period. It might, then,
have been necessary for the local girls to
marry men from some other place. [Calculating
descent through females] is a very common cus-
tom of American Indian tribes, frequently asso-
ciated with matrilocal residence. It is not
at all unlikely that the Indians at Irene fol-
lowed the sai: pattern. (Caldwell and McCann

Finally, it is of interest to note that the Irene

site excavators chose to report their zooarcheological

findings. Admittedly, this is done in tabular form only,

but nonetheless represented a developing interest in

eliciting information regarding the prehistoric utiliza-

tion of the environment. Among the faunal remains were

white-tailed deer, black bear, opossum, raccoon, beaver,

bobcat, turkey, alligator, sturgeon, gar, drumfish, blue

crab, Virginia oyster, and others. The species list pro-

vided at least the evidence that the Irene population sub-

sisted by extensive hunting, fishing and shellfishing.

While no direct evidence of agriculture was found, it must

be borne in mind that ethnobotanical recovery techniques

of water-separation and chemical flotation were not in

use in American archeology at the time of the Irene site


It is fortunate that this preliminary concern with

prehistoric environmental use was continued to a far more

detailed extent in the work of subsequent archeologists,

particularly that of Lewis H. Larson. In his doctoral

dissertation, Aboriginal Subsistence Technology on the

Southeastern Coastal Plain during the Late Prehistoric

Period, Larson provided not only a detailed environmental

discussion of the resource availability of the pine-barren

inland regions, but presented as well the hypothesis that

the barrier islands were only seasonally occupied, in that

the late prehistoric tribes would migrate inland during

the fall and winter for acorn gathering and hunting sub-

sistence activities (Larson 1969). These and other mat-

ters of coastal protohistoric subsistence will be examined

in some detail in later chapters.

The above research, as well as the considerable

body of ethnohistorical material relating to the region,

were both insturmental in generating the research hypothe-

ses of the present investigation. These hypotheses, as

well as their respective experimental designs, will be

discussed in the section that follows.

Research Hypotheses

The present research was organized in relation to

a central aspect of the culture-concept paradigm, i.e.,

the concept of functional integration as a result of

sociocultural adaptation. Specifically, it was held,

a prior, that

Adaptation, as the concept is used in biology
and anthropology, refers to the ability of a
population both to tolerate selective forces
in its environment and to master environmental
problems by developing effective structures,
behaviors, and forms of social organization.
...Natural selection, then, applies as
much to man's cultural development as it does
to his biological development. Indeed, the
two are inseparable, for a new tool, technique,
or social innovation cannot persist if it
leads to the extinction of any population
which accepts it. On the other hand, an
innovation which siginficantly improves the
adaptation of a group to its environment will
give that group an advantage over other groups
which lack the innovation. (Bock 1969:208)

It is this selective process which ultimately pro-

duces, over a sufficient length of time, the functional

integration of any sociocultural system. As summarized bly

Harris, this integration consists of

arrangements of patterned behavior, thought,
and feeling that contribute to the survival
and reproduction of particular social groups.
Traits contributing to the maintenance of a
system may be said to have a positive function
with respect to that system. Viable systems
may be regarded as consisting largely of
positive-functioned traits, since the contrary
assumption would lead us to expect the system's
extinction. (Harris 1971:141, emphasis in

For heuristic purposes, this integrated system

may be understood, following the British anthropologist

A. R. Radcliffe-Brown, as possessing three major adaptive

aspects: the ecological patterns, the social structure,

and the ideological organization. Since it was the ulti-

mate purpose of the present research to reconstruct the

sociocultural system of a protohistoric coastal society,

the deductive assumptions and their associated research

hypotheses involved in this task will be presented below

under the particular aspect of the sociocultural system to

which they were most directly related. We will begin with

the ecological patterns.

Ecological Patterns

The ecological patterns involve

the culturally given technology of energy pro-
curement, transformation, and distribution.
These technological items interact with the
conditions of the natural habitat to yield
characteristic levels of energy outputs in the
form of food, fuels, and other disposable en-
ergy rations. (Harris 1971:144)

Research hypotheses regarding ecological patterns were the


1) Coastal adaptation involved the utilization of

high-hammock, tidal flat, and sea-beach-and-dune habitats.

The rationale here was that the scattered-resource environ-

ment of the coastal strand would necessitate utilization

of all of its respective habitats, in that any single one

of them would not be a sufficient resource supply for the

coastal societies, and moreover, would soon be depleted

if the other habitats were left untouched. Test impli-

cations of this hypothesis involved the research tech-

niques of zooarcheology, malacology, and spectrochemical


The zooarcheological investigation involved the

classification of faunal material found in arbitrary 25 cm

levels of a 3m x 6m column sample of shellfish midden

matrix. If the hypothesis was accurate, then representa-

tive fauna from all three habitats should be encountered.

The malacological investigation consisted of

classifying the shell content of the above column sam-

ple. Representative shells of both beach-and-dune and

tidal-flat habitats were anticipated.

Spectrochemical analysis involved the determina-

tion of amounts of stable strontium present in the matrix

of human skeletal material as evidence of the proportion

of animal and plant food in the aboriginal diet. Given

the hypothesis, it was anticipated that significant dif-

ferences in amounts of stable strontium would exist be-

tween the individuals in the archeological sample. The

rationale here is that the society subsisted by hunting,

fishing, and shellfishing, with possible gathering activi-

ties as well, and that differential access to the different

resource areas would produce significant Sr differences

between the individuals of the sample. This situation

would be comparable to the same type of differentiation

observed by Antoinette B. Brown for a formative popula-

tion in the Oaxaca Valley of northern Mexico (Brown n.d.:

1-4), in which high-status hunters manifested signifi-

cantly lower amounts of strontium in the matrix of the

skeletal material.

2) At the time of European contact, the coastal

tries ereundrgonga technological shift from a hunting,


fishing, horticultural, and shell fishing economy to an

economy more pronounced_]y horticultural. This hypothesis

was suggested both by the architectural massiveness of the

Irene site (implying an energy surplus, possibly agricul-

tural) as well as by the early ethnohistorical source

materials, to be discussed in a subsequent chapter.

There is good reason for exercising considerable care

in the formulation of this particular hypothesis. Above

all, the near synchronousn~ess of culture contact and the

condition of a developed horticultural system in the

sixteenth century should not imply an automatic causal

relation between the two. While it is true that Spanish

missionaries were interested in instituting a full-time

agricultural economy on the coast, it should not be over-

looked that the earliest European encounters with the

coastal strand tribes (ca. 1526; 1562) report the exis-

tence of an already-developed, indigenous agriculture.

The Spanish priests may simply have sought to transfer

more of the inland horticultural activities to the coast

by making coastal cultivation more efficient. Use of hoes

or other technological improvements could have been use-

ful to this end. The procedures involved in investigat-

ing this hypothesis are described below.

Zooarcheol ogl~~ic~_a~l analys was conducted as dis-

cussed for hypothesis (1), but in this instance, it was

anticipated that the overall vertebrate animal biomass

would decrease through time. The rationale was that an

intensification of horticultural activities would parallel

a decline in hunting and fishing, which would be reflected

in a reduced osseous weight of faunal materials in the

more recent levels.

Malacological analysis was conducted as discussed

in hypothesis (1), but in this instance, it was of inter-

est to determine whether or not the shell species directly

associated with a beach-and-dune environment underwent a

decrease throughout time. It was reasoned, in this case,

that an increase in horticultural activities would be

accompanied by a decrease in the sea fishing activities,

during which time the beach-and-dune shells would have

been incidentally procured.

Spectrochemical analysis was applied to the skele-

tal remains of a protohistoric and early historic site to

determine biochemically if a noticeable transition in

dietary patterns took place at the time of culture con-

tact. In this instance, it was anticipated that the more

recent skeletal material would have concentrations of

stable strontium closer to that of a known herbivore,

while earlier material would be greater than the terres-

trial primary consumer level because of the greater con-

centration of Sr present in malacological and marine

faunal materials.

Ethnobotanical analysis was the most crucial tech-

nique of all, however, in establishing the existence of

aboriginal horticulture. In this investigation, two

separate techniques were utilized. One of these was the

method of water separation and zinc chloride chemical

flotation. This was applied to two arbitrary levels of

the column sample of midden matrix discussed above. It

was anticipated that evidence of cultigens (i.e., charred

seeds, corn kernels) should appear in the arbitrary levels.

