Aftermath of a feast

MISSING IMAGE

Material Information

Title:
Aftermath of a feast human colonization of the Southern Bahamian Archipelago and its effects on the indigenous fauna
Physical Description:
xv, 279 leaves : ill. ; 29 cm.
Language:
English
Creator:
Carlson, Lisabeth Anne
Publication Date:

Subjects

Subjects / Keywords:
Ethnozoology -- Turks and Caicos Islands -- Grand Turk   ( lcsh )
Antiquities -- Grand Turk (Turks and Caicos Islands)   ( lcsh )
Colonization -- Grand Turk (Turks and Caicos Islands)   ( lcsh )
Anthropology thesis, Ph. D   ( lcsh )
Dissertations, Academic -- Anthropology -- UF   ( lcsh )
Genre:
bibliography   ( marcgt )
non-fiction   ( marcgt )

Notes

Thesis:
Thesis (Ph. D.)--University of Florida, 1999.
Bibliography:
Includes bibliographical references (leaves 246-278).
General Note:
Printout.
General Note:
Vita.
Statement of Responsibility:
by Lisabeth A. Carlson.

Record Information

Source Institution:
University of Florida
Rights Management:
All applicable rights reserved by the source institution and holding location.
Resource Identifier:
aleph - 021550229
oclc - 43698124
System ID:
AA00022200:00001

Full Text











AFTERMATH OF A FEAST:
HUMAN COLONIZATION OF THE SOUTHERN BAHAMIAN ARCHIPELAGO
AND ITS EFFECTS ON THE INDIGENOUS FAUNA















By

LISABETH A. CARLSON


A DISSERTATION PRESENTED TO THE GRADUATE SCHOOL
OF THE UNIVERSITY OF FLORIDA IN PARTIAL FULFILLMENT
OF THE REQUIREMENTS FOR THE DEGREE OF
DOCTOR OF PHILOSOPHY

UNIVERSITY OF FLORIDA


1999

































Copyright 1999

By

Lisabeth A. Carison














ACKNOWLEDGMENTS


Although I have always been the kind of person who is reluctant to

accept help from others, this process is one that you cannot, and would not

want to, do alone. I have a great many people to thank. I would never have

made it through graduate school if I had not been introduced to Bill Keegan.

He gave me a place to work in such receptive surroundings that I couldn't help

but flourish. In this field, I am what I am because he is what he is.

Somehow, through sheer good fortune, I got to do research during the

past seven years in the Turks and Caicos Islands. The alliance between the

University of Florida and the Turks and Caicos National Museum was made

possible through Grethe Seim, who has been working in association with Bill

Keegan for twenty years. Grethe has been a tireless digger over the years, as

well as providing financial support, logistical assistance, a gracious hospitality,

and an unrelenting interest in the history and preservation of her beloved

Grand Turk. Brian Riggs, of the Turks and Caicos National Museum, has

helped me in so many ways and on so many occasions that it is impossible to

relate. He has made invaluable contributions to this work through his vast

knowledge of the Turks and Caicos, his personal connections throughout the

islands, his ability to get any problem solved, and his enduring friendship.

This project involved multiple years of fieldwork and depended on the

assistance of numerous people. Special thanks go to my field assistants.









Kimberly Martin worked much too hard that first year and we are still in her

debt. Barbara Toomey, my ally and kindred searcher, has over the years

always been willing to help with anything, and I mean anything, that needed to

be done. Sofia Marquet is the best field companion, co-worker and colleague

anyone could ask for. I am indebted to many other researchers who brought

their talents and specializations to the archaeological work on Grand Turk

including Mary Collins, Ann Cordell, John DeBry, Aline Gubrium, Reid

Hardman, Peter Harris, David Heuberger, Elise LeCompte, Lee Newsom, Iriv

Quitmyer, Sylvia Scudder, and Corbett Torrence.

Many volunteer hours went into the excavation and lab analysis portion

of this study. I am grateful for the hard work of 90 Earthwatch volunteers. I

would never have been able to organize and retrieve all the data recovered by

this army of laborers if I had not been taught so well by Kathy Deagan. There

are a few individuals who have generously contributed their time, skills, energy

and interest to multiple projects. I am indebted to each of them for their

enduring field support, especially Reed and Barbara Toomey, Robert Hoffman,

Jean Borchardt, Bob Gezon, Phyllis Kolianos, Ralph and Mary Lou Pax, Ben

Castricone and Patti Yamane and Lorie, Dan, Lindsey and Caroline Keegan.

Major funding for the project has been provided by the Wenner-Gren

Foundation for Anthropological Research, Earthwatch-Center for Field

Research, the Goggin-Fairbanks Foundation, and the Florida Museum of

Natural History. The Turks and Caicos National Museum, Grethe Seim, Reed

and Barbara Toomey and Andrew Newlands and William McCollum of Coralie

Properties, Ltd. on Grand Turk, all made donations toward this research.









Coralie Properties, the Turks and Caicos National Museum and the Turks and

Caicos Ministry of Natural Resources and the Environment each played a role

in the preservation of this archaeological site and in the safe-guarding of this

property against future development.

A great advantage in working in the Florida Museum of Natural History

(FLMNH) at the University of Florida is the day to day presence of experts in all

fields of natural science. The following people have graciously shared with me

their knowledge and their comparative specimen collections-Gary Morgan and

Erica Simons in Vertebrate Paleontology, Alan Bolten and Karen Bjomrndal of the

Archie Carr Center for Sea Turtle Research, Perron Ross, David Auth and

especially Dick Franz in Herpetology, George Burgess in Ichthyology, Laurie

Wilkins in Mamimology, David Steadman in Ornithology and Kurt Auffenberg in

Malacology. I am especially grateful to Walter Auffenberg, who was the first

person to look at and take an interest in my tortoise bones. I greatly

appreciate the guidance of all the people who work in the Zooarchaeology

range, especially Liz Wing, Sylvia Scudder, Susan deFrance, Laura Kozuck and

Irv Quitmyer. Most important, however, have been the members of my

committee, Bill Keegan, Bill Marquardt, Lynette Norr, David Steadman and Liz

Wing who have all been extremely supportive and helpful in guiding the

progress of different aspects of this work.

In the preparation of this document much needed computer assistance

was kindly provided by Scott Mitchell and by Abe Christian and Bill Paine of

the FLMNH Office of Museum Technology. Base maps for some of the figures

within were prepared by Matt Allen and Corbett Torrence. Marianne Franklin









provided the little book of conversions and frequently even beach front

property. I am most grateful to my friend, Anne Stokes, for working out with

me all these years and for doing everything in the program first, so I could see

how it was supposed to be done. A few friends on Grand Turk have made all

my time spent there memorable. My gratitude goes out to David Bowen for the

yoga, Dave Warren for the music, and Bob Gascoine and Jane Minty for

sharing their boat and sheltered comer of North Creek.

I have shared these last few years most closely with Blair, Gary, Kelley,

Kim, Marianne and Bob. They have each contributed to this work by

replenishing me with companionship, canoe trips, beach days, pool games,

music-making, rum drinks and complicated meals. My sister and Dev

provided food for my soul. My parents have supported me unendingly

throughout my extended education, but soon we will get to talk about

something else besides this paper. Ultimately, my greatest appreciation goes to

Bob who has been there through it all. Our good fortune together continues to

confound and comfort me.














TABLE OF CONTENTS

page

ACKNOWLEDGMENTS................................................................................ iii

LIST OF TABLES.......................................................................................... x

LIST OF FIGURES.......................................................................................xii

ABSTRACT................................................................................................ xiv

CHAPTERS

1 INTRODUCTION.................................................................................... 1

2 PREVIOUS RESEARCH........................................................................... 6

Who are the Taino?................................................................................ 6
Prehistoric Southern Bahamian Archaeology............................................ 9
Relevant Issues in West Indian Prehistory............................................... 11
Aceramic Period................................................................................. 12
Ceramic Period.................................................................................. 13
Why Study Islands?............................................................................... 15

3 THE PHYSICAL LANDSCAPE................................................................... 19

Geography of the Bahamian Archipelago................................................. 19
Sea Level Change................................................................................... 23
Climate................................................................................................. 26
Soils...................................................................................................... 29
Flora and Fauna.................................................................................... 30

4 THE PROJECT: STRATEGIES AND METHODS......................................... 34

Team Composition................................................................................. 34
Excavation Chronology........................................................................... 35
Field Procedures.................................................................................... 40
Lab Procedures...................................................................................... 43
Methods of Faunal Identification and Diet Reconstruction........................ 47
Temporal Controls................................................................................. 51









5 RESULTS OF FAUNAL ANALYSIS: VERTEBRATE AND INVERTEBRATE
REMAINS................................................... ............................................ 58

Diet Reconstruction............................................................................... 59
Green Turtle ...................................................................................... 59
Species Identified and Quantified........................................................ 61
Other Reptiles......................................................... .. ..................... 71
Fishes............................................................................................... 73
Invertebrates................................................... ................................. 77
Biogeographically Important Species....................................................... 81
Tortoise ............................................................................................ 82
Birds............................................... .................................................. 85
Contexts of Faunal Remains................................................................... 88
Middens and Turtle Roasting Hearths.................................................. 90
Contexts of Bird, Iguana, and Tortoise Remains................................... 97

6 RESOURCE USE................................................................................. 102

Horticultural Practices.......................................................................... 102
Environmental Consequences........................................................... 103
Material Evidence ........................................................................... 105
Hunting and Fishing Practices.............................................................. 111
Material Evidence............................................................................. 111
Inferred Evidence............................................................................. 117
Location of Turtle Harvests............................................................... 127

7 DIETARY COMPARISONS THROUGH TIME AND SPACE......................... 132

Temporal Changes in Fauna at the Coralie Site..................................... 132
Terrestrial Resources........................................................................ 133
Marine Resources............................................................................. 140
Comparison of Coralie Results with Other Regional Sites........................ 143
Diet at a Later Grand Turk Site......................................................... 145
Diet at Caicos Island Sites................................................................ 146
Diet at Other Bahamian Sites............................................................ 154
Overexploitation and Extinction............................................................ 158
Explanations of Diet Change............................................................. 158
Zooarchaeological Evidence of Overexploitation.................................. 161
Extinctions in the West Indies........................................................... 164

8 ISSUES OF COLONIZATION AND THE SETTLEMENT OF GRAND TURK... 171

Principles Governing Colonization Decisions.......................................... 172
Settlement Theory............................................................................. 174
Examples of Colonization Progressions from Other Regions................. 176
Population Movement in the West Indies............................................... 179
Saladoid Migration........................................................................... 180


viii









Ostionoid Migration .......................................................................... 185
Explanations for the Ostionan Expansion .......................................... 189
Faunal Remains at Other Ostionan Period Sites..................................... 195
Puerto Rican Data............................................................................ 196
Haitian and Jamaican Data.............................................................. 198
Settlement of Grand Turk..................................................................... 201
Settlement Location and Size............................................................ 202
Nature of Settlement......................................................................... 207
End of Occupation............................................................................ 212

9 CONCLUSIONS.................................................................................... 215

APPENDICES

A SYSTEMATIC ACCOUNTS FOR IDENTIFIABLE FISH REMAINS............... 220

B SYSTEMATIC ACCOUNTS FOR IDENTIFIABLE BIRD REMAINS.............. 228

C DESCRIPTION AND DISCUSSION OF GRAND TURK TORTOISE
REMAINS............................................................................................ 233

REFERENCES......................................................................................... 246

BIOGRAPHICAL SKETCH......................................................................... 279














LIST OF TABLES


Table page

1. Radiocarbon Chronology for the Coralie Site........................................... 52

2. Vertebrate and Invertebrates Species List from the Coralie Site, Grand
Turk, with Scientific and Common Names................................................. 63