A supplementary analysis involved food pits present be-

neath the plow-zone in a sterile sand matrix and within

a protohistoric context. Here, all seeds were simply

hand-separated from the soil contents of the pits and

transported to the University of Florida Department of

Forestry herbarium for classification. Specifically an-

ticipated in this instance was direct evidence (supple-

menting the ethnohistorical data) of the cultivation of

Zea mays or "Indian corn."

Social Structure

The social structure promotes

the maintenance of orderly relationships
among the individuals and groups responsi-
ble for technoenvironmental processes and
for the breeding and care of children.
Social structure is also concerned with the
orderly transfer and distribution of energy
and labor power among the various produc-
tion units. (Harris 1971:145)

Research hypotheses involving the coastal aboriginal social

structure were the following:

1) The coastal population was characterized by

ranked groups. For reasons that are presented in a sub-

sequent chapter, it seemed wisest not to attempt any

specification as to the descent rule (if any) that de-

termined membership within the groups. It was reasoned,

however, that the scattered-resource ecology of the coas-

tal strand would be most viably utilized by a system of

competitive, or rivalrous redistribution, a system which,

in turn, could serve as a basis for determining differen-

tial group prestige within the protohistoric society.

There was, additionally, ethnohistoric documentation

suggesting that such a system was in effect. Archeologi-

cal examination of this question involved mortuary data

exclusively, and consisted of the following investiga-


a. The statistical determination that all

burial samples were taken from the same population. While

it was acknowledged that there is certainly a chronological

ordering for the three sites investigated, it was also

determined by ceramic and mortuary analyses that there

exists no significant temporal gap between any of the

three. In view of this, it was seen as legitimate to

investigate statistically the possibility of combining

the three sites for further analysis. This was accom-

plished by expressing grave goods content and appearance

of the burial as a set of presence-absence traits. Two

of the three sites were then compared in this way, the

number of similarities between burials being noted. Mean

similarity and standard deviation were calculated. This

procedure, plus ceramic similarity, served as the rationale

for combining two sites. The third site, heavily dis-

turbed, was added to the sample on the basis of ceramic

similarity and ethnohistorical documentation relating to

aboriginal living structures.

b. The architectural analysis of the sites.

With the combined samples, it was hypothesized that rank

differences between aboriginal groups would be manifested

in mortuary differences between them. In this investiga-

tion, three different locations for aboriginal burials

were encountered, and an explanation was presented demon-

strating the manner in which at least two ranked groups

were involved. To avoid circularity, the explanation was

supplemented by relevant ethnohistorical data.

2) The ranked groups were characterized by a

matrilocal postmarital residence pattern. This hypothesis

was suggested by the analysis of the Irene site (discussed

above), but could not, due to the limited size of the

burial sample, be investigated by the techniques of mor-

phological and metrical analysis. As a consequence, the

hypothesis was investigated through the statistical tech-

nique of Pielou-Peebles "nearest-neighbor" analysis. In

this investigation, it was anticipated that burial dyads

would principally include female-female and female-male

pairs with only a limited number of male-male dyads. The

rationale was that young, unmarried females would be

buried together, married females would be placed with

their affines, and few males would be found together,

since local males would soon depart to find a wife, and

"imported" males would be buried with their w~ives. (Given

partilocality, a mirror image mortuary pattern would be

anticipated.) As a supplement to this investigation, a

subsample of female-female dyads was examined by the same

methodology to determine whether or not there was a sig-

nificant clustering according to age. In this instance,

it was anticipated that the predominant age grouping would

be the young-young dyad, since mature/aged individuals

would be interred with affines, and would not be a sig-

nificant part of the female-female subsample. In both

the above investigations, the relevant variables were cast

in a 2x2 contingency table and tested through the use of

Fisher's Exact Probability Test.

Ideological Organization

The ideological organization was the final con-

cern of our attempts at sociocultural reconstruction. It

may be defined as

the entire realm of socially patterned
thought. It includes the explicit and
implicit knowledge, opinion, values, plans,


and goals that people have about their ecologi-
cal circumstances: their understanding of
nature, technology, production and reproduc-
tion; their reasons for living, working, and
reproducing. (Harris 1971:146)

In this instance we have been required to rely,

almost exclusively, upon ethnohistorical materials to re-

construct this dimension of the coastal tribes' socio-

cultural adaptation. We have been aided, however, by the

investigations into the possibility of ranked groups (dis-

cussed above) which facilitated the use of cross-cultural

correlations between the number of hierarchical groups in

a technologically primitive society and the existence of

a belief in a supreme deity. It was thus possible to

substantiate at least one aspect of the ethnohistorical

ideological data through the combined use of archeological

findings and cross-cultural statistical correlations.

The above research hypotheses provided direction

for field and laboratory investigation conducted from the

spring of 1973 through the fall of 1974. More extensive

details on experimental procedures will be provided in the

appropriate chapters. At the moment, however, we will be

concerned with describing the region in which the research

was conducted, the plant and animal resources located

within that region, and the relationships existing between

them. These ecological matters (wEhich ultimately have a

bearing on virtually everything that follows) will be

presented at some length in the following chapter.



St. Simon's Island, where the 1973 and 1974 exca-

vations were conducted, is a part of a chain of ecologi-

cally similar barrier islands extending from just south of

the Florida-Georgia border along the complete coastlines

of both Georgia and South Carolina. These islands are the

product of sea level changes, tidal action, and stream-

carried sedimentation, all interacting with one another

during the Pleistocene. One view of barrier-island forma-

tion posits that the estuarine situation of fresh-water

streams on the mainland flowing to meet the sea results

in much sedimentation occurring in the quiet water at the

estuary mouth. The sediment eventually develops into a

delta which, if it becomes sufficiently high to be exposed

at low tide, will become a lagoon or tidal flat region.

Simultaneously, however, the erosion of the coast through

tidal action is occurring, and materials are being deposited

on the exposed delta, if continued for a sufficiently long

period of time, barrier islands will develop which protect

both the mainland and the brackish flats on their landward

side from the turbulence of the ocean tides (Smith 1966:

210, Larson and W~aring 1965:263).

Figure 2. United States Geological Survey Map of
St. Simon's Island

*~ ** ( ' .

N ....

:..-T o NT;.

... o:

,..~.:. .. .i .*

.. *.

.. ..=~.

s a

Another view of the matter is held by marine geo-

logist John Hoyt, which seems to be a more economical ex-

planation. Essentially, Hoyt maintains that the tidal cur-

rents intersect the sloping landscape of the mainland,

causing dunes or beach ridges to form adjacent to the shore-

line. Eventually, the landward side of these ridges be-

comes submerged, and a barrier island is created (Hoyt 1967:


In any case, the resulting ecological situation is

one of a multiplicity of habitats in close contiguity.

This is sometimes referred to as an "ecotone" environment,

the ecotone being the borderline between two or more of

the habitats (Fairbanks 1968:37). There are at least

three major habitats within the barrier-island system ,

and many further subdivisions could be made of these three.

They are the beach-and-dune, live-oak hammock, and the

estuary and tidal flat habitats. A brief discussion of

the biotic communities within these habitats, as well as

the interaction between the various communities, is pre-

sented in detail below.

Beach-and-Dune Community

The sandy beach on the seaward side of the barrier

islands is the product of the weathering action of the

waves upon the island. It is a habitat that supports a

less complex biotic community than the others, and is of

considerably less value to human utilization. Neverthe-

less, it does provid-e some resources, and is of particular

interest ecologically in that it offers a striking example

of a delicate relationship between organic matter and

saprophytic bacteria (decomposers), an interaction which

provides not only a food source for littoral fauna, but

functions as well as a site of biogeochemical recycling

of essential nutrients (such as phosphates and nitrogen)

to the sea (Smith 1966:237).

The life that exists on the beach below the high-

water mark must necessarily exist belpw the sand, as the

surface provides neither protection nor features for at-

tachment. A few inches below the surface, however, there

exist conditions of relatively constant temperature and

salinity. It is here that organic matter brought in by

the tide, as well as anaerobic bacteria, which abound in

the subsurface low oxygen conditions, are brought into

juxtaposition. The result is a constant decay of some of

the organic matter into phosphorus and nitrogen (which

are returned to the sea), and the provision of a food

source for the littoral primary consumers (Smith 1966:235).

These primary consumers are generally burrowing

fauna. They include worm species such as lugworms and

trumpet worms, which tunnel through the sand to the site

of the organic detritus, amphipods (small, often parasitic,

crustacea) such as sand fleas and ghost shrimp, and "surf

fishers," which are beach fauna that follow the ebb and

flow of the tide. This last group includes the razor-like

clams, Ensis directs and Taselus plebius (the latter

species utilized in the aboriginal diet), herbivorous ani-

mals which employ a powerful anterior foot to bring them

to the organic "belt" below the sand, and the periwinkle

(Littorina sp.), a European immigrant to Atlantic shores,

which is a littoral snail generally found near the average

high-tide limit, and is also a possible food resource for

humans (Morris 1947:147, Milanich 1972a:3-6, Smith 1966:238).

Ths succeeding trophic level,'.that of the carni-

vores, includes not only such predatory gastropods as the

moon snail (found from Massachusetts to the Gulf of Mexico),

but also crustacea such as the blue crab Collinectes

sapidu (another human food resource) which, in both cases,

prey upon the mollusca (Richards 1938:173, Smith 1966:238).

Additionally, the gulls and the shore birds which probe

through the wet sand in search of food, would belong to

the carnivore level (Smith 1966:239).

The dune and back-dune areas (or subhabitats) are

the product of the interaction of tidal litter, wind-blown

seeds, and the organic nutrient supply of the tides and

tidal creeks. Unlike the flat, sandy beach, the higher

beach (or embryonic dune area) is characterized by debris

brought in by both the tides and tidal creeks, which makes

possible a surface attachment of wind-blown seeds. In the


embryonic stages of the dune, these colonizing seeds will

take root in the organic belt below the surface of the sand

and will extend horizontal roots very close to the sand's

surface, an activity made possible by the temperature re-

duction due to the shelter of the tidal debris (a signifi-

cant difference of some 870F in open sand versus 44.60F

beneath the region of the litter). These horizontal feed-

ers aid the debris in halting both wind-blown sand and seeds

(they are sometimes referred to as "tillers," in this con-

nection) thus initiating a positive feedback relationship

between the wind-blown materials, and the horizontally

extended "tiller" plants culminating in the development of

a dune (Ranwell 1972:135-40),

The passage of time accomplishes the improvement

in the dunes of the various conditions of plant growth.

Stability increases with the colonizing vegetation dis-

cussed above, subsurface nutrient supply is supplemented by

the regular decay of the vegetation, moisture is main-

tained by tidal creeks and barrier-island rainfall (which

is adequate all year round, with almost daily convectional

thunderstorms in summer), and temperature is held relatively

stable (mean annual temperature is around 60"F.), partly

as a result of the shade provided by the developing vegeta-

tion community (Gibson 1948:50-51, Ranwell 1972:138).

This community is our first real instance of a

primary production level sufficiently developed to provide

a food source for nonlittoral herbivorous animals. One

finds, in this subh~abitat, such plants as cabbage palm

(Sabel palmetto), saw palmetto (Serenoa repens), china

briar (Smilax bonanox), Spanish bayonet (Yucca gloriosa),

cassina (Ilex vomitoria), sea oats (Uniola paniculata),

wild grape (Vitis rotundifolia), and wax myrtle (Myrica

cerifera) (Milanich 1972:5). Vegetation such as this pro-

vides food for herbivores such as the marsh rabbit (Sylvila-

gus palustris), for browsers such as white-tailed deer

(0docoileus virginianus), and for omnivores such as the

raccoon (Procyon lotor) and the opossum (Didelphis marsup-

ialis) which can feed not only on the plants, but on crus-

tacea, worms, and insects as well. All of these animals

can additionally serve as human food resources, and the

last three species were in fact encountered in food pits (to

be subsequently discussed) during the 1973-74 excavations.

The beach-and-dune system, in short, is a habitat

which (with the notable exception of surf-fishing activi-

ties) is of comparatively little nutritional value for a

human society. The high solubility of the soil in combi-

nation with wave action, high salinity level and low nitro-

gen content act in concert to prohibit any lush vegetation

which could provide a primary production base for a signifi-

cant amount of animal biomass. As a consequence, the beach-

and-dune system was of comparatively less importance in the

adaptive strategies of prehistoric communities than the


other habitats and subhabitats found within the coastal

strand region (Ranwell 1972:40-59).

Live-0ak Hammock Community

A "hammock" habitat is "a dense growth of mesophy-

tic (i.e., having moderate moisture requirements) broad-

leaved trees on a slightly raised substratum and not wet

enough to be a swamp" (Shelford 1963:67-68). It is char-

acteristic of the barrier island interiors and the main-

land as well--both of which are necessarily removed from

the biotically restricting conditions'.of salinity and

nutrient deficiency which are diagnostic of the beach-and-

dune habitat.

The major chemical evolutionary process that takes

place in the embryonic forest environment is the develop-

ment of the lime and clay components. Ground water com-

bines with CO2 in the soil to form carbonic acid (H2CO3 '

an unstable compound which can either return to water and

carbon dioxide, or can dissociate into a proton and nega-

tively charged bicarbonate ion (HC03). In the latter case,

combination with calcium present in rock will produce lime

(CaCO3), which, over time, accumulates deeper in the soil

material. In the case of clay, we are dealing with miner-

als such as aluminum and silicon which react with oxygen

to form a large, "sheet"-type lattice structure with a

large surface area. Both the clay and lime (as well as

other minerals) are carried downward in the soil, where

they ultimately consolidate to form a "B" horizon (Smith

1966:244, Colinvaux 1973:45, 201-202).

The aspect of this horizon of particular ecologi-

cal significance is its clay component. The surface of

the "sheet" lattice is negatively charged, which allows

it to retain nutritionally crucial cations (atoms bear-

ing a positive charge) such as calcium and potassium, as

well as some important trace elements.1 This lattice

"trap" of crucial nutrients thus enhances the possibility

of soil colonization by a plant communlity (Colinvaux 1973:


The plants which colonize the weathered rock ac-

celerate its breakdown into smaller granules by sending

down roots. The root systems additionally function as a

retrieval system for nutrients which have diffused deeply

into the soil. The end result is the generation of a cycle

of nutrient retrieval, plant growth, death, decomposition,

diffusion of nutrients downward, growth of new plants, and

nutrient retrieval. In the temperate climate of the south-

east, the decomposition rate is slow in comparison with,

for example, a tropical rain forest, resulting in an overall

The relation of soil chemistry and morphology to
Sr-38 analysis is dealt with in Chapter VII.

pumping upward of nutrients into an "organic reservoir"

or humus zone which is close to the surface of the soil

(Smith 1966:244, Colinvaux 1973:201-203).

The nutritional richness of this zone is further

enhanced by worms and insects which consume the fresh

litter deposited on the forest floor and excrete it in a

partially decomposed form. This organic detritus is then

further broken down by saprophytic bacteria into carbon

dioxide, water, minerals, and salts--a role directly com-

parable to that of the saprophytes in the "organic belt"

below the surface of the shore (Smith 1966:244).

The continual replenishing of the organic content

of the humus through forest litter and decomposers in com-

b~ination with the comparatively small loss of nutrients

through leaching eventually culminates in a developed for-

est community of higher plant and animal species. These

mixed deciduous and coniferous forests are vertically

stratified (i.e., they consist of multiple vegetation

"canopies"). Close to the forest floor are low shrubs such

as scrub palmetto, Sabal minor; gallberry, Ilex qlabra;

as well as a host of shrubs common also the back-dunes area

(yucca, cabbage palm, cassina, and saw palmetto). Above

this shrub stratum one encounters the vegetation of the

dominant trees, the live-oak, Quercus virginiana and the

magnolia, Magnolia grand_!~fliflora The primary ecological

significance of this vertical stratification of vegetation


is the accompanying stratification of conditions of light,

temperature, and moisture which it creates, resulting in

a multiplicity of stratified subhabitats for a variety of

different faunal species (Milanich 1972:6, Smith 1966:305).

The oak trees provide mast for white-tailed deer

(0docoileus virginianus), a herbivorous animal, while the

small insects, worms, and ground-layer vegetation support

the omnivore populations of raccoon (Procyon lotor), and

oppossum (Didelphis marsupialis). The forest floor addi-

tionally provides acorns and other seeds for herbivorous

fowl such as the turkey (Meleagris gallopavo). There are

thus both herbivorous and omnivorous vertebrate tropic lev-

els present in the hammocks, all of which represent potential

food resources for an aboriginal society (Milanich 1972:6).