3. Quantified Totals on Vertebrate Fauna from the Coralie Site.................... 67

4. Quantified Totals on Invertebrate Fauna from the Coralie Site.................. 69

5. Fish Size Estimates from Vertebrae of Identified Species.......................... 75

6. Ranking of Fishes at Coralie by MNI and Estimated Meat Weight..............76

7. Fine Mesh Samples of Fauna from the Coralie Site.................................. 78

8. Comparison of % of each Green Turtle Element from the Coralie Site
to % of each Element in a Complete Turtle Skeleton.................................. 93

9. Comparison of % of each Rock Iguana Element from the Coralie
Site to % of each Element Category in a Complete Iguana Skeleton............ 98

10. Habitats of the Marine Species found at Coralie.................................. 119

11. Feeding Practices of the Marine Species found at Coralie..................... 122

12. Possible Fish Procurement Technologies used at Coralie...................... 125

13. Bird Distributions through Time at the Coralie Site.............................. 134

14. Modem Distributions of Avifauna Identified at Coralie......................... 137

15. Faunal Remains Comparing Nine Units of Material from an Early
Context to Nine Units of Material from a Late Context.............................. 138

16. Radiocarbon Dates for Turks and Caicos Sites.................................... 144









17. Ostionan Period Sites with Completed Faunal Studies......................... 196

18. Amerindian Names of Islands in the Southeastern Bahamas and
Turks and Caicos Islands...................................................................... 202














LIST OF FIGURES


Figure page

1. Bahamas and Turks and Caicos Islands.................................................. 2

2. Turks and Caicos Islands....................................................................... 20

3. Grand Turk with Archaeological Site Locations........................................ 22

4. Sea Level Fluctuations Recorded for Florida............................................ 24

5. Northern Tip of Grand Turk Showing Site Location.................................. 38

6. Amerindian Canoe Paddles from the Bahamian Archipelago.................... 39

7. Excavation Units at the Coralie Site....................................................... 41

8. Locations and Time Periods of Radiocarbon Samples............................... 53

9. Profile Map of East/West Trench........................................................... 55

10. Photograph of Two Fire Pits (Fl 1 and F6) from Different Periods
of Occupation......................................................................................... 56

11. Percentage of Total Biomass Provided by the Primary Vertebrate and
Invertebrate Rem ains............................................................................... 61

12. Percentage of Total MNI Provided by the Primary Vertebrate Species....... 66

13. Features in the Central Section of the Coralie Site.................................. 89

14. Drawing and Photograph of Turtle Roasting Hearth (F25)....................... 92

15. Tortoise Plastron in situ at the Coralie Site.......................................... 100

16. Queen Conch (Strombus gigas) Tool.................................................... 106

17. Im ported Stone Ax............................................................................. 107

18. West Indian Topsnail (Cittariumpica) Shell Tool.................................. 109










19. Sea Turtle Pleurals with Harpoon-produced Fractures......................... 112

20. Fishing Implements from tie A Rat, Haiti.............................................. 116

21. Decline in Number of Boobies through Time at Coralie......................... 135

22. Decline in Number of Green Turtles through Time at Coralie................. 141

23. Caicos Archaeological Sites with Completed Faunal Studies................. 147

24. Bahamian Archaeological Sites with Completed Faunal Studies............ 154

25. Shallow Banks North of Hispaniola..................................................... 187

26. Post-mold Pattern Indicating Structure at Coralie................................ 204

27. Windscreen Post with Buttress Post.................................................... 205

28. Drawing of Oyster Shell Pendant and Conch Shell Bead....................... 210

29. Drawing of Plastron of Grand Turk Tortoise (Geochelone sp.) showing
Scute Pattern........................................................................................ 235

30. Photograph of Interior of Tortoise Plastron in situ................................ 236














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

AFTERMATH OF A FEAST:
HUMAN COLONIZATION OF THE SOUTHERN BAHAMIAN ARCHIPELAGO
AND ITS EFFECTS ON THE INDIGENOUS FAUNA

By

Lisabeth A. Carlson

December 1999


Chairman: William F. Keegan
Major Department: Department of Anthropology

This dissertation explores how a small island in the Bahamian

archipelago was initially colonized by humans and documents the

consequences to its animal resources. Archaeological investigations at the site

of Coralie on Grand Turk in the Turks and Caicos Islands have allowed a re-

examination of the relationship between humans and their environment at the

all important earliest stage of island settlement. The analysis of faunal remains

from different periods in an island's history provides data on shifting

subsistence strategies. Most difficult and vital to ascertain are the resources

available at the moment of initial human colonization. The Coralie site is the

oldest excavated site in the Bahamian archipelago with radiocarbon dates from

the first half of the 8th century A.D. It is also the colonization site for this

island. Its zooarchaeological material provides the first record for human









utilization of a pristine island fauna, never before subject to human predation.

Finds include species unknown to the West Indies and species that are locally

extinct. The remains are dramatically different from those in other West Indian

archaeological sites.

In addition to tracing changes in subsistence practices over time in these

islands, this dissertation looks at how the available resources, and small island

life in general, affected human decision making regarding migration,

colonization, and the establishment of settlements in this region. The events

on Grand Turk are tied to a period of extensive population movements in the

West Indies, which occurred at the transition between the Saladoid and

Ostionoid periods. The nature of this island and its history of occupation leads

into a theoretical discussion of why and how populations moved out of Puerto

Rico during the 6th century A.D., and why Grand Turk would have been a

desirable place to settle at that time. The results of this study clearly reveal the

role resources have in affecting human actions and, concurrently, the role of

humans in altering their landscapes.














CHAPTER 1
INTRODUCTION


When an oceanic island is colonized by humans for the first time, the

faunal composition of that island begins the process of being irrevocably and

dramatically altered. Plentiful animal resources, which can make an island

attractive to settlers initially, become increasingly rare through

overexploitation, and eventually human survival itself in these isolated

environments is threatened. This degenerative process has been charted

through the identification of faunal remains from Taino Amerindian sites in the

Turks and Caicos Islands, British West Indies (Figure 1), the southernmost set

of islands in the Bahamian archipelago.

The site of Coralie (GT-3), on the island of Grand Turk, is the earliest

excavated settlement in the southern Bahamian archipelago, with radiocarbon

dates from the first half of the 8th century. It is also the initial colonization site

for the island, containing evidence of the first contact between human

horticulturists and an oceanic island fauna that had been evolving undisturbed

for many thousands of years. Though this island is small, dry, and today

nearly barren, in the past it was densely populated with an array of terrestrial

species that complemented the plentiful marine resources of its extensive

banks. Humans harvested isolated, and therefore pristine, populations of

birds, reptiles, fishes, marine mammals, and invertebrates. The consequence

































Figure 1: Bahamas and Turks and Caicos Islands.


of this meeting for many of the terrestrial species was extinction.

Zooarchaeological remains from Coralie and other Turks and Caicos sites aid in

deciphering the timing of these extinctions, pinpointing some to the pre-

Columbian period. In addition, excavations at later sites in this region show

how, with the animal life diminished, the Amerindians were forced to develop

new technologies to increase the exploitation potential of certain species and

habitats, in order to sustain themselves and the resources they had left.

People have inhabited the West Indies since 6000 B.P., but no

populations had ever ventured into the Bahamian archipelago until the time of

the "Ostionoid expansion," at A.D. 650. This was a period of sudden

migrations of people out of Puerto Rico and into the remaining islands of the









Greater Antilles and the Bahamian archipelago. Cuba and Hispaniola had

sustained ceramic populations for thousands of years, but the Bahamas had

not. The settlers of Grand Turk found an island that had never been exposed

to human activity. Because of this, the resources encountered by the

colonizers were initially exceptionally abundant and very different from other

Antillean islands. Few archaeological sites from the Early Ostionan time period

have been excavated outside of Puerto Rico. This investigation provides

insights into Amerindian lifeways during this dynamic, and little known, period

in West Indian prehistory.

This is a biological study, but it is also an anthropological study because

it investigates insular human behavior. Equal time is spent understanding the

human population that first lived in the Turks and Caicos Islands and the

animal populations that also lived here before, during, but not in all cases,

after the Taino.

The following two chapters provide the historical and environmental

context for this study. West Indian prehistoric terminology is introduced and a

brief history of research in the Turks and Caicos Islands is outlined. The

presentation of previous research in West Indian archaeology exploring diet

and migration issues allows this current research to be put in a broader

regional and theoretical context. How the environment of the Turks and Caicos

Islands has changed over time is presented to inform discussions of habitat

use and possible environmental reasons for population migration. Chapter 4

details the project, the strategies for investigation, and the methods used to

reconstruct changing faunal distributions. Chapter 5 presents the Amerindian









diet at the Coralie site through the identification of animal species and by

exploring the contexts in which the remains were found. This is followed by an

extended discussion of the zooarchaeological remains in this and other regional

sites (Chapters 6 and 7). Interpretations of habitat use, procurement

technologies, and past animal population densities and distributions are

presented. Temporal differences in the results are deciphered, first within the

sediments of Grand Turk, and second in comparison to later sites in the Turks

and Caicos Islands and Bahamas, in order to show how overexploitation had

depleted the available subsistence base by the later time period. The issue of

West Indian faunal extinctions, and the role humans may have played in the

resulting situation is explored.

The final chapter of this dissertation discusses colonization issues in the

West Indies, centering on how this first human settlement of the southern

Bahamian archipelago was founded. The Coralie site is compared with other

Early Ostionoid period sites to show how different the fauna was on an

unexploited island, and to reveal how this might have contributed to

colonization decisions. Lastly, I discuss the nature of this settlement, the

social implications of this isolated island life, and how and why this settlement

came to an end. Although this analysis is limited to a single occupational site

on a small, isolated island, the findings have implications to more far reaching

issues in Caribbean prehistory.

There are no other sites in the West Indies where archaeological deposits

have captured such a complete picture of an animal population unadulterated

by previous human predation. We see what used to be available in these





5


environments, and come to realize that in all situations where humans endure,

they do so at great cost to the diversity of animal life, which in turn threatens

the stability of human settlements in these fragile island habitats.














CHAPTER 2
PREVIOUS RESEARCH


A century of archaeological investigations have been conducted

throughout the West Indies and in the Turks and Caicos Islands in an attempt

to learn more about the Taino culture. The Amerindians who met Columbus in

1492 were vividly portrayed in historic chronicles of the period. At the time,

however, their complex culture was in a state of rapid disintegration.

Archaeological excavations are the best vehicle available to decipher the way of

life of these island societies before European contact, before this culture

disappeared. A brief review is provided here to introduce the prehistoric

inhabitants of the West Indies and to present a timeline of the major cultural

events of this region. This current study will be put in the broader context of

West Indian prehistoric research. The reader will, concurrently, be led to

further research done in the Turks and Caicos, diet and migration studies from

around the West Indies, and similar cross-cultural studies centering on the

phenomenon of island life.


Who are the Taino?


Names for the Amerindians of the West Indies are only known for the

contact period. The inhabitants of the Bahamas were referred to as

"Lucayans," which in the native Arawakan language translated as "island men"









(Granberry 1991). The inhabitants of the Greater Antilles were called "Taino"

by the Spanish. The "Classic Taino" was the society described at contact who

inhabited Hispaniola, Puerto Rico and eastern Cuba. They were different from

the Tainos of Jamaica, central Cuba and the Virgin and Leeward Islands only

in that the Classic Taino had a greater elaboration in its material culture, and

appeared to be the political and economic center of this society (Rouse 1992).

The Classic Taino material culture and stratified social organization are found

at some sites in the Virgin and Leeward Islands and in the Turks and Caicos

region (Faber Morse 1997; Hofman 1993; Peterson and Crock 1999; Sullivan

1981). The Windward Islands of the Lesser Antilles were inhabited by a

different group called "Caribs" by the European colonizers. They are today

referred to as Island Caribs to distinguish them from mainland tribes. The

Island Caribs had the greatest longevity of all the Amerindians of the

Caribbean islands. During the 17th and 18th centuries, the French extensively

reported on their way of life (Breton 1665; Dutertre 1667; Labat 1742; LaBorde

1704; Rochefort 1665). Examples from Island Carib ethnohistory, as well as

Spanish chroniclers' descriptions of the Taino (Columbus 1959; Dunn and

Kelley 1989; FernAndez Mendez 1976; Las Casas 1951; Martyr 1970; Navarette

1825-37; Oviedo 1959; Pane 1974) will be used throughout this text for

comparative purposes.