The live-oak hammock habitat, then, is nutritionally

richer than that of beach-and-dune. Shielded from the ef-

fects of tidal action and wind, containing less soluble

soils, and characterized by a rela tively stable organic

reservoir in its upper humic level, it is a habitat that

could have been utilized either for prehistoric horticul-

ture, for hunting-and-gathering activities (or, conceivably

for both), and could thereby have been a significant habi-

tat in Georgia coastal subsistence activities.

Estuary and Tidal Flat Community

The estuary and tidal flat subhabitats evolved by

the geological processes described at the beginning of this

chapter. Their significance in the maintenance of the

coastal strand environmental system can probably best be

understood, not by discussing them as discrete units (the

procedure thus far), but rather by describing their inter-

actions with the other habitats discussed above. By con-

cluding the discussion in this way, we can more easily ap-

preciate the way in which the various coastal habitats are

knit together into a mutually interacting biological system.

An estuary/tidal flat habitat is an area of mixed

salinity, temperature, and turbulence. In an estuary, the

meeting of the river and the incoming tides creates a mixed

salinity as well as sufficient turbulence to shift nutri-

ents (primarily brought in by the tide) throughout the

depth of the water--a "vertical mixing" process. Tempera-

ture in the estuary fluctuates on a seasonal and a daily

basis, and is strongly affected by the temperature of the

incoming tides. In the tidal flat subhabitat, the salinity

and turbulence are not as strikingly variable as the temper-

ature. High tide on the flats either heats or cools them,

depending upon the season, while there is also a great

diurnal fluctuation as the "tidal flats are submerged in

salty water and then exposed to the full insolation of the

sun" (Smith 1966:215, 211-212).

The biotic community of an estuary is largely a

function of the fluctuation of salinity, and the turbu-

lence of the river currents meeting the incoming ocean

tides. The majority of the estuarine live forms are found

in its lower depths (i.e., they are benthicc"), and most

are securely attached to the bottom. The motile forms are

primarily crustaceans and fish, particularly the young of

the various species. Almost no fresh-water forms are

found (Smith 1966:213).

The organic matter brought in by the tide and the

marine plants of the estuary itself f,orm the primary pro-

duction base for its trophic levels. Herbivorous fish such

as mullet (Mugil cephalus) and the herbivorous Virginia

oyster (Crassostrea virginica) are both found in abun-

dance. A carnivorous marine trophic level is also present.

The sheepshead (Archosargus probatocephalus) feeds upon

barnacles and mussels, the black drum (Pogonias cromis)

feeds upon oysters and small fish, and the red drum

(Sciaenops ocellatus) feeds upon small shrimp and mullet.

Other marine carnivores include the salt-water catfish

(Arius felis) and the jackfish (Caranx hippos), both of

which will eat smaller fish (Milanich 1972:7, La Monte 1945:

36, 72, 79, 86). All of these marine fauna are potential

food resources for an aboriginal society, and, in the case

of the Virginia oyster and the schools of mullet, could

easily have been procured in abundance.

The tidal flats, as explained earlier, are the

result of an estuarine delta eventually becoming separated

from the sea by the development of barrier islands. They

are interlaced with tidal creeks that flow to the estuary

and the sea, and are supplied as well with some drainage

(or "leaks") from both the island and mainland biogeochemi-

cal systems. Organic debris from the leaks and the tides

provide a nutrient base for a durable form of vegetation,

salt water cord grass (Sprt~ina alterniflora), which ex-

tends throughout the tidal flats (Smith 1966:216).

Within the mud of the flats are found the Virginia

oyster (Crassostrea virginica), which feeds upon the organic

nutrient supply, and many herbivorous insects which feed

upon the salt-water cord grass. Both the insects and the

plants furnish a food supply for omnivorous reptiles such

as the snapping turtle (Chelydra serpentina), the box tur-

tle (Terrapene carolina), and the diamond-back terrapin

(Maclemys terrapin) (Smith 1966:216-217, Milanich 1972a:6).

All of these last, of course, are easily procured by humans

and can be gathered in abundance.

The systems of the habitats and subhabitats described

above are in constant interaction with one another and are,

like all ecological systems, open--or nondiscrete. This

is particularly true of the reservoir of organic nutrients

described in relation to tidal flats, estuary, and beach-

and-dune areas. In the tidal flats, we have discussed the

organic nutrient base furnished primarily by the tides,

and making possible the lush growth of Spartina. But only

5 percent of this primary production level is utilized by

tidal flat herbivores. The remainder is either decomposed

by saprophytic bacteria into methane and hydrogen sulfide

(some of which is drained to the sea) or is transported

by tidal creeks into the estuary, where it supplements the

amounts of organic materials brought in with the tides

(Smith 1966:217).

There are additional interactions involving the

biotic communities of the vertebrates. Live-oak hammock

herbivores such as rabbits and white-tailed deer utilize

back-dune vegetation, while forest omnivores such as rac-

coon and opossum can consume not only small plants in the

back-dune and tidal-flat areas, but can feed upon small

fish and crustacea as well.

In sum, the live-oak hammock barrier islands pro-

vide the shelter and protection that makes the tidal flats

a possibility. The tidal flats provide food resources for

high-hammock herbivores and omnivores, and additionally

transport an extensive nutrient supply to the waters of

the estuary. The estuarine fauna, particularly the small

fish and crustacea, are protected from extensive predation

by the limited salinity of the estuary, to which larger

carnivorous fish are less adapted. The estuary thus func-

tions as a "nursery," for organisms which eventually will

move to the sea. The sea, in its turn, both "receives"

fish and crustacea from the estuary, and provides in turn

an organic nutrient supply to both the estuary and the

beach. The changes along barrier-island coastlines, how-

ever, result in flat beach becoming dune, with the nutrient

reservoir forming a food base for xerophytic back-dune veg-

etation. This back-dune vegetation, in turn, furnishes a

primary production level for high-hammock herbivorous and

omnivorous mammals, while the back-dune permanence and high-

er contour provides a "barrier" of its own between the high-

hammock forest and the high salinity of the beach.

Interaction between the habitats, then, is pronounced

--and occurs on a multiplicity of levels. We have not yet

considered, however, the way in which the various habitats

were utilized by the coastal strand peoples of protohistoric

and early historic times. It is that question, and, spe-

cifically, the methodological procedures involved in

its archeological investigation, to which we will next di-

rect our attention.



Location and Methodological Problems

"Taylor Mound" is the name assigned by the inves-

tigator and Dr. Jerald T. Milanich to a historic period

(ca. A.D. 1600-1650) ceremonial mound with associated bur-

ials located to the southwest of Taylor Fish Camp on the

northern end of St. Simon's Island, Georgia, at 31' 15'

north latitude and 810 21' west longtitude. The structure

was located in a grove of trees in an area that was undis-

turbed by plowing. This advantage was offset, however, by

an earlier unsystematic excavation by local residents which

had removed 13 burials from a trench dug east-to-west across

the apex of the mound. This earlier excavation had de-

voted only slight attention to an artistic and photographic

recording of the burials in question (and apparently none

at all to trained osteological identification) before the

skeletal material was redeposited, in no apparent order,

into the trench from which it had been taken.

This earlier disturbance creates a number of ser-

ious analytical problems, all of which should be made clear at


the beginning of this discussion. The rimajor difficulties

would seem to be (1) the identification of the strati-

graphic and, when applicable, the mortuaEry context of

artifacts recovered in the earlier excavation (i.e., were

materials found above a burial, within a construction

feature of the mound, in the burial pit rmatrix, or in juxta-

position with the skeleton itself?), (2) ti, identification

of sex and age of the burials, and (3) the l ated problem

of sample size. We will discuss each of th .e problems in

its turn.

The report of the earlier excavation did not include

vertical control in terms of distance below datum plane

(a concept to be presented in the methodological discussion

below), but rather in terms of distance below surface, and

even this approach was limited primarily to the discussion

of the burials and their associated artifacts. The result

of this omission is the impossibility of cross-checking

any earlier statements with regard to the stratigraphic

contect of artifactual materials. We may take, as an

example, the fundamental matter of the stratigraphic con-

text of the burials (and, by implication) of their arti-

facts. The report on the earlier excavation described

"Burial 10" as being located on the eastern edge of the

mound's shell core, its surface corresponding to the base

of that core (Pearson and Cook n.d.:20). As this is a

burial that allegedly was associated with chronologically

significant historical materials, it is essential to know

if the burial was intrusive into (and thus postdated) the

core and the fill (as the excavator contends) or whether

it was located within the sterile matrix beneath the mound.