Each of these Amerindian groups-the Taino, the Classic Taino, the

Lucayans, and the Island Caribs-are recognized archaeologically by a

corresponding material culture. For the West Indies, pottery has been the

primary means used to decipher separate societies (Rouse 1992). The









Lucayans of the Bahamian archipelago are recognized by a type of locally

manufactured pottery called "Palmetto ware," which had burnt, crushed conch

shell temper. This ceramic type is established throughout the Bahamian

archipelago by the 12th century A.D. There are three major styles of pottery in

the Greater Antilles in the Ostionoid period. The initial "Ostionan" sequence is

named after the Puerto Rican type site of Punta Ostiones (Rouse 1952). The

ware is a plain, thin, often red-slipped ceramic. This style was prominent in

the Greater Antilles between the years A.D. 600 and 800. The two other

sequential pottery styles that follow the Ostionan are the Meillacan between

A.D. 800 and 1200, and the Chican, from 1200 to contact. Though all three

are substyles of a single series (the "Ostionoid'), they do not all occur on every

island, and there is much overlap in time ranges for each style in different

regions (Keegan n.d.). This is evident in the investigations on Grand Turk

where Ostionan pottery continued very late, centuries after it ceased to be

made in Puerto Rico.

All the Ostionoid series pottery types are manufactured in the Greater

Antilles. They are considered trade wares when they occur in the Bahamian

archipelago (Cordell 1998). Palmetto ware has not been found on Grand Turk,

which based on the evidence to date, suggests that this island was never

occupied by the Lucayans. Ostionan style pottery is the only ceramic at the

Coralie site. This pottery came to this island with the colonizers and continued

to be imported for centuries to follow. Summing up, the site of Coralie is an

Early Taino colony which, through its ceramic remains, is classified as an

Ostionan Ostionoid site.









Prehistoric Southern Bahamian Archaeology


Archaeological investigations into the prehistory of the Bahamian

archipelago have been intermittent to date. Even though the Turks and Caicos

Islands form a small part of this archipelago, they have benefited from the

interest of a few exceptional researchers. Theodore DeBooy visited the Turks

and Caicos in 1912, on behalf of the Heye Museum in New York City, with

hopes of documenting Amerindian sites and collecting relics for the museum.

He completed the first archaeological survey here, locating 61 open air village

sites and cave sites from six of the major islands. Undisturbed Amerindian

artifacts were found in some of the caves, and items were purchased from local

residents. He reported on or acquired wooden idols, bowls, and dujos (seats),

ceramic vessels, stone celts or axes and hafted celts, shell ornaments, and

Amerindian petroglyphs (DeBooy 1912, 1913). His ceramic collections from

Grand Turk were of Meillacan style pottery only. Chican pottery has not been

found on Grand Turk as of yet. Few other investigations ensued in the Turks

and Caicos Islands or Bahamas until the 1970s (Granberry 1955).

In 1933-34, Froelich Rainey of Yale University conducted a survey of the

Bahamas, visiting the islands of Great Abaco, Eleuthera, Long, and Crooked,

among others (reported in Olson 1982). He located a total of 15 Amerindian

sites on 11 islands. The most important of these was the Gordon Hill Cave site

on Crooked Island. Of the seven caves Rainey visited on Crooked, one

contained two Lucayan burials and one, the Gordon Hill cave, had a human

occupational deposit. The site contained Palmetto ware, fishing implements,









and hutia and bird bones. This site is important for later biogeographic

considerations (see Chapter 7).

In 1978, Shawn Sullivan completed his PhD research in the Caicos

Islands focusing on the settlement of Middle Caicos and on modeling resource

use for the island (Sullivan 1981). He mapped, collected surface artifacts, and

did preliminary subsurface testing at the site of MC-6, which was hypothesized

to be a late period Classic Taino "outpost" for the acquisition of salt from a

nearby salina. Salt was a valuable Taino trade item. He also completed some

excavations at the earlier site of MC-12, which has been subsequently

destroyed by construction. Surveys of Middle Caicos to date have uncovered

39 sites. All the radiocarbon dates processed from Middle Caicos post-date

A.D. 1000, which is 300 years after Grand Turk was initially settled. In the

years preceding European contact, Middle Caicos was the center of the

Amerindian population in the Turks and Caicos Islands.

William Keegan also has worked in the Turks and Caicos Islands since

1978. He excavated sites on Pine Cay, Middle Caicos, and Grand Turk, and

surveyed nearly all the islands and cays. A focus of his research has been the

relationship between the Turks and Caicos Islands and Hispaniola, as

evidenced in trade, craft specialization, and population movements (Keegan

1992, 1997a). One issue that generated a great deal of research in these

islands was the question of the first landfall of Columbus (Sadler 1997). Grand

Turk was one of a few islands vying for the right to claim this position. In

addition to historical document research and navigational inquiries,

archaeologists sought the Amerindian villages described by Columbus. Sears









and Sullivan (1978) concluded that the very dry environment of the Turks bank

precluded Amerindian settlement. However, Keegan discovered the first

prehistoric settlement known on Grand Turk-the Governor's Beach site (GT-2).

GT-2 is a Meillacan site dating to the 13th century. Activity at this location

concentrated on the production of small, red beads from the cherry jewelbox

shell (Chama sarda) for export (Carlson 1993, 1995). Although this finding did

not strengthen the case for Grand Turk as Columbus' landfall, it did begin a

period of intensive archaeological research into the prehistory of Grand Turk.

Grand Turk was the site of a different type of first landfall, at least as

significant as the arrival of Columbus. It was the arrival of the first

Amerindian, the first person to settle in the southern Bahamian islands. This

current study of the Coralie site fills in the research gap for the earliest period

in the history of the Turks and Caicos Islands, explaining how and why these

islands were first settled, and the consequences of that settlement.


Relevant Issues in West Indian Prehistory


Certain basic facts regarding the prehistory of the West Indies, especially

which populations migrated where and when, must be presented in order to

understand the context of the settlement of Grand Turk. Many theoretical

issues surround the study of migrating populations. A brief history of the

questions pertinent to migration research in the West Indies will be presented

here. I will also introduce the kinds of questions that have been addressed

through the study of zooarchaeological remains from this region.









Aceramic Period

The West Indies were first settled about 6000 B.P. by people who entered

the region by crossing the Caribbean Sea between the Yucatan peninsula and

the western tip of Cuba (Rouse 1992; Wilson et al. 1998). These foragers are

recognized by their manufacture and use of flaked-stone tools (Pantel 1988). A

ground-stone producing culture developed 2000 years later, and may have

been the result of a second migration, this time from the South American

continent up through the Lesser Antilles to the larger islands. There are both

"Lithic" and "Archaic" sites, referring to the first and second migrations,

respectively, on the islands of Cuba (Dacal Moure and Rivero de la Calle 1996;

Osgood 1942; Rouse 1942), Hispaniola (Veloz Maggiolo and Ortega 1976),

Puerto Rico (Rouse and Alegria 1990), the Virgin Islands (Lundberg 1991), and

some of the Lesser Antilles, particularly Antigua (Nodine 1987, 1990). To date,

no ceramic sites have been identified on either Jamaica or the Bahamian

archipelago.

The West Indian Lithic sites exhibit a subsistence strategy of mangrove

zone gathering along with the hunting of medium-sized terrestrial fauna (Veloz

Maggiolo and Vega 1982). The Archaic sites show a diet of primarily marine

gathering (Davis 1982, 1988; Narganes Storde 1991). The extent to which all

these ceramic sites (both Lithic and Archaic) represent different cultural

groups (Rouse 1992; Veloz Maggiolo and Vega 1982), or differing, perhaps

seasonal, procurement strategies among a cohesive culture (Lundberg 1989,

1991) is still unclear.









The Archaic period populations introduced a number of fruit tree species

and edible seed species to the region (Newsom 1993), evidently practicing a

form of arboriculture and garden tending. The species represented originated

from either Mexico or Central America and South America, providing evidence

for multiple migration homelands. The first ceramic period migration, the

"Saladoid," originated from the Orinoco River basin area of South America.


Ceramic Period

The ceramic period history of the West Indies is characterized by two

major population movements. The Saladoid period is named for the

Venezuelan type site of Saladero (Rouse 1992), where the prototypical white-

on-red painted ceramic style was first encountered. The Saladoid series

describes the movement of a distinctive pottery style into the West Indies, but

the term Saladoid has come to refer to an entire culture and its people. These

ceramic bearing people first entered the oceanic islands of the Lesser Antilles

about 500 B.C., quickly settling on islands as far northwest as Puerto Rico.

The progression abruptly ceased and the Saladoid populations developed in

place for the next 1000 years. The next wave of migration began about A.D.

650, and this Ostionoid period (called Post-Saladoid in the Lesser Antilles)

continues up until European contact. The Ostionoid people expanded into

Jamaica, Hispaniola, Cuba, and the Bahamas.

In discussions of population movements throughout the West Indies, the

primary inquiry has revolved around the instigation of these migrations.

Population pressure, warfare, the search for new resources, increasing socio-









political complexity, ecological decline, and technological innovations have all

been proposed as the impetus for these migrations (Curet 1996; Keegan 1989;

Roe 1989; Siegel 1991, 1996). The ethnicity of the migrating populations is

still under question. Some scholars propose that multiple migrations occurred

during the Saladoid period (Chanlatte Baik 1981; Haviser 1997; Rodriguez

1989; Roe 1989), while others find evidence for a single population movement

(Rouse 1989; Siegel 1991). Both scenarios will be discussed in greater detail in

Chapter 8.

Other debates have centered around the geographical origins of the

Saladoid people (Evans and Meggars 1968; Lathrap 1970; Rouse et al. 1985),

and the timing of their migration into the Caribbean (Rouse 1986; Sanoja and

Vargas 1983). Discussions later turned to the way in which these populations

adapted to their new, island environments, whether applying mainland

adaptations to island life with little modification (Keegan 1985; Roe 1989), or

approaching the restrictions of islands with adaptive flexibility and

opportunistic behavior (Siegel 1991; Watters and Rouse 1989).

Much of the dietary research from the West Indies has focused on the

Saladoid period populations. Zooarchaeological studies have been concerned

with similarities in subsistence between the Saladoid and their ancestors in the

Orinoco basin (Petersen 1997), the amount of marine foods in early Saladoid

diets (deFrance 1989; deFrance et al. 1996), how settlement locations reflected

inland or marine subsistence orientations (Haviser 1997; Siegel 1992), and

what caused the sudden shift in diet between the Saladoid and Ostionoid

periods (Carbone 1980; Goodwin 1979; Jones 1985; Keegan 1989; Rainey









1940). More generally, faunal data have been used to answer questions

regarding habitat use, resource procurement, animal introductions, and

settlement exploitation areas (Keegan 1986; Sullivan 1981; Wing 1993; Wing

and Reitz 1982; Wing and Scudder 1983). Issues such as human-caused

faunal extinctions and habitat alterations have been dealt with, more from a

paleontological point of view to date (Morgan and Woods 1986; Pregill and

Olson 1981), but data from archaeological excavations are starting to fill the

gap between the pre-human and the historic contexts (Pregill et al. 1994;

Steadman et al. 1984). The Early Ostionoid settlement of Grand Turk, with its

unadulterated biological component, has provided zooarchaeological data from

which to address many of these aforementioned questions.


Why Study Islands?