The latter would seem to me unlikely, but not impossible,

and the absence of sufficient vertical control precluded

the possibility of either cross-checking the stratigraphic

accuracy of the earlier report or investigating the inter-

esting possibility that the original rationale of the shell

core of the mound was to serve as a marker for a cluster

of burials beneath it. This burial characteristic has been

previously noted for the coastal tribes (Larson 1953:15)

and would serve as an additional line of evidence for the

contemporaneity of the Taylor Mound site with Couper Field

(to be discussed in detail below). In view of this problem,

then, it seems wisest to use mortuary and artifactural data

recovered at the time of our own investigation, when dis-

cussing cultural traits of the builders of Taylor Mound,

and when relating Taylor Mound to other sites. The second

problem is more readily disposed of. Information regarding

any significant association of grave goods of a certain type

(e.g., "exotic" versus "local"), or, for that matter, pres-

ence of absence of grave goods, according to the status of

sex or age is only as reliable as the osteological examina-

tion of the skeletal material involved. In this regard,

then, only the Taylor Mound skeletal material recovered by

the writer (examined in situ by a physical anthropologist)

will be utilized in discussing cultural practices reflected

in mortuary behavior.

Implicit in the above limitations is the special

problem of sample size. Eleven burials were recovered in

our own investigation, one of which was too fragmented

(and too young) to be identified as to sex, the other in

a matrix which made it difficult to determine if artifac-

tual materials were deliberately associated. As a conse-

quence, there are at least two statistical strategies that

are available. One of these is the utilization of tests

(such as Fisher's Exact Probability Test) which are speci-

fically designed for statistical problems in which anti-

cipated frequencies are small (Freeman 1965:227). The

other approach is to investigate the possibility of signi-

ficant similarity between mortuary practice at Taylor Mound

and Couper Field (in which many undisturbed burials were

found) so that these two samples (each small in itself)

might be legitimately combined when assessing the archeo-

logical evidence for cultural practices along the Georgia

coast. We will deal at some length with these alternative

possibilities in the presentation which follows.

Methodology of Excavation

The excavation was conducted by means of the stan-

dard-block system. Horizontal control was maintained

through a grid system consisting of three-meter squares,

the primary excavation unit. Vertical control was main-

tained by means of measuring distances below a datum plane

swept by the horizontal movement of the scope of a Dietzgen

Top-Site transit. Fixity of the plane was maintained by a

permanent transit station.

An east-west grid was established across the mound,

oriented at 600 east of magnetic north. A stake placed

on the western edge of the mound was arbitrarily desig-

nated 100N 100E, and all other stakes were numbered accord-

ingly. All units (i.e., all three-meter squares) through-

out the excavation were designated by the coordinates of

the stake in the southwest corner of the unit, a convention

of the University of Chicago grid system.

Excavation of all three-meter squares proceeded by

removal of the "natural," i.e., the geophysical, strata

(including cultural strata) within each unit. Vertical

control over these strata was maintained by means of datum

shots from each corner of the unit, in addition to a cen-

ter datum shot, for the top and the bottom of each stratum.

Bone and cultural materials from each zone were maintained,

and were placed in bags labelled according to unit coordi-

nate and physical/cultural stratum. The materials were

removed from their matrix by shovelling out the soil in

each unit, usually in thin layer. amounting to approximately

an inch in depth (or two and one-half centimeters). The

soil was placed on a motorized, vibrating screen of one-

quarter inch mesh, powered by a five horsepower engine.

Bones and cultural materials remaining on the screen were

maintained. The soil was not maintained, but was sampled

as described below.

A feature was operationally defined as any aspect

of archeological significance in the site which could not

be moved intact to a laboratory for analysis. Features of

Taylor Mound included a fire pit in a matrix of projecting

shell (a construction feature to be described below) and

sterile submound sand, a line of worked conch shells in the

mound fill on the northeastern edge of the shell core, and

a boat-shaped depression cut into the old humus on the east-

ern edge of the mound, which contained a cache of broken

pottery conventionally designated as representing San Marcos

Stamped, Irene Incised, and Irene Plain types, and which

was covered by soil removed from the mound fill. These

features were mapped horizontally when they were first

encountered, and vertical control was maintained by datum

shots on both the surface and the bottom of each feature.

Morphology and contents were recorded on feature forms,

and cultural contents were bagged and labelled with pro-

venience. A soil sample was taken from the fire pit men-

tioned above, and was water-screened through graduated

cylinders in an unsuccessful attempt to locate charred

fragments of corn as evidence of aboriginal horticulture.

Other than determining the sequence of mound con-

struction and use, the only remaining methodological as-

pect of major importance in the Taylor Mound excavation

was the removal of the associated burials. The procedure

that was followed varied in accordance with the nature of

interment as well as with special problems which were en-

countered in a number of cases. If the burial was within

a pit, the contents of the pit were removed by a trowel.

A bag of this soil was maintained in such cases for the

graduated cylinder water-screening referred to above. Once

all soil had been removed from the pit by trowelling, the

small amount in direct contact with theskeleton was re-

moved by a brush. At this point, the nature of the asso-

ciated cultural materials (if any), as well as tentative

osteological analyses were recorded on burial forms. If

the burial was not in a pit (e.g., intrusive into either

the shell core or the "shell shelf," or placed on top of

submound sterile soil), a modification of the above pro-

cedures was required. In the first instance, the burials

in question were highly fragmented, and could not be given,

in one case, osteological analysis sufficient to determine

the sex of the individual. Nevertheless, burial forms were

prepared in as much detail as possible and a photographic

record was maintained. In the second instance (that of the

burials placed on top of submound sterile soil and covered

with fill), all soil except that in direct association

with the skeleton itself was removed by trowel, such that

the skeleton was left supported on a "pedestal" of soil.

This provided a kind of "working table" situation in which

both the brushing and subsequent osteological analyses of

the skeletons were facilitated. Photographs were taken,

burial forms were prepared, and the skeletons were sealed

in plastic until the systematic physical anthropological an-

alysis was done.

This analysis was conducted by Dr. William R.

Maples of the Department of Anthropology and the Florida

State Museum at the University of Florida. It consisted

of basic demographic data of sex and approximate age of

the individual. All visible pathologies were noted and

described (e.g., enlargement of left tibia, cranial osteo-

porosis), but diagnostic labels were not assigned. Upon

completion of the analysis of each burial, the skeleton

was removed from its context. This removal involved the

individual bagging of each skeletal element, and the placing

of the bags in a labelled box. The burials were then sent

to the Social Science Department of the Florida State

Museum for further analysis.1 Upon completion of the

The osteological analyses of Taylor Mound, Couper
Field, and Indian Field skeletal materials are still con-
tinuing at the time of this writing. These are being con-
ducted by James Zahler under the direction of William R.
Maples. It is possible that these investigations may ne-
cessitate a later reconsideration of some of the sociocul-
tural findings of this report.

excavation, the area was backfilled by Sea Island Com-

pany, Sea Island, Georgia, who served as cosponsors of the


The Architectural Characteristics of Taylor Mound

Taylor Mound is probably best described as a small

ceremonial mound with associated burials. The term "cere-

monial" rather than burial seems advisable in view of the

architectural elaboration of the southern portion of the

structure and the presence of the "east-side pottery cache,"

both of which have possible ceremonia? significance, as

evidenc ed by ethnohistorical documentation to be discussed

in more detail below.

Essentially, the mound consisted of a shell "core"

which was placed in an area in which much of the topsoil

had been previously scraped away, and which was surrounded

by soil removed from a (nearly) surrounding "borrow pit."

The eliciting of the probable steps of construction was

made possible by considering the most economic explanation

of the stratigraphy of the mound which was evident in pro-

file, particularly the strata of a trench 19.77 meters in

length (15.40 meters of which was oriented grid east-west

across the apex of the mound, and is partially illustrated

below, while the remaining 4.37 meters was the profile of

a trench extending grid northeast from the shell core).

Figure 5.