When studying the prehistoric peoples of the West Indies, the

background for all the research is the unique environment in which they lived,

namely, islands. For much of the West Indies, especially the southern

Bahamian archipelago, those islands are very small and limited in their

terrestrial resources. Animals and humans that are adapted to island life are

different from those on the mainland. Because of this, certain research issues

are particular to island studies, regardless of which set of islands and which

group of islanders. Investigations with similar theoretical underpinnings to

this Grand Turk study, can come from areas very distant from the West Indies.

There is a large body of work from researchers focusing in the Pacific and the

Mediterranean on the mechanics of colonizations and the predictability of









settlement locations (Cherry 1990; Held 1993; Irwin 1992; Keegan and

Diamond 1987). Cross-cultural comparisons will be drawn.

Because this investigation deals with issues of changing animal

distributions, it is essentially a biogeographic study. Biogeography, according

to Davis (1988), is the "study of geographic distributions of species in historic

perspective, using ecological concepts about interspecific relationships." In

other words, it is the study of how organisms interact in time and space.

Issues that are encompassed by the study of biogeography include extinctions,

changes through time of species distributions, and in the case of island

biogeographical studies, colonization. Because of the time period of the first

settlement of Grand Turk and the nature of the faunal remains discovered

there, these three issues are of primary importance in the interpretation of the

early prehistory of this island.

The West Indies, with the exception of Trinidad and Tobago and the ABC

Islands (Aruba, Bonaire, and Curaqao), are true "oceanic islands" rather than

continental islands. Oceanic islands are "those surrounded by deep water

beyond the continental shelf, remaining separate from the continent even

during marine regressions of glacial age" (Martin 1984: 355-6). They are

geographically closed systems, bounded by the water that surrounds them,

making colonization and dispersal events rare. This isolation is one factor that

makes oceanic islands such great laboratories for scientific study (Patton 1996;

Rosenzweig 1995; Wallace 1880).

The most famous island researcher was Darwin (1839, 1859), whose

Galapagos experiences helped lead him to his thesis of independent evolution








after isolation. In 1967, MacArthur and Wilson published their theory of island

biogeography. Their attempts to isolate and quantify the variables related to

island colonization and subsequent evolution and extinction, initiated decades

of further work on island ecosystems (Brown 1971; Case and Cody 1983;

Diamond 1969; Lomolino 1984; Rosenzweig 1995; Simberloff and Wilson 1969,

1970).

Petersen and Crock (1999) suggest that certain archaeological research

questions are best answered through investigations on very small islands. This

is because environmental limitations such as available habitat and resources,

and island size, shape, and location, can be carefully isolated and controlled by

the researcher, thus elucidating which factors most affect populations.

Furthermore, the small island of an island magnifies changes by reducing the

number of variables, and provides a manageable data set on an entire

ecosystem.

There are other advantages in studying changing faunal distributions on

islands rather than on mainlands. Island faunas have a relatively small gene

pool, which permits rapid genetic changes to occur. Adaptations unique to

island life may include loss of flight, size changes (smaller or larger), loss of

wariness and protective coloring, and radiations into specific niches to lessen

competition (Case 1978; Diamond 1978; Livezey 1993; McNab 1994;

Schonewald-Cox et al. 1983). Because of these factors, there is a profusion of

endemic island species-ones that have evolved locally and inhabit only one

location and are, therefore, rare to begin with. These island animal

populations tend to be stable and enduring, though limited in distribution.








However, they are fragile in the sense that they have no defenses, and changes

to these populations can happen very quickly after human colonization occurs.

Many island species went extinct before ever being historically described, and

are known now only through their skeletal remains (Olson 1982; Olson and

James 1982; Pregill et al. 1994; Steadman 1993, 1995; Steadman et al. 1984).

Even though this dissertation is based on an archaeological excavation,

many of the same issues dealt with in biogeographical studies will be

investigated here. By tracing changes in animal distributions on Grand Turk

and through the Turks and Caicos, human-caused faunal extinctions can be

pinpointed. This study explores how the Turks and Caicos were colonized, first

by certain animal populations and then by humans, documenting the

consequences to this environment and the effects of this environment's decay

on its human inhabitants. Building on the work of previous researchers in the

Bahamian islands, and particularly the Turks and Caicos, this study presents

another perspective on the prehistory of these islands.














CHAPTER 3
THE PHYSICAL LANDSCAPE


The relationship between people and their physical environment is

central to this investigation. In order to understand how each affected the

other, we must first discuss the geography of this region. Environmental

studies look regionally and locally at soils, hydrology, sea level and climate

changes, and present and past flora and fauna in order to elucidate an island's

past ecology (Dincauze 1987). The second stage of these inquiries is to study

the impact human habitation has on island ecosystems through sediment cores

and paleontological and zooarchaeological investigations. By reconstructing

the environmental history of this region, and especially Grand Turk, the causes

of the changes seen can be pinpointed and we can begin to assess the role of

humans in altering the landscape in which we live.


Geography of the Bahamian Archipelago


The British colony of the Turks and Caicos Islands is the southernmost

cluster of islands in the geographical Bahamas, located 145 km north of

Hispaniola. Though it is politically separate from the Bahamas, the Turks and

Caicos are biologically and geologically part of the Bahamian archipelago. The

Turks and Caicos Islands are made up of over 40 islands and cays on two,











North Caicos

Provideca~cla Cay^, JL Mddle Calicsc ^












OMaadTuT.
Figurtweo Caico sa c


G d Tk Grand Turk






0 10 20














(Sealey 1985), which includes part of northern Cuba and the southern tip of


the peninsula of Florida. It began forming 150 million years ago (MYA),

through a combination of sedimentation in shallow water (about 1 cm in 500

years) and subsidence, and has a stationary geologic history. The Greater and

Lesser Antilles, rather, are a complex and dynamic mix of volcanism, plate

tectonics and carbonate sedimentation (Donnelly 1989). Less than 10% of the

Bahamian platform (11,400 km2) is above water today (Morgan 1989),

consisting mainly of shallow banks, with deep channels separating portions of

the platform. About 80 MYA, the southeastern bank began to break up into









many small banks divided by these deep water troughs (Schlager and Ginsburg

1981).

During the ice ages, changing sea levels alternately exposed and

submerged these banks. In the exposed cycles, easterly trade winds blew the

sediments into dunes, and higher, fossilized, sand dune ridges formed on the

windward side of the islands (Doran 1955). These windward ridges average 30

m in elevation with the highest point in the Bahamas being on a 63 m high Cat

Island ridge. The highest point on the Turks and Caicos bank is 48 m above

sea level (Flamingo Hill, East Caicos). The windward ridgeline on Grand Turk

peaks at 32 m above sea level (Bahamas Ministry of Education 1985).

The topography of Grand Turk, which corresponds to most southern

Bahamian islands, consists of a sandy leeward beach, a rocky windward coast,

inland salt water lagoons edged by mangrove vegetation, and inland salt flats,

or "salinas," which produce sea salt in the summer months. The dry-adapted

vegetation is denser on the eastern side of the island. Its terrain consists of a

consecutive series of diminishing dunes, which run perpendicular to the trade

wind direction, leading toward the leeward side of the island (Macpherson

1975). The trade winds blow continuously east-southeast at an average of 26

km per hour (Sealey 1985). The lithification process is rapid in this region. It

is not unusual to find Amerindian and historic period conch shell refuse buried

in the beachrock. The fact that the topography of these islands changes

quickly must be considered when assessing prehistoric settlement patterns.

The north half of this island is dominated by a large, inland lagoon

called North Creek, with the Coralie site (GT-3) located on its northwest shore,









about 1/2 km from the mouth (Figure 3). A few dune ridges to the west is the

Turks bank with its shallow coral reef environment extending 1/2 km from

shore. The bank edge then drops off to the Turks Island Passage, 36 km wide

and up to 2500 m deep, which separates the Turks bank from the more

extensive Caicos bank. The Coralie site location provided easy access to many

habitats including mangrove, inland lagoon, tidal flats, rocky shore, sea grass

beds, coral reefs, deep water, and inland scrublands, which could all be


GRAND TURK
0n .,2 w km


TURKS BANK


Figure 3: Grand Turk with Archaeological Site Locations.









exploited for their animal inhabitants. The only other Amerindian settlement

known on this island, GT-2, is located on the leeward shore of the southern

half of this island.


Sea Level Change


Because the islands of this archipelago are so flat, even a slight change

in sea level can radically alter the amount of exposed land. At the end of the

final glacial advance, approximately 18,000 years ago, sea level was 120 m

lower than today (Fairbanks 1989). Modem land formations can give clues to

fluctuating sea levels in the past. Blue holes are formed in dry, low sea level

cycles, with the depth of a blue hole equaling the amount of sea level drop

(Gascoyne et al. 1979). Another indicator of a lower sea level is submerged

beach rock slabs, such as are seen off the island of Bimini (Richards 1988).

Periods of higher than present sea level occurred during interglacials. The

entire Bahamian archipelago has been submerged multiple times in the past,

most recently at 65,000 B.P. (Bloom 1983; Imbrie et al. 1983; Pregill and Olson

1981). Physical evidence of elevated sea levels can be seen today in exposed

fossil corals and beach platforms now high and dry, and by uplifted coastal

notches once eroded by the action of the sea (Neumann and Moore 1975).

About 3800 B.P., corresponding with the beginning of the Archaic

Amerindian period in the West Indies, sea level was as much as 7 m lower than

today (Lighty et al. 1982; Tanner 1991; Watters et al. 1992). Because of this, it

is probable that many Aceramic period sites are currently inundated. The

waters rose steadily, reaching present day levels by A.D. 200 (Fairbanks 1989).










Small sea level fluctuations in recent times have influenced settlement

decisions at Coralie. Tanner (1991) reports broad-scale changes interpreted

from beach ridge data in the Gulf of Mexico, noting a sea level drop between

A.D. 450 and 750. The depth of the drop is not reported. In central Polynesia

at this same time, A.D. 500, sea levels were beginning to lower from a

sustained period of up to 1 m higher than present (Lepofsky et al. 1996). Work

in southwest Florida has produced some fine-scale results of sea level changes

for this period (Walker et al. 1995). There, sea level dropped from a peak at

A.D. 400 to a low of 50 cm below present at A.D. 600 (see Figure 4). These

figures are specific to Florida, and the exact levels of change are not known for

the Caribbean. In general, it is possible to say that sea level was high in the




120
> 100

~80
S
J 60
40\

|i 20

0- ------------ ---------
0 0
.w -20
-40

E -60
C
U r 0 0 0 0 0 0 0

Calendar Years



Figure 4: Sea Level Fluctuations Recorded for Florida.
Data from Walker et al. 1995.









Saladoid, then dropped substantially in the Early Ostionoid. The affects of

these varying sea levels and accompanying climate changes during the Late

Saladoid have yet to be truly measured and understood as a possible factor in

the disintegration of the Saladoid culture.

Excavations at the Coralie site provide evidence for a lower sea level

when Grand Turk was first occupied in A.D. 700. The deepest deposits in the

site are below the present-day water table, and some early period features and

artifacts were altered by exposure to water. At minimum, a 50 cm sea level

difference is necessary for this settlement to have been on dry land.

By A.D. 900, the center of activity at Coralie had moved away from North

Creek and atop the next dune to the west. Carbone states (1980) that sea level

began to rise again about A.D. 900, after some time spent below present levels,

and rose until stabilizing at A.D. 1200. This period, known as the "Medieval

optimum," was a time of relatively quick sea level rise. According to Tanner

(1991), the waters rose by 1.5 m. Recent investigations on Guadeloupe

(Delpuech et al. 1999) use evidence of coastal transformations to confirm a sea

level rise that began slowly sometime after A.D. 600 and accelerated another

1.8 m after A.D. 1000.

Further evidence pointing to a change in sea level, and changes in the

habitat surrounding the site comes from land snails. There are over 100

species of land snail in the West Indies (Emerson and Jacobson 1976) and

some are specific to certain environments. A fine mesh sample was taken near

the base (80 cmbs) of the earliest 8th century occupation, in a unit closest to

the mangrove edge of North Creek. It contained hundreds of very tiny land









snails of the species Truncatella pulchella (Beautiful Truncatella). These

creatures grow to a maximum length of 5 mm and would not have been

discovered but for the fine mesh sampling. Truncatella pulchella is not a

mangrove related species. It prefers shady, humid crevices above the high tide

mark, and usually lives in the rotted leaf matter below sea grape trees

(Emerson and Jacobson 1976). The provenience where this sample was taken

is saturated by the present-day water table. The presence of this small land

snail confirms that this provenience was once much drier. Sea grape trees are

a coastal plant usually occupying the frontal zone nearest the sea, sharing the

space with cocoplum, manchineel and sandburr (Correll and Correll 1982).

Because of the trade winds, these islands tend to grow, or migrate, toward the

leeshore. In the 8th century, the settlement may have been closer to the

leeward coast of the island, with the mangrove edge of North Creek farther to

the east. The site was certainly on higher and drier ground than it is today.


Climate


The transition between the Pleistocene and the Holocene, from 12,000 to

10,000 B.P., marked great ecological change. How much these changes

directly caused massive animal extinctions is an ongoing debate (Martin and

Klein 1984; Martin and Steadman 1999), and will be addressed in greater

detail in Chapter 7. The temperature of the oceans rose by four degrees

centigrade between 13,700 and 12,000 B.P., and sea levels began to rise (Lynts

and Judd 1971). The climate, which was dry at the beginning of this cycle,

experienced a wet phase between 7000 and 4000 B.P. (van der Hammen 1974).









Sediment cores taken in southwest Haiti (Hodell et al. 1991) show a

general return to aridity during the last 3,200 years, with the most extreme

aridity occurring in Late Saladoid times; a slightly wetter period occurred in the

Post-Saladoid times. These data are in contrast to the increasingly dry

conditions seen in the Yucatan Peninsula during the second half of the

millennium (Curtis et al. 1996; Hodell et al. 1995). Data from multiple sites in

this region showed that the dry interval continued from A.D. 280 through 1080

and pinpointed multiple drought events in that period. The first of several

major droughts identified was dated to A.D. 585 +\- 50 years. The authors

state that this was the "driest interval of the last 3500 years" (Curtis et al.

1996:45).

Both these studies indicate that the Late Saladoid period was

characterized by extremely arid conditions, but there is conflicting data on how

long this dry period lasted. Hodell et al. (1995) explain the 300 year difference

for peak drought conditions in these two regions by noting a possible dating

error in the Haiti study. This may be the case, but it is interesting that the

Mayan civilization collapsed during the peak of the drought in their region, and

the Saladoid culture collapsed 300 years earlier, also during possible peak

drought conditions. The timing of these arid cycles may not have been

regionally consistent. A study of sediment cores taken from mangrove habitats

on the island of Grenada (McAndrews 1996) identified a vegetation change

between A.D. 450 and 950, possibly indicating a dry interval at this time and

corroborating data from the Yucatan. Further work may resolve these issues.









Nevertheless, arid conditions during the Late Saladoid surely contributed to the

cultural changes of the time.

In regard to Grand Turk, it is nearly impossible to imagine a settlement

here over 1000 years ago if it were any drier than it is today. Currently, there

are no fresh water sources on the island. Average rainfall, with records dating

back to 1955, is 575 mm annually, but it can range between 300 and 1125

mm (Bahamas, Meteorological Office 1997). The rainfall comes in definite

seasons, with October to December providing 44% of the annual precipitation

and February to April only 13%. The driest month is March and the wettest is

November. There is a rainfall gradient between the northern and southern

islands of the Bahamas, with Abaco receiving ca. 1550 mm, annually-three

times that of Grand Turk (Little et al. 1977). Because of high temperatures,

low rainfall and constant, drying winds, the evapotranspiration rate today on

Grand Turk is 1500 to 1875 mm annually-three times greater than the

amount of rain received (Sealey 1985). A richer vegetative state in the past

could have somewhat mitigated this imbalance.

It is because of this evapotranspiration effect that plentiful, high quality

salt forms along the edges of interior salinas. For most of its recorded history,

Grand Turk's livelihood depended on the salt trade (Sadler 1997). Sullivan

(1981) speculated that salt collection was the reason for the settlement of the

largest and most elaborate site on Middle Caicos (MC-6). The role of salt in

prehistoric times on Grand Turk is difficult to ascertain from archaeological

remains, but it was surely a utilized and valued resource.









Soils


The islands in the Bahamian chain are formed from precipitating

calcium carbonate, corals or algae. Bahamian soils are sandy and stony

sediments with little humic content. The primary mineral constituents of these

soils are calcite, salt and trace amounts of aragonite. At Coralie, the soil

matrix is 98% sand, and 2% silt and clay with broken shell inclusions

(Hardman et al. 1998; Scudder 1997). Grain size of the sand is variable with

90% falling within the small or medium categories. The beach sands below the

site contained the coarsest grains. The buried anthropogenic horizon

contained evidence of the Amerindian settlement. Hardman et al. (1998:13)

concluded this horizon was an "essentially stable surface for an extended

period of time."

Mann (1986) has hypothesized that the source of all the non-calcareous

soil components in the Bahamas are a result of aolian deposited dust, blown

over from the deserts of North Africa. The locally manufactured Lucayan

pottery, Palmetto ware, was fashioned from these occasional deposits of red,

clayey soil called "Bahamas red loam." The poor quality of the clay deposits is

reflected in the friable, coarse nature of the locally made pottery. There are no

such clay deposits on Grand Turk.

The soils of these drier, southern islands are not rich, but nevertheless

are adequate for root crop agriculture, especially the cultivation of manioc, the

staple food of the Taino. On Grand Turk, planting of root crops was possible

on the elevated and well drained series of beach ridges. Even though the








matrix is nearly all sand, manioc could flourish in this environment as long as

it was planted in correct relationship with the infrequent rains (Lee 1980).


Flora and Fauna


The plant communities can be divided into two broad subdivisions-

v'whitelands' and 'blacklands' (Campbell 1978). Pioneering whiteland species

such as palmetto, seagrape, and the introduced Casuarina pine live in a soil of

lime and white sand. Other whiteland species include cacti, buttonwood, and

bay cedar. Coppice vegetation dominates the blacklands with species such as

pine, mahogany, mastic, ficus, and lignum vitae (Correll and Correll 1982).

Pine tree (Pinus caribaea) monocultures cover much of the northern Bahamas,

and the occasional southern island such as Pine Cay in the Turks and Caicos

Islands (Henry 1974). Pine Cay can support pine vegetation because of a

geologic anomaly that traps a fresh water lens below this island (Iverson 1979).

Before the Bahamian hardwood forests were cut by initial European settlers, a

variety of "fine madera, mahagony, cedar and pine...fit for building of vessels"

(Craton 1962:111) reportedly covered these islands. These harvested vegetative

communities were never able to reestablish themselves. Today, the flora of

Grand Turk is all low dry-adapted scrub species.

Through the identification of charcoal remains, the types of trees and

shrubs inhabiting Grand Turk over 1000 years ago can be reconstructed.

From the Coralie site, paleobotanist Lee Newsom identified wild lime (cf.

Zanthoxylum sp.), palm (Palmaceae), buttonwood (Conocarpus erectus), black

torch (Erithalisfruticosa), ironwood (cf. Krugiodendronferreum), and remains








from the bittersweet family (Celastraceae). All these taxa (especially palm and

wild lime) are associated with dry environments and are relatively fast growing

trees, which normally exhibit wide growth rings. These particular charcoal

samples showed variably sized growth rings, further suggesting dry or

occasional drought conditions. This is some corroborating evidence for the

proposition of a continued dry climate up to A.D. 1000 in these islands.

Today the most abundant tree near the site is black mangrove (Avicennia

germinans), a common Amerindian fuelwood (Newsom 1993). Buttonwood,

rather than black mangrove, appear in these samples and may indicate a less

swampy environment in this area in the past. The wild lime and bittersweet

family specimens identified at Coralie have also been identified at sites in the

Lesser Antilles and the Virgin Islands. Palms (Palmaceae) are one of the most

widespread plant families found in West Indian archaeological sites. Parts of

the palm are edible and these trees were probably tended from Saladoid times.

Ironwood appears in a West Indian archaeological site for the first time at

Coralie. A dyewood tree (Andira sp.) was identified from a historic context,

although it is not certain whether it is a native or introduced species. It does

not grow on Grand Turk today. These botanical samples, which came mainly

from early period deposits, provide no evidence to suggest that Grand Turk ever

sustained a heavily vegetated hardwood environment.

All of the various species of flora and fauna in the Bahamian archipelago

are Late Quaternary period colonizers (Pregill and Olson 1981). However, the

Greater Antilles have a long history of animal evolution beginning as early as

the Eocene period (55 MYA) (Morgan and Woods 1986). The fauna reached the









islands of the Caribbean via overwater dispersal, originating from South or

Central America (Darlington 1938; Simpson 1956). Because of adaptive

radiation after arrival, all of the various island species could have evolved from

very few colonization episodes (Morgan and Woods 1986). When the Great

Bahama bank was exposed during the last glaciation, the distance to the

Cuban landmass was only 12 km (Franz et al. 1996). This accounts for the

Cuban origin of much of the Bahamian flora and fauna (Buden 1981; Correll

and Correll 1982; Franz et al. 1996; Morgan 1989). Except for modem

introductions, North America has been the source of few animals to the nearby

Bahamas. Vast tracts of plant and animal habitat were lost in the Bahamas as

sea level rose at the Pleistocene/Holocene boundary (Pregill and Olson 1981).

Radical environmental changes such as these certainly effected the distribution

of Bahamian faunas over time.

The following environmental characteristics of the Turks and Caicos

Islands are important to keep in mind. The islands were formed of precipitated

limestone; there is no local source of metamorphic or igneous rock. All the

landmasses in the Turks and Caicos complex are low in elevation and small in

area; the largest, Middle Caicos, is only 120 km2. Because of this, terrestrial

habitats are not diverse. The climate is hot and very dry, the vegetation

stunted and drought resistant. The land itself here is fluid with dunes rising

anew in the west, ridges eroding in the east, and beach rock forming under

foot. Due to its lack of elevation, slight changes in sea level had the potential

to radically alter the amount of exposed land. Hurricane-grade storms, which

constantly endanger the region, could restructure the island overnight and





33


threaten populations of land animals with extinction. Fluctuations in rainfall

and temperature could compromise the ability of the island to support human

life, regardless of the richness of the food base, because fresh water sources

may disappear. Amerindian settlement on Grand Turk must have been a

tenuous proposition.














CHAPTER 4
THE PROJECT: STRATEGIES AND METHODS


This chapter provides a chronology of all the work done at the Coralie

site, both in the field and in the lab. Explanations are provided for how the

faunal assemblage was reconstructed and what methods were used to

document changes in that assemblage over time. Investigating the faunal

remains of this site was my primary contribution to this project. Other

researchers have completed work on separate aspects of this study (Cordell

1998; Gubrium 1998; Hardman et al. 1998; Harris 1996; Keegan 1997a,

1997b; Scudder 1997). Some of the broader goals of this extensive project will

be presented in this chapter.


Team Composition


Explorations at the Coralie site were instigated and led by William

Keegan, Curator of Caribbean Archaeology at the Florida Museum of Natural

History in Gainesville, Florida. His work was in collaboration with many

specialists and volunteers. Brian Riggs, the manager of the Turks and Caicos

National Museum on Grand Turk, provided a year-round presence on the

island and handled all the unexpected developments with the site. I oversaw

the excavations and the analysis of the remains. Lee Newsom from Southern

Illinois University at Carbondale did the archaeobotanical work at the site.