Stratigraphy of Taylor Mo'und

A. Modern Shell and Humus

B. Mound Fill

C. Mound Wash and Borrow Pit Fill

D. Humus

E. Borrow Pit Humnus

F. Sterile Sand

G. Gray Sand

H. White Sand

I. Leached Borrow Pit Humus


97 E Sca e 1:20





F .

... ~10 0N
I I I1

The strata evident in profile were as follows: (1) modern

humus, (2) mottled brown soil immediately beneath the humus,

which sensedd out" at approximately 1.5 meters from the

edge of the shell core, (3) darker soil (similar to modern

humus) immediately beneath the mottled-brown zone, (4) a

moderately lighter zone beneath this darker soil, (5) tan

sterile sand, (6) mottled tan soil in contact with the

core, which sensedd out" at the onset of (2), and, finally,

(6) the shellfish "core" itself.

The simplest explanation of the mound construction

sequence would seem to be the following: The greater part

of the humus of the site of the shell core was scraped

away prior to the placement of the core. A nearby mid-

den, consisting primarily of shellfish, was scraped up,

brought to the prepared area, and deposited on the subsoil.

This core was square measuring 7m on a side (measured at the

top) and 1m thick. The sides sloped outward at an angle

of about 300 above horizontal. A large moat-like "borrow"

pit extending some 4 meters from the edge of the fill was

dug down into the sterile subsoil on the north, west, and

south sides of the core. The soil from the pit was placed

against all sides of the core, extending out 2.10 meters.

This sand was termed the "mound fill." Upon completion

of the structure, the sherds from as many as ten broken

vessels were deposited in a shallow depression dug into the

topsoil at the eastern edge of the fill. In the course of

time, burials were added to the structure, and (with one

exception which involves a unique construction feature

to be discussed below) they were either placed in pits

dug into the core, within the fill itself along the eastern

side, or within the sterile soil on the eastern side, with

mound fill subsequently added.

The southern side of the mound was the only side

exhibiting any type of structural elaboration. In this

instance, the shell core itself was modified by the crea-

tion of a projecting "shelf" of compacted shell. The shelf

extended outward approximately two meters (being eroded at

the edge), and followed the southern edge of the core for

at least two meters, until traces of it were obscured by

a combination of a hole dug into the southeastern edge of

the core during the previous excavation and a tree which was

rooted in the core. Beneath the projecting shelf was ster-

11e sand, while above it was mound fill. Extending south-

ward from the shelf approximately one meter were two molded

"steps" of sterile soil leading downward until they became

the bottom of the borrow pit. The only other unusual con-

struction feature was a hypothetical sand "topping" mounted

on the surface of the core. This was evidenced by sand

that was present throughout the core's matrix as well as by

sand washed into the borrow pit. Whether this sand topping

was present or not would not be a question of any signifi-

cance beyond that of descriptive architectural history, if

Figure 6. Steps of Mound Construction

1) Shell core with projecting shell shelf
(B) is mounted on partially cleared

2) Soil from surrounding borrow pit (B)
is placed around the core (A)

3) Sterile sand is molded to create an
extra step (B)

4) A sand topping (A) is added, to create
six steps on the southern side (A
through F)

co us sV c U A

J', uo

~,-U 57
I-. ~ Ya


it were not for a possible ethnohistorical significance

involving a structural reflection of aboriginal magico-

religious systems. Properly, this will be presented in

more detail in the appropriate chapter, but here we may

take note that, if the "topping" were actually present, it

would have resulted in a series of six distinct steps be-

ing evident on the southern portion of the mound. These

would have been: (1) the sand "topping" (2) the exposed

portion of the core between the edge of the "topping" and

the beginning of the fill (that the topping was not coter-

minous with the core was suggested by the centralized con-

centration of the sand), (3) the mound fill, (4) the pro-

jecting "shelf," and (5) and (6) the two molded "steps"

of sterile sand. The account of the Guale Festivals in

the writings of Father Juan Rogel describes them as

having occurred six times during the year (Rogel 1861 [1570]:

328). Further, the account provided by Peter Martyr

d'Anghera states that the south was the location of the af-

terworld, in the belief systems of the coastal strand tribes

(Martyr in Swanton 1922:44-54). Combining these two reports,

we are left with the possibility that the elaboration of the

southern portion of the structure through the construction

of six discrete levels did indeed occur, and would consti-

tute an architectural reflection of an aboriginal magico-

religious system.

The above sequence of construction would seem to

be, then, the simplest explanation of the mound's strati-

graphy. It remains only to be said that the additional

strata can be explained with equal facility as the re-

sult of the combined processes of erosion and deposition

through time. Apparently the mound fill was washed by

rains (which occur all year around, with convectional thun-

derstorms occurring almost daily in the summer) into the

surrounding borrow pit. This process, however, would

have been slower than that of deciduous forest humus ac-

cumulating in the bottom of the pit. As a consequence,

mound fill was being washed down on top of accumulated

borrow pit humus (stratum (3) above) which was itself

leached out (creating stratum (4)). Finally, on top of

this "mound wash and borrow-pit fill" (2), there exists

an accumulation of modern humus (stratum (1)).

The final feature of interest in the ceremonial

structure was the east-side pottery deposit. This is a

common feature of aboriginal mounds in Southeastern United

States. Basically, it is a deposit of potsherds from ce-

ramonial vessels that were possibly broken as part of a

ritual commemorating the completion of the structure; nor-

mally, the deposit is on the eastern side of the mound.

The pottery cache at Taylor Mound was located to the east

of the shell core. It was a boat-shaped hole dug into

the old humus, filled with pottery fragments, and covered

with mound fill. Ceramic types present included San

Marcos (rectilinear complicated stamped), Irene incised,

Irene Filfot, and Irene Plain. At the Florida State

Museum, University of Florida, five vessels have been

largely restored from the sherds. They are the follow-

ing: (1) Irene incised vessel, circumferences: 39cm,

top; 47cm, middle; 2) Irene incised, circumferences: 36.8cm,

top; 58.0cm, middle, 3) San Marcos Complicated Stamped,

circumferences: 64.0cm, top; 59.5cm, middle; height 19.3

cm, 4) Irene Plain, circumferences: 53cm, top; 45.7cm,

middle; height 14.0cm; 5) Irene Incised, circumferences:

46.0cm, top; 49.5cm, middle; height 14.0cm. While the

presence of the deposit is important in establishing af-

finities between the Guale end inland Southeastern tribes,

it cannot be used in establishing the chronological rela-

tionship between Taylor Mound and other St. Simon's Island

sites, due to the misleading effect of comparing ceremonial

with utilitarian ceramics as well as the inevitable skewing

effect arising from sample variation.

It should also be noted that a number of European

artifacts were encountered at Taylor Mound. These included

(1) a -brass spike encountered in the mound fill to the im-

mediate south of the burials, 10.5cm long with a square,

tapered shaft, 1.10cm on a side near the top, and .80cm

on a side near the bottom; (2) a round ship's spike located

in the modern humus on the immediate north of the shell core:

30.00cm long and 6.6cm in circumference; (3) two nails

(one 18.4cm long, the other 15,00cm long) and one spike

15cm long, all encountered near the top of the mound fill

to the immediate east of the shell core. The presence of

these last artifacts within the fill adds greater terminus

post quem certainty to the protohistoric dating of the


Such were the architectural characteristics of

Taylor Mound. In concluding, we might stress that, while

its gross dimensions would not have made it imposing (in-

deed, it would be miniscule by comparison with many cere-

monial mounds), its southern elaboration and east-side

pottery deposit would nevertheless characterize it as a

ceremonial structure. It did, however, serve a dual

role (the other being that of burial) and it is to the

matter of the Taylor Mound burials and their cultural

significance that we will next direct our attention.

The Burials at Taylor Mlound

Eleven burials were recovered from Taylor Mound.

In this section, we will deal first with a description of

each of these burials, and secondly, with cultural infor-

mation that may be elicited from them. In the second in-

stance, we will examine (1) the degree of similarity among

the Taylor Mound burials, (2) the similarity among the

Couper Field burials, and (3) the homogeneity between

burials when the Taylor Mound and Couper Field populations

are combined. This determination is a necessary prior

step toward determining the appropriate test to be used

(e.g., chi-square versus Fisher's Exact) when investigat-

ing (for example) the association of types of grave

goods with sex and age, as an index of differential pres-

tige. At the moment, however, a brief description of the

Taylor Mound burials is in order.