Several researchers from the Florida Museum of Natural History contributed

their expertise to specialized studies. Sylvia Scudder collected and analyzed

soil samples. Irvy Quitmyer examined land snails from the vicinity of the site.

Ann Cordell did petrographic analyses of the ceramic remains. Elise LeCompte

supervised an excavation of waterlogged sediments on the banks of North

Creek. Mary Collins, of the Soil and Water Sciences Department of the

University of Florida, undertook pedological investigations at the site. All the

work was accomplished, through the assistance of some 100 volunteers who

provided over 10,000 person hours of labor in the excavation phase of this

study alone.


Excavation Chronology


Because the site is deeply buried, it went unnoticed by surface surveys

and remained relatively undisturbed until its discovery in 1991. The property

where the site is located was one of many plots of land being sectioned off and

cleared for development in this tract by land owner Andrew Newlands. A

dense, low brush covers this entire area, so surface visibility was minimal. The

plot was being cleared by hand of this vegetation when Brian Riggs visited to

check for any indication of Amerindian sites. He noted surface finds of pottery

and burnt conch shell debris.

As the Turks and Caicos has legislation to protect antiquities such as

archaeological sites, William Keegan was asked to complete an archaeological

impact statement. In February, 1992, he and Barbara Toomey shovel tested

this area. A total of 35 test pits defined the sites' boundaries and determined









varying deposit densities within the settlement. Judging by the vast amount of

turtle bone found in these test pits, the initial reaction was that this site was a

sea turtle butchering area. Charcoal from one of these test pits produced what

was, at the time, the earliest radiocarbon date for the region-ca. A.D. 900

(1120 +/- 120 B.P., corrected and calibrated).

In 1994, a water pipe was laid east/west, bisecting the site. A small

trench was dug by hand about 20 cm wide and 50 cm deep. Again, Brian

Riggs was there to note what was disturbed. He bagged the artifacts found

with fairly precise horizontal and vertical provenience. This preliminary, make-

shift trench was used, in association with the test pit data, in deciding where to

plot the first excavation units when Keegan returned in January of 1995.

Multiple excavation seasons were carried out between 1995 and 1997 in

affiliation with the Earthwatch organization (Keegan 1996). During the month

of January, 1995, two teams of 15 volunteers each worked at both the Coralie

site and GT-2. The next two seasons were spent entirely at Coralie. In January

and February of 1996, three teams of 10 volunteers each worked for a total of

six weeks. The final year of 1997 had three teams again of 10 volunteers each

for six weeks in January and February. These three concentrated periods of

excavation provided the data for this faunal study. Soon after the significance

of this site was determined, the Turks and Caicos government acquired the

land from the developer and is currently holding it in trust as a site of national

historical importance.

A few remote sensing survey techniques were utilized to help determine

the layout of the site and where units should be placed. In May of 1995,









Keegan spent one week completing an electromagnetic survey of the site area.

He determined that this technology was not conducive to the nature of the soils

and would not be helpful in meeting his goals.

A second attempt at subsurface survey employed a ground penetrating

radar system. Mary Collins and two of her graduate students spent one week

in January 1996 completing this radar survey and doing soil studies in the

area of the site. The unconsolidated sand of this locale provided a good matrix

for sending the radar signal, and features within the site did show up on the

visual printouts. Our excavations were used to test the effectiveness of this

technique. Unfortunately, the water table itself provided the strongest radar

signal and detecting other inclusions in the sands proved difficult.

Figure 5 depicts the region surrounding the site. The settlement

boundaries are sketched and general geography is shown. This is a small site

running northeast/southwest for 160 m with a maximum width of 40 m. The

mangrove fringed edge of North Creek is 45 m from the eastern edge of the site

today. The ocean is 300 m to the west across a series of small dune ridges.

Also shown on this map is a small wooden dock on the North Creek shore. It

was 10 m east of this dock, in the peat sediments under North Creek, that the

most spectacular artifact associated with the Coralie site was found.

In 1996, Captain Bob Gascoine was deepening a channel between this

dock and his anchored live-aboard dive boat when he pulled a piece of wood

out of the muck below his feet. To his astonishment he recognized it to be a

complete, one-piece canoe paddle. What is more, it was a Taino artifact (Figure

6), clearly associated with the Coralie site. He contacted the Turks and Caicos





38
.. . . . . . . .. ,. . . .. _. _.. . . . -j ~ i

. . . . . . . . . ..i!ii ii ii ii iii. . . . .. .
. . . . ..o .o o . . . I . . . . ,


.. : : :::::::::: :::..::::::::::::::. :
...... :: ::: :::::::: :::::: .
i~ i .! ....................... .:. .:. .:.






......~ contou 'vra .. . ..mR N



Nationald Mueumand---:;y Bra Rigsimeitey ad inswremdet



conserv the1 fid Thewod as enatvel ienifid y ee Newsom as:
bu w ,ood (P, \ a a t o b ip o l, an m t....h e. n


t....... Grand Turk .owth ihbtn o.the Coralste. T
aio...... .......e and roid a cirediage rang a 2 ma) of
":... .. .. .: R E K .






















995-1125, lte find The wocupastion.Thisel sugestsfe tha peope sale between

HisanioaandTi pcisi o konfo the Turks and Caicos Islandsmutpeiesdrn the ay
. .. ...:::::::: .:::....~: .. .:: : :

. .. .. .. ..... .. . . . ..:: : :: : :: : :: : :: : :: :
.............. . . . ............ . .: .:. . .

6 m O .'~ ~.'..t .... .....' '.....












Fiue5:NrhrnTpo Grand Turk aln ihteihbtnshofin Ste Locationit. Tewo a
Nationalbouse.- and .r.i.. .i....ib ediatel ange lan were ma e o..






995-1125, lthe iund The wooup enation.y Thsdugentise bya Leope Nesaie beaee

Bhisanioas Thsseaensdo nonf the Turks and Caicos Islandsmutpeiesdrg tode y
..... ....l .a.aebe.cre.nHspnoadlotatrwrkn t a
.... :rn "lurk* .ln itteinaint fteCriese. .:-%NORTH:::`
radiocarbon~~~..... ....n rvde airtd g ag a : im) .D..
99-15,lt i h ocptin hs ugst ht epe6mld ewe
Hi...i... ............d..i.s.slnd.muti leti es..rngt.









































Figure 6: Amerindian Canoe Paddles from the Bahamian Archipelago.
(top) Scale drawing comparing Grand Turk paddle to one from Mores
Island, Bahamas (Reproduced from Sadler 1997:33). Mores
Island paddle sketch first reproduced in DeBooy 1913: Fig. 1.
Grand Turk paddle sketch courtesy of the Turks and Caicos
National Museum.
(bottom) Photograph of Grand Turk paddle. Courtesy of the Turks
and Caicos National Museum.


occupation. The paddle was carved from a single piece of wood. It is 135 cm

long. The pointed blade section is 85 cm long. It has a crosspiece at the top of


0cm 20 40 60 80 100 120 140




'
I I I I I I I I I I I I I I I




-_- ,D,... -z-i Z' ._Z Z -1 "- .,_.









the shaft. This paddle is nearly identical to one found on Mores Island (Little

Bahama Bank) in 1913 by DeBooy (1913: Fig. 1). Las Casas, a chronicler of

the contact period, described these paddles "like long handled bakers' shovels,

but sharp" (in Olazagasti 1997:133).

In response to this find, investigations were carried out in the sediments

below North Creek in the area where the paddle was discovered. Conditions

were apparently excellent for the preservation of wood, fiber, and other

perishable materials in these waterlogged strata. Elise LeCompte spent two

weeks in January of 1997 exploring the area but no other Amerindian period

artifacts were found.


Field Procedures


The Earthwatch teams excavated 271 m3 of sediment. Most units were 2

by 2 m blocks laid out in a single long trench that paralleled the creek, with

perpendicular extensions opened up east of this main trench (see Figure 7).

The average depth of each unit was 1 m, which coincided approximately with

the water table. Datum readings came from a single permanent transit station.

These numbers have been converted so that they describe a generalized below

ground surface measurement. This is possible because the site's surface is

nearly flat. The datum readings give an idea of the depth below surface as well

as an accurate across site system for recording relative depths. Though the

unit size may seem large, much of the material was mapped in situ as it was

uncovered; 126 field maps were made. Most units contained feature material

(either hearths or midden deposits), which was excavated, bagged, and









catalogued separately from the surrounding matrix. These techniques allowed

sufficient control over the horizontal provenience of the artifacts.

The site is buried by a sand layer, which grades from 25 to 65 cm deep

as it nears North Creek. This overburden (Zone 1) was mostly sterile but

contained some 19th and 20th century historic materials. At the turn of the





0.-
140 \O /o
lw-~ \ "^
N \ \ .

120- \
1m I."" 1 \ P





100- 0

00


Do 100 120
CASTING

Figure 7: Excavation Units at the Coralie Site.









century, this tract of land supported a coconut palm plantation for a short

time. The only surface indication of a historic occupation is a broken down

stone wall that cuts diagonally across the site toward the northwest. In some

of the units, historic remains do lie atop the Amerindian deposits. Excavations

identified a few intrusive features from the historic period-four historic post-

molds, one reinforced post-hole constructed from coral and rock, and two

examples of wood. The woods preserved because they went deep enough to

have been saturated from constant contact with the water table. One piece was

an old fence post, the other a tree root; both began at 35 cmbs (centimeters

below surface). Other than these examples, there was very little mixing of the

two contexts. Careful excavation mitigated most contamination. There was no

evidence of tilled soil or other activities that would destroy the integrity of the

prehistoric deposits.

The overburden was removed by shovel and screened. However, after it

was determined how consistently deep the Amerindian deposits were, and how

sterile this layer was, this material was discarded without further investigation.

Zone 2 is the dark, anthropogenic soil that comprises the site. It varies in

thickness between 10 and 40 cm. It is one uniform stratum, which was

removed carefully in 10 cm levels with trowel and brush. The soil was full of

fragile bone, so the excavations proceeded cautiously. All matrix was dry sifted

through 1/4" mesh, with 10-liter soil samples taken from features, representative

zones, and any areas with noticeably denser animal bone inclusions to be used

for fine mesh faunal analysis. It was determined at the beginning of these

excavations, through testing with various mesh sizes, that /4" mesh would









propel expedient excavations while at the same time recovering most of the

bone material in the site. Separate soil samples for analysis of macro and

microbotanical remains and mineralogical content were also collected.


Lab Procedures


Many of the field specimens were processed simultaneously on Grand

Turk while excavations continued. Shell remains were washed, sorted,

identified to species, and analyzed. Lab analysis procedures included

recording shell weights, counting the number of specimens for each species

(abbreviated NISP), and calculating an MNI (minimum number of individuals)

measurement, which estimates how many individuals of each species are

represented by the remains. This is done by counting a unique anatomical

constituent for each species. For example, for invertebrates with univalve

shells, the spire of the shell was counted. With clams and other bivalves, the

number of left and right hinges were counted, and the larger number was used

to calculate the MNI. However, simply counting the number of hinges and

dividing that number in half produced the same result. Measurements were

sometimes taken for use in allometric formulas, which can calculate individual

animal sizes. This technique is just beginning to be used for invertebrates in

the Caribbean area (Wing 1998). Any secondary use of the shells was noted

and shell tools were described in detail. All food shell debris was discarded

after analysis. The shell tools have been curated either with the Turks and

Caicos National Museum or the Florida Museum of Natural History. Other tool

forms were found made from coral and local and imported stone. These were









similarly analyzed and curated. Pottery and bone were transported to

Gainesville for analysis. Pottery is not abundant in this site, perhaps because

every vessel and griddle (which make up 10% of the collection) had to be

imported from the Greater Antilles. The total amount of pottery excavated was

? only 1423 sherds, weighing 63 kg. If pottery distributions were averaged

across the site, each 1 m3 would have produced only seven sherds.

Furthermore, 39% of these sherds were smaller than 2 cm across. Very little

could be done in terms of vessel shape reconstruction because of the high

breakage and eroded nature of this collection. The pottery from this site is

wholly unremarkable, yet typically Ostionan. Only 2% displayed any

decoration, not including red-slipping. The finely made redwaree" comprised

27% of the sample; 45% was plainware. The remaining sherds possessed an

exterior only "self slip" (Rice 1987).

Further results from the pottery analysis can be found in two studies by

researchers from the University of Florida. The entire collection of pottery was

incorporated into a Master's thesis on the Ostionan pottery style by Gubrium

(1998), who described the assemblage and confirmed its Ostionan associations.

Using thin-sectioned sherds from Coralie, Cordell (1998) identified three paste

types along with three surface treatment types, and made correlations between

them. She further traced these paste types to ones from Hispaniola as part of a

larger study to locate the source areas for imported pottery in this region.

The bone remains were not washed, rather, they were laid out to air dry

and brushed clean when necessary. Much of the initial sorting process was

completed in the field. Thousands of sea turtle shell fragments came from each









provenience. These were often crushed into very small pieces. Time was not

spent counting the fragments under 2 cm in diameter in proveniences with

ample turtle remains; these were only weighed. In calculating the NISP for

turtle, only the internal skeleton and some of the individual carapace and

plastron bones were counted. So this number is a gross underestimation

(perhaps 90% went uncounted) of the number of turtles bones in the site. No

attempt was made to estimate the number of uncounted turtle bone fragments

through sampling. Similarly, there were very large samples of burnt, broken

conch shell. Many proveniences of conch were laboriously counted and

weighed, so the total NISP for conch is an estimation calculated from sampling.

The NISP for fish remains is underestimated somewhat (less than 10%)

because some proveniences of unidentifiable fish fragments were not counted,

only weighed. These imprecise methods had no impact on the dietary

calculations where MNI is customarily used.

Every provenience had 100% recovery of bone in the 1/4" screen. All

these remains were used in the calculation of these results. In addition to this,

fine screen samples were analyzed to understand what portion of the remains,

and the diet, was missed in the 4" screens. The fine mesh samples came from

10 proveniences and totaled 106 liters of matrix. The soil was processed

through nested geologic screens of 4 mm, 2 mm, and 1 mm mesh to recover all

the smaller faunal remains. The 4 mm and 2 mm screen sizes provided nearly

all of the bone material. Very little was recovered in the 1 mm mesh.

In a sampling experiment from the site of Trants, Montserrat, Reitz

(1994) found that through the use of 2 mm mesh, the amount of bone









recovered in the sample doubled, although the number of individuals (MNI)

remained the same. This was also the result with the fine mesh samples from

Coralie. No new MNI was found but because there were so many mostly

unidentifiable bone scraps, the NISP was raised substantially. The 1/4" mesh

samples provided an accurate reflection of the deposited remains, which is not

the case at many later West Indian sites.

The bone sample recovered from Coralie is particularly large when

compared with other Caribbean sites. In order to compare faunal results

between sites, the samples should ideally be of similar size, and be sieved

through the same mesh diameter. For many studies in the Caribbean this level

of comparability does not exist. For a sample to provide a reasonably complete

picture of the faunal remains at a site, 200 MNI must be collected from a

minimum of 1400 bones (Wing and Brown 1979). Many sites fall short of this

requirement.

The largest biosamples have come from the extensive Saladoid and Post-

Saladoid sites of Maisabel, Puerto Rico (deFrance 1989), Sugar Factory Pier, St.

Kitts (Wing and Scudder 1980), Tutu, St. Thomas (Righter n.d.), and Golden

Rock, St. Eustatius (Versteeg and Schinkel 1992). Each has over 500 MNI and

over 10,000 NISP for the vertebrate remains alone. The Coralie sample had

963 MNI and over 30,000 NISP (not including all the broken turtle shell

fragments). This is comparable to the sites previously mentioned in sample

size, if not in overall site size. These four sites all had different sampling

strategies than those used at Coralie, but as long as the results truly represent

what was in the ground, the data sets can be compared with reasonably









accurate conclusions. Unfortunately there are no other comparatively large

faunal samples from sites in the Bahamian archipelago, or from any other

Ostionan period sites. What data are available will be discussed in Chapters 7

and 8.


Methods of Faunal Identification and Diet Reconstruction


The sorted, faunal materials were identified to the lowest taxon by

comparing elements in the sample with those contained in the zooarchaeology,

ornithology and herpetology comparative collections at the Florida Museum of

Natural History. MNI was counted for all taxa based on the most numerous

element, taking into consideration any differences in the size of paired elements

(Wing and Brown 1979). For example, all the right humeri of the sea turtles

were counted, then any individuals that did not fall into the size range dictated

by the humeri were added to the total. These included two hatchlings and two

very large turtles, larger than any of the recovered humeri. For iguanas, left

ilia were counted, considering only gross size differences. The estimate of the

number of individual tortoises was based on the number and the size of

plastra. Fish MNI was determined by counting the most common single

element for each taxon. This was most often an atlas or in the case of the

parrotfish, a pharyngeal grinding plate. It is very difficult to estimate MNI for

shark because of the nature of their remains, namely numerous teeth and

vertebrae. A very conservative estimate was made by looking at great size

differences in vertebrae and noting the amount of temporal or spatial distance

between specimens. Because the number of bird remains found were few,









usually only one or two of each taxon, location of the remains was primarily

used in assigning MNI. The most abundant element was counted in the more

populous bird species.

After determining the number of individuals of each taxon the next step

was to estimate how much meat these individuals contributed to the Taino diet.

There are multiple ways to do this. One is the use of allometric conversions

(Reitz and Cordier 1983; Wing and Brown 1979). Relevant measurements were

recorded on the well preserved elements from each species (Morales and

Rosenlund 1979 for fish; Driesch 1976 for bird and mammals) for use in this

type of formula. This technique worked best with the fish remains. By

measuring the anterior diameter of the centrum of the vertebrae, then inserting

that number in the following formula (Wing and Scudder 1980:204), the live

weight of a fish can be determined.


Logy =2.047 (Logx) + 1.162

(x=centrum diameter in millimeters)
(y=body weight of fish in grams)


When allometry was not possible, a "proportional method" was used

instead, where the size of individuals was estimated by likening the bones with

remains of individuals in a comparative skeletal collection. Finally, if no

method could provide an actual meat weight based on the bone size, then an

averaged species weight was used. This was the case with birds, invertebrates,

and most of the rock iguana remains. Even so, juvenile bones were taken into

consideration when estimating the total meat weight provided by each species.









All bone modifications were noted; this includes evidence of burning or

butchering. Recording the extent of epiphyseal fusion of the long bones

allowed an estimation of the age of death for some species. For the two species

represented by many individuals and thousands of bones, namely green turtle

and rock iguana, calculations on the recovered vs. expected number and

percentage of bones provided information on possible butchering techniques,

field processing evidence, and deposition practices.

Sometimes assumptions must be made when identifying species based

on comparisons with complete modem skeletons. This was the case in working

with so many sea turtle remains. Elements such as mandibles are very

diagnostic to species but others, such as carapace and plastron fragments, are

not. Those bones that were identifiable to species because of unique

anatomical structure, were all, with one exception, believed to be Chelonia

mydas, the green turtle. So, it was an extrapolation to say that nearly all of the

bones in the site are green turtle, but it is based on skeletal observations, and

the fact that green turtle is the preferred eating turtle among the world's

populations.

Being familiar with the current animal inhabitants of the terrestrial and

marine environments of the Turks and Caicos was helpful in narrowing down

possible species identifications, particularly within the fishes. An additional

aid was a survey of underwater resources of Grand Turk completed by

ichthyologists George Burgess of the Florida Museum of Natural History and

David Snyder of Continental Shelf Consultants, Jupiter, Florida (Burgess and

Snyder 1995). Current inhabitants may or may not reflect zooarchaeological








Current inhabitants may or may not reflect zooarchaeological findings, but this

type of information is what informs us about changes in ecosystems.

Because I worked with the Coralie site bones for multiple years and

learned what would be expected in the sample, only a very small percentage of

the remains were unidentifiable (UID), meaning not even identifiable to class.

According to Wing and Brown (1979), the average UID vertebrate remains in a

sample is 23%, but it ranges widely. At Golden Rock, St. Eustatius, 58% of the

bones were UID (Versteeg and Schinkel 1992). At Coralie, only 6.5% of the

remains were not identified to at least fish, bird, sea turtle, or reptile. I am

quite certain that these remains are mostly crushed bone fragments of sea

turtle, but based on their individual merits, they were classified UID.

All of these methods were combined to arrive at a reconstruction of the

diet consumed at this site. Situations that could bias the findings (Wing 1989)

were considered at all phases of the inquiry, and minimized as much as

possible. Poor bone preservation, resulting from acidic soils, is the most

common bias. The sandy soils within this site contain a high density of shell

inclusions. These shells, especially the burnt conch shell, release calcium

carbonate into the sandy soil matrix, increasing the preservation of delicate

materials, including bone. Proximity to the water table and indications that the

site was at one time inundated can have a negative effect on preservation. This

is most evident in some of the pottery remains which are, in certain deposits,

very friable (21% of the sample was considerably eroded and crumbly).

However, this had little effect on the bone remains. Overall, the preservation at

this site was excellent.









Other biases that can skew the subsistence picture include field

processing of certain species away from the site, and carrying raw materials

into the site for a secondary use. Here, only the data on conch remains may

have been influenced by these processes. Techniques of recovery and analysis

can severely bias the results. At Coralie, I supervised all excavations and

performed the faunal analysis for each year's work, resulting in consistent and

cumulative results.

The collection of bone from Coralie is catalogued and housed at the

Florida Museum of Natural History. All the bird, tortoise, and large rock

iguana remains are accessioned along with the entire sample from the 1995

season. The fish, sea turtle, and smaller rock iguana remains from the other
7
two seasons were primarily analyzed in the field and in most cases discarded '

after analysis. The pottery, stone pieces, shell tools, beads, and pendants all

remain in the Turks and Caicos at the National Museum on Grand Turk.


Temporal Controls


An essential part of this study was to trace changes in the faunal

assemblage over time. The site does not contain deep, continuous,

stratigraphic deposits, yet it was possible to pinpoint time ranges for different

areas of the site, using a combination of horizontal and vertical stratigraphy

and radiocarbon dating.

Table 1 presents the radiocarbon chronology for this site, while Figure 8

shows the general location of each of the dated samples. The three charcoal

samples taken from the easternmost units provided the earliest dates, from the









Table 1: Radiocarbon Chronology for the Coralie Site.


Laboratory Provenience Context Material & Age B.P. cal. AD Calibrated
Number Wood Type (with C13/C12 Intercept Age Range
correction) (2 sigma)
Beta-80911 110N 110OE Ash lens Charcoal 1280+/-60 AD 710 AD 650-885
FS #81 Area 10 Wild Lime
92-93.5 cmbd
Beta-93912 100N 108E Midden Shell* 1170 +/- 60 AD 770 AD 665-905
FS #168 Feature 23
78-90 cmbd
Beta-98698 100N 110E Hearth Charcoal 1230 +/-60 AD 790 AD 670-970
FS#178 Feature 25
70-80 cmbd
Beta-80910 124N 100E Hearth Charcoal 1160+/-60 AD 885 AD 720-1105
FS #35 Feature 5 Palm
47-62 cmbd
Beta-61151 ca. 120N 110E Zone2 Charcoal 1120+/-120 AD895- AD650-1160
Test pit (1992) 939
50-60 cmbd
Beta-114924 148N 104E Zone 2 Charcoal 1120+/-50 AD 960 AD 800-1015
FS #353 Level 2
70-80 cmbd
Beta-98697 110N 102E Post Layer Charcoal 1010 +/-50 AD 1020 AD 970-1165
FS #41 Zone 2
70 cmbd
Beta-93913 99N 99E Post Layer Shell* 930+/-60 AD 1020 AD 895-1145
FS#216 Zone 2
60-70 cmbd
Beta-96700 Mangroves Peat Layer Wood 940 +/-60 AD 1045 AD 995-1125
Paddle cf. Bullwood AD 1105
AD 1115
Beta-98699 96N 100OE Hearth Charcoal 900 +/-50 AD 1170 AD 1040-
FS#198 Feature 28 1215
55-74 cmbd
Beta-112311 124N 102E Tree root Wood Modem
50-100 cmbd Andira


Dates processed at the Beta Analytic Radiocarbon Dating Laboratory,
Miami, FL. Calibrations according to Stuiver et al. (1993).
*Shell dates adjusted for local reservoir effect.

