The description of all burials to be considered

in both Taylor Mound and Couper Field will utilize the

following arbitrary categories: (1) sex, (2) age (infant,

child or adolescent, adult, aged), (3) burial type (primary/

secondary), (4) position of bones (extended/flexed/other),

(5) position of head (left side/right side/face down/face

up/other), (6) [+, -] marine fauna, (7) [+, -] terrestrial

fauna, (8) [+, -] unmodified stone, (9) [+, -] modified

stone, (10) [+, -] modified shell, (11) [+, -] modified

bone. In the case of the more general categories (such

as "worked stone"), the artifact will be specifically

identified (e.g., "projectile point," "stone axe"). The

description, then, of the Taylor Mound burials is provided


Burial One was located on the southeastern side

of the mound, intrusive into the projecting "shell shelf"

and the molded "step" of sterile sand immediately beneath

Scale 1:4

Figure 7.

Taylor Mound Burial 1

it. Osteological analysis determined that the burial was

an infant aged approximately 36 to 38 weeks. The nearly

2000 beads in association with the burial were arranged

in the form of a bracelet, an anklet, and an apron. The

bracelet and anklet were further characterized by alter-

nating strands (2 white, 2 red, and 2 white) for a total

of six different strands. This number is the same as the

total levels on the southern portion of the mound (dis-

cussed above), implying (along with the fact that the bur-

ial was the only one with a southern location) that the

placement of this individual possessed a possible magico-

relisious significance--and may have been an infant sacri-

fice. Distinctive features of the burial include the fol-


1) indeterminate (IND) 7) no terrestrial fauna (-)

2) infant (INF) 8) no unmodified stone (-)

3) primary (PR) 9) no modified stone (-)

4) flexed (FL) 10) modified shell (1954 shell

5) left (L) beads and a painted clam

6) no marine fauna (-) shell) (+)

11) no modified bone (-)

Burial Two was located in the northeastern portion

of the mound fill. Apparently, it has been placed on the

submound surface (primarily sterile sand) and covered with

fill. The following distinctive features were noted:

Figure 8.

Taylor Mound Burial 2

1) female (F) 7) no terrestrial fauna (-)

2) adult (AD) 8) no unmodified stone (-)

3) primary (PR) 9) no modified stone (-)

4) flexed (FL) 10) no modified shell (-)

5) right (R) 11) no modified bone (-)

6) marine fauna (two clam

shells, one contained

inside the other) (+)

Burial Three was located in a pit intrusive into

sterile soil beneath the northeastern part of the mound

fill. Distinctive features were the ,following:

1) male (M) 7) no terrestrial fanua (-)

2) aged (AG) 8) no unmodified stone (-)

3) primary (PR) 9) no modified stone (-)

4) other (supine with legs 10) no modified shell()

flexed) (SUP-FL)

5) other (vertical) (VERT)

6) marine faunal (13 clam

shells and one busycon

shell) (+)

Burial Four was located in a pit intrusive into

the sterile soil beneath the mound fill on the southeastern

Determination of the "worked" or unworkedd" status
of busycon shells is especially difficult. In this case,
we are assuming that it was unmodified.

Scale 1:5

Figure 9.

Taylor Mound Burial 3

Taylor Mound Burial 4

Figure 10.

side. Distinctive features were as follows:

1) female (F) 7) no terrestrial fauna (-)

2) adult (AD) 8) no unmodified stone (-)

3) primary (PR) 9) no modified stone (-)

4) flexed (FL) 10) no modified shell (-)

5) right (R) 11) no modified bone (-)

6) marine fauna (38 frag-

ments of shell) (+)

Burial Five was located in a pit dug into sterile

soil beneath the east-central portion of the mound fill.

Distincive features were:

1) female (F) 7) no terrestrial fauna (-)

2) adult (AD) 8) no unmodified stone (-)

3) primary (PR) 9) no modified stone (-)

4) flexed (FL) 10) no modified shell (-)

5) right (R) 11) no modified bone (-)

6) marine fauna (113 frag-

ments of shell) (+)

Burial Six was located within the east-central part

of the mound fill, and was above Burial 9, described below.

Its distinctive features included the following:

1) male (M) 7) no terrestrial fauan (-)

2) adult (AD) 8) no unmodified stone (-)

3) primary (PR) 9) no modified sonte (-)

4) flexed (FL) 10) modified shell (78 shell

5) right (R) beads) (+)

6) marine fauna (36 shell

fragments) (+)

Figure 11.

Taylor Mound Burial 5

Figure 12.

Taylor Mound Burial 6


Burial Seven was located in a pit dug into sub-

mound sterile soil on the east-central part of the mound,

and was covered with mound fill. Distinctive features

were the following:

1) female (F) 7) terrestrial fauna (two

2) adolescent (ADO) box turtle shells placed

3) primary (PR) along vertebral column)(+)

4) flexed (FL) 8) no unmodified stone (-)

5) right (R) 9) no modified stone (-)

6) marine fauna (oyster 10) no modified shell (-)

shells) (+) 11) no modified bone (-)

Burial Eight was a highly pathologic individual

exhibiting premature closure of cranial sutures, osteo-

porosis of both parietals and the posterior frontal, a

vertical forehead, billowed and unfused cranial condyls,

ilium unfused to ischium, and cortical thickening of the

right tibia. The tentative in situ field analysis by

William R. Maples resulted in noting the possibility of

either an endocrincological pathology or abnormal hemo-

globin. Subsequent flouroscopic analyses of the skeletal
material of Burial 8 conducted in the J. Hillis Miller

Medical Center at the University of Florida (and which

are still in progress at the time of this writing) have

provided a tentative diagnosis of some type of hemolytic

animia, which may be thalassemia. Essentially, this is

a pathology in which normal hemoglobin, a conjugated

Figurre 13.

Taylor Mound Burial 7

Figure 14.

Taylor Mound Burial 8

Scalte 1:5

protein responsible for oxygen transport to the tissues,

is not synthesized, but a variant form, in which one of

the two amino acid chains is suppressed, is synthesized

instead (Buettner-Janusch 1973:453-5). Its 01d Wlorld

distribution included most of Spain, and it is relatively

frequent among Arab populations. Moreover, it appears in

two forms (major and minor), which vary strongly in likeli-

hood of fatality, severity of anemia, degree of jaundice,

enlargement of spleen and liver size, and so forth. The

less severe form (minor) is generally not manifested

strikingly in the skeletal material; ,the more severe form

is. Finally, the major and minor variations are homozygous

and hetereozygous conditions, respectively.

Burial Eight was located in a pit dug into the sub-

mound sterile soil on the east-central side of the mound,

and was covered with mound fill. Its distinctive features

were the following:

1) female (F) 7) no terrestrial fauna (-)

2) adolescent (ADO) 8) no unmodified stone (-)

3) primary (PR) 9) no modified stone (-)

4) flexed (FL) 10) no modified shell (-)

5) right (R,) 11) no modified bone (-)

6) no marine fauna (-)

Burial Nine was located in a pit dug into sub-

mound sterile sand on the east-central side of the mound,

and was covered with mound fill. Distinctive features

included the following:

Figure 15.

Taylor Mound Burial 9

1) female (F) 7) no terrestrial fauna (-)

2) adult (AD) 8) no unmodified stone (-)

3) primary (PR) 9) no modified stone (-)

4) flexed (FL) 10) no modified shell (-)

5) right (R) 11) no modified bone (-)

6) no marine fauna (-)
Burial Ten was located in a submound sterile soil

pit on the east-central side of the mound, and was covered
with mound fill. Its distinctive features were as follows:

1) female (F) 7) no terrestrial fauna (-)

2) adult (AD) 8) no unmodified stone (-)

3) primary (PR) 9) no modified stone (-)

4) flexed (FL) 10) no modified shell (-)

5) other (face up) (FU) 11) no modified bone (-)

6) no marine fauna (-)

Burial Eleven was located in a pit intrusive into

the shell core. It was located beneath a tree, and was

totally in fragments, probably as a result of root action.

Associated materials presented below were those in direct

association with the skeletal fragments.

1) female (F) 7). no terrestrial fauna (-)

2) adult (AD) 8) no unmodified stone (-)

3) primary (PR) 9) modified stone (flintchip)

4) flexed (FL) (+)

5) right (R) 10) modified shell (shell beads)
6) marine fauna (one whelk (+)>

and one oyster shell)(+) 11) no modified bone (-)

Scaile 1:5

Taylor Mound Burial 10

Figure 16.

of the chart is

the apparently pronounced homogeneity between the burials.

Our next task will be the utilization of these same cate-

gories for a description of the Couper Field burial complex

to determine, by examination, if the burials appear to pos -

sess the same type of homogeneity as the burial complex at

Taylor Mound.

Taylor Mound burials may be summarized as in the

table below:

Table 1

Summary of Mortuary Distinctive
Features at Taylor Mound

Burial (1) (2) (3) (4) (5)





(10) (11)

L -


RT +

R +

R +-

R +

R +

















- - +

+ - -

+ + -

Probably the most evident aspect

Summary of Investigation of Taylor Mound

Taylor Mound, a protohistoric ceremonial mound with

associated primary burials, was the subject of two arch-

cological investigations, the first of which presented ser-

ious methodological problems for the second. Essentially,

these were problems of the reliability of earlier strati-

graphic and osteological data, both of which were respon-

sible for a related problem of sample size. The decision

which seemed most advisable was to use only the data re-

trieved during the second excavation, and to examine

statistically the possibility of combining the resultant

small sample of mortuary data with the samples from other


The primary result of the second excavation was

the eliciting of detailed architectural and mortuary infor-

mation. In the former case, it was determined that the

mound was constructed by placing a shell midden atop a

prepared area of sterile soil, surrounding it with soil

removed from a borrow pit, and possibly capping the struc-

ture with a topping of fine sand. In the second instance,

it was determined that a pronounced mortuary homogeneity

with little deviation appears to exist among the burials,

suggestive of only limited rank distinctions in the cul-

ture. 'his finding, however, requires a larger burial

sample for corroboration, a matter to be explored in the

following chapter.



"Couper Field" was the name assigned by the in-

vestigator to an abandoned modern field on the northern

part of St. Simon's Island, Georgia, immediately south of

the remains of the ante-bellum Couper mansion. Unlike

Taylor Mound, the area had been heavi,1y plowed, and the

sparseness and height of the pine trees on the site sug-

gested that secondary succession had begun only within the

last few decades. Nevertheless, the majority of burials

which were recovered from this protohistoric burial-

complex site were undisturbed by plowing and afforded a

larger data base than those which were recovered at Taylor

Unlike the mound excavation, the Couper Field

topography was suggestive with regard to possible con-

centration of archeological materials, but was not con-

clusive. There were throughout the area many small

circular rises of shell (ca. 50 cm high at the apex) which

were probably deposited around dwellings. In order to

determine specifically the midden accumulations that wecre

directly associated with living areas or concentrations

of features, a sampling design was necessary.

The paramount considerations in the design were

uniformity (i.e., that there should be no clustering of

sampling units in any particular sector of the site)

as well as the limitations of time and personnel. In view

of these considerations, the entire open field was not

sampled, but instead, was covered by a 60m N-S, 90m

E-W grid which created a 5400m2 belt" across its middle.

Within this framework, it was possible to establish what

was deemed the appropriate sampling design. This w~as a

"chain" or "systematic" sample, an interval sampling tech-

nique in which every nt unit is measured (Slonim 1960:

57-59). In content analysis, it might be every tenth

page of a document; in sociological surveys, it could be

every fifth house along a street. The special advantage

of the technique is the avoidance of the sample-variation

skewing which can occur in random sample designs (e.g.,

numbered excavation units being selected through a table

of random numbers), but there is the special disadvantage

of the sampling pattern coinciding with a "natural" pattern

that is present in the data itself (e.g., every fifth

house is an expensive corner lot, and the chapters of the

document are usually ten pages long). In bte present

instance, the interval selected was 15 meters: every 15

meters rectilinearly within the grid a three-meter square

was excavated, for a total of 28 such units. There was no

apparent coinciding of this pattern with degree or depth

of plowing, nor with any vegetative or topographic vari-

ations. Each unit was excavated to the depth of sterile

soil (mean depth of the plow zone was .25m), and horizontal

maps were prepared of all features present within each unit.

Time and personnel limitations, sample objective, and the

disturbed context of materials in the plow zone made

motorized screening undesirable. Upon completion of the

sample, a composite map was prepared. Examination of this

map revealed a clustering of features Iparticularly

shellfish and food bone concentrations) in the western

part of 5400m2 belt. As a consequence, a sector 45m N-S

and 15m E-W, which enclosed this clustering of features,

was designated for intensive excavation.

Within the above sector, a smaller grid system- of

three-meter squares was established. Horizontal and verti-

cal control were maintained in the same manner described

for Taylor Mound, and methodology involving features and

burials was the same. The latter was the overall archeo-

logical characteristic of the site--and it is therefore

to the burials and their cultural implications that we

will next direct our attention.

The Burials at Couper Field

Sixteen burials were recovered at Couper Field.

In the present chapter, we will continue the discussion

and analysis initiated with the Taylor Mound burials:

(1) the description of the burials according to an arbi-

trary set of distinctive features, (2) the investigation of

site homogeneity by systematic comparison of these fea-

tures, (3) the examination of the legitimacy of combining

the Couper Field and Taylor Mound burials into a single

sample, and (4) the investigation of sociocultural ques-

tions regarding aboriginal ranked groups and postmarital

residential patterns. We begin, then, with a description

of the Couper Field burial complex, incorporating the same

distinctive features that were previously used.

All burials (with limited exceptions to be noted

below) were located in pits dug into the tan sterile sand

beneath the plow zone. Burial One was characterized by

the following features:

Burial One:

1) female (F) 7) no terrestrial fauna (-)

2) aged (AG) 8) no unmodified stone (-)

3) primary (PR) 9) no modified stone (-)

4) flexed (FL) 10) modified shell (+t) (shell beads)

5) right (R)

6) marine fauna (oyster shell) (+)

Concent at i on

Figure 17. Couper Field Burial 1

Scale 1:5

Burial Two:

1) female (F)

2) adult (AD)

3) primary (PR)

4) flexed (FL)

5) left (L)

6) marine fauna (burial pit was

lined with oyster shell) (+)

Burial Three:

1) female (F)

2) adult (AD)

3) primary (PR)

4) flexed (FL)

5) face down

6) no marine fauna (-)

7) terrestrial fauna (+) (animal

bone was located along the spine)

8) no unmodified stone (-)

9) no modified stone (-)

10) no modified shell (-)

11) no modified bone (-)

no terrestrial fauna (-)

no unmodified stone (-)

no modified stone (-)

modified shell (shell beads in

association with cervical

vertebrae) (+)

Burial Four presents a problem comparable to that of

Burial Eleven at Taylor Mound. It was located within the

plow zone, was heavily disturbed, and included several

individuals. One of these, burial four (c) consisted only

of a vertebral fragment in association with a greenstone

celt. The celt was spatulate, and was similar to those

frequently encountered in Mississippian sites. Its length
was 3.16 cm, and its width was 6.3 cm at the base and 4.5 cm

at the top. Its circumference was 13.4 cm at the midline.

Figure 18. Couper Field Burial 2

Figure 19. Couper Field Burial 3

Only one burial (4D) was sufficiently undisturbed to be

characterized by our present set of features. These were

as follows:

female (F)

adult (AD)

primary (PR)

flexed (FL)

other (vertical) (VERT)

no marine fauna (-)

7) no terrestrial fauna (-)

8) no unmodified stone (-)

9) no modified stone (-)

10) no modified shell (-)

11) no modified bone (-)

Characteristics of Burial Five were the follow-


male (M)

adult (AD)

primary (PR)

flexed (FL)

left (L)

marine fauna (seven

shark's teeth, two clam

shells) (+t)

7) terrestrial fauna (bird claw,

small animal mandible and skull (+)

8) unmodified stone (igneous rock) (+)

9) modified stone (two flintchips;

one slate celt, 35.7 cm long, 5.5

cm wide at the "flared" base, 4.9

cm wide immediately above the

flare, 3.5 cm wide at the top, and

a midline circumference of 10.5 cm;

one stemmed projectile point) (+)


Scale 1:5

Figure 20.

Couper Field Burial 4C

Figure 21.

Couper Field Burial 4D

0 10 CM

Figure 22. Couper Field Burial 5

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 2.9.9 - mvs