\N


[I 8th century
[0] 9th & 10th centuries
[ 1th & 12th centuries


100


120


EATING


Figure 8: Locations and Time Periods of Radiocarbon Samples.


140


120







100


meters
0 5 10









early to late 8th century. Excavations were usually stopped by the water table

at about 100 cmbs, yet soil tests revealed turtle bone occurring as deep as 110

cmbs in one of the easternmost units. The earliest radiocarbon date (intercept

A.D. 710) came from 93 cmbs. The very deepest deposits, below the water

table, have not been radiocarbon dated. It is possible that the site was first

settled before the 8th century. The main trench, approximately 10 m to the

west of the area with the earliest dates, sampled a distance of 90 m running

parallel to the creek edge. Radiocarbon dates calculated from samples along

this trench produced five intercept dates ranging between A.D. 885 and A.D.

1170. These dates become steadily later as the excavations move south.

Stratigraphic considerations further refine the above chronology. Figure

9 is a simplified profile map designed to show the east/west slope of the site.

Zone 2 was the original ground surface during the Amerindian occupation. It

slopes down from the west toward the creek, so the overburden layer varies in

thickness. As well as deepening toward the creek, this discolored horizon

lightens and thins, at points to 10 cm.

Most of this sites' features came from within the Zone 2 layer, but some

began below this in lighter colored sand (see Figure 9). In this figure, some of

the features are compressed along a single east/west axis to simplify the

presentation. The lower layer of features (see Fl 1, ash lens and shell midden

pockets) have top depths ranging from 70 to 90 cmbs. These are all associated

with the earliest occupation. Feature 11 is a fire pit located in the area of the

later, westerly units. It is the only feature below 70 cmbs in the western units.

Figure 10 is a photograph of these two fire pits (F 1 and F6) from different











1022 116B
I ground saurfoe I



' 7 35
cmbs







F igureiii 9: Prfl Map of~ i EatWs rnh
pe..ods of oua...... e p a
cmbs

. . . ... .-100






dicloe .ol asoiae it .t .h ale .cuain .obno -70
cmbs


,anth.ropohen horaion aZon, 2) ae Shelle ddefnd poet
.o. .no . ... ...



Fire pit







periods of occupation. The depth and discontinuity of F1 1 with the above

discolored soils associates it with the earlier occupation. A combination of

stratigraphy and radiocarbon dates helped defined the temporal boundaries

within the site.

The last characteristic of the site depicted in this profile is a wedge

shaped deposit of beach sand that covers only the eastern 10 m of the site. It

sits directly atop the buried anthropogenic horizon. This deposition effectively

ended the 8th century occupation. The boundary between this sand wedge and

the Amerindian deposit is abrupt, leading some authors (Hardman et al. 1998)

to conclude that a heavy storm surge or hurricane suddenly buried the site.

Alternatively, citing factors that argued for a low energy deposition, Scudder

(1997) concluded that this feature was the result of a gradual rise in sea level.































Figure 10: Photograph of Two Fire Pits (F 11 and F6) from Different
Periods of Occupation.


A horizontal separation of time periods occurs at this site, evidenced by

differing radiocarbon dates for separate areas and the lack of deep, continuous

stratigraphy. The settlement was moved at one point away from the creek. It

also seems to have moved south over time. As Rouse pointed out (1977), the

residents of West Indian sites moved their houses often and dropped their

refuse everywhere. This pattern can be seen on St. Eustatius (Versteeg and

Schinkel 1992), and on St. Croix, where Faber Morse (1997) demonstrates this

phenomenon at multiple sites. Even so, intra-settlement patterns are usually

discernible from the remains.

Taking all this into consideration, roughly designated locations within

the site were assigned time ranges, resulting in three general phases for the

occupation. Using more fine-grained distinctions noted in the vertical


2 t
.. " "- ." ".
. ,--" .-4 .
"*-, . J:.,^' M








stratigraphy, each of these areas and phases were then sub-divided into early

and late segments. Phase 1 is the colonization period occupation, with

deposition occurring only in the 8th century; 17 units produced artifacts only

from this time period. Phase 2, with 28 units, is roughly the period between

A.D. 800 and 1000. Phase 3, with 19 units, is the final period between A.D.

1000 and 1200. Temporal changes in faunal use were charted along these

subdivisions.

It has been eight years since explorations began at the Coralie site on

Grand Turk. Many people have been involved in numerous aspects of the

project. Because it is a small site, a great amount of very detailed information

has been collected. The strategies and methods used to accomplish the goals

for this investigation evolved over the years as new questions were raised by

the findings. The stage is now set to discuss what was found and to put forth

an interpretation of what happened when people first colonized the island of

Grand Turk.














CHAPTER 5
RESULTS OF FAUNAL ANALYSIS:
VERTEBRATE AND INVERTEBRATE REMAINS


There is nothing of greater human concern than sustaining oneself.

From analyses of the zooarchaeological remains at the Coralie site, it is

possible to reconstruct the diet consumed at this settlement, and to assess how

well the occupants provided for themselves from the resources available on this

small island. Because this is a colonization period settlement, the diet

consumed tells us something about the process of choice in human

subsistence. We get a glimpse of what the Amerindians preferred to eat, with

selections made from a variety of obtainable species. When looking at only

bone and shell remains, the focus is necessarily restricted to the consumption

of animal protein and fat from vertebrate and invertebrate species living in

either marine or terrestrial habitats. The total diet, which includes the

contribution of plant foods, cannot be understood from these methods (see

Stokes 1998).

Some of the animals recovered at Coralie are of interest beyond their

ability to provide calories to a diet. Certain identifications revealed animals of

biogeographic importance to this region, with species unknown from the Turks

and Caicos Islands, and a few undescribed in the West Indies. This chapter

will also report on what contexts and in what condition these bones were

found, which can give clues to how these animals were processed in the site.








Diet Reconstruction


The terrestrial habitat of Grand Turk supported an array of species that

are of biogeographic interest and provide evidence for overexploitation within a

limited space. But when looking at the total biomass making up the overall

Taino diet, it is the marine species that make the greatest contribution,

providing 94% of the estimated meat. Tropical inshore waters are "one of the

most complex, stable, and biologically productive ecosystems on earth"

(Nietschmann 1972:1). It is this habitat that truly sustained the inhabitants of

Grand Turk. The vast majority of meat consumed at this site was from sea

turtle. According to optimal foraging theory, sea turtle would have been a

preferred food when available (Keegan 1992), yet few Caribbean sites have

produced abundant turtle remains (for an exception see Allaire 1977). Coralie

provides evidence for the popularity of turtle meat in the Amerindian diet.


Green Turtle

Green sea turtle (Chelonia mydas) is the primary faunal deposit in the

site, with 75% of the bone, by mass, belonging to this one species. The three

seasons of excavations recovered 45 kg of turtle bone. Over 6000 identifiable

elements were counted, not including small shell fragments. The estimated

usable meat weight (referring only to the edible portion of the animal) from this

amount of bone was 2386 kg (over 5000 Ibs), providing 77% of the meat in this

diet. An individual sea turtle provides a very large food package. Even though

there were six times as many individual iguanas captured here as turtles, the

turtles were more important as a meat source.








The MNI for sea turtle was conservatively estimated at 50. Small

juveniles, with a carapace length (C.L.) of 30 to 50 cm, comprised 1/4 of this

population and weighed between 5 and 20 kg each. Mid-sized, sub-adult

turtles (C.L.: 50-85 cm) made up 60% of the collection, weighing between 20

and 70 kg each. The remaining 15% were large turtles over 1 m in length. The

largest green turtle in this site is estimated to have weighed 160 kg. These are

rough size estimations based on humeri measurements (Bjomdal et al. 1998).

The turtles harvested on Grand Turk were 85% sub-adults, and the few

adults captured were relatively small. Carr (1952) reported that adult green

turtles weigh on average 110 to 180 kg, but that specimens had been reported

up to 390 kg. Green turtles have an extremely slow growth rate, particularly

once adulthood has been reached (at ca. 70 kg or 85 C.L.), growing only scant

centimeters each decade (Bjomdal and Zug 1995). They can live and continue

to grow for beyond a century. Past populations, with no human predation,

would have supported very large adults, yet these were not harvested at

Coralie.

The Miskito Indians of Nicaragua are a modem turtle hunting society

with a subsistence pattern similar to that seen at the Coralie site. This culture

was intensively studied by Nietschmann (1972) and will be used as a point of

comparison for this investigation. The Miskito and the Taino exploited similar

resources, with an especially heavy reliance on turtle meat for the protein and

fat portion of their diets. The average size of the turtles harvested by the

Miskito was 86 kg, larger than those on Grand Turk. Their 6-m-long by 1-m-

across dugout canoes can carry a load of three to four adult turtles. Miskito










turtlers are technically capable of harvesting 180 kg turtles, but these are

increasingly rare within local populations.

Neitschmann calculated the amount of meat consumed in one Miskito

village over one year's time. A large village of 1000 people ate 819 turtles

(35,000 kg or 77,000 Ibs of meat). This is just less than one turtle per person

per year. Yet, turtle still contributed 70% of the animal portion of this diet.

Fish provided 6% of the meat consumed, and iguanas, small birds, and

mammals a total of 3%. The remaining 20% came from store-bought foods.

Even though the Coralie site is comparatively a much smaller habitation, the

percentages of meats consumed are roughly similar (see Figure 11).




100%
90%
80%
70%_
60%
60%-
40%
30%
20%
10%
0%
turtle fish iguana conch inverts, tortoise bird small
reptile



Figure 11: Percentage of Total Biomass Provided by the Primary Vertebrate
and Invertebrate Remains.


Species Identified and Quantified

Turtle may be the dominant meat source, but a great variety of animals

from various habitats were also harvested and deposited in the Coralie site








including fish, rock iguana, tortoise, conch, small invertebrates, and birds.

Table 2 is a list of the vertebrate and invertebrate species identified in the

remains. There are 32 species of fish, 20 species of bird, and seven species of

reptile. No mammal bones were found. For the invertebrates, there are 33

species of gastropod, 22 bivalves, five crustaceans, two echinoderms, and one

chiton. In looking over this simple list of names, a few unusual discoveries

emerge; the first is tortoise. Only historic period, introduced tortoises survive

in the West Indies, and all the endemic tortoises were thought to have gone

extinct before humans first arrived in these islands (Auffenberg 1967; Schwartz

and Henderson 1991; Watters et al. 1984). There are other unusual finds in

the bird identifications including two parrots, only one of which is known from

this region (Snyder et al. 1982), the South American scarlet ibis, and the

double-striped thick-knee, a bird thought to be long extinct in the Bahamas

(Buden 1987; Raffaele et al. 1998; Olson 1982; Pregill and Olson 1981).

Other species common in West Indian archaeological sites are absent

here, notably rats and hutias. Rice rats (Oryzomyines) were regularly

consumed in Lesser Antillean sites, but no species of rat inhabited the

Bahamian archipelago before the contact period (Deagan 1988; Morgan 1989).

The only rodent native to this region is the Bahamian hutia (Geocapromys

ingrahami). Once common throughout these islands, this species survives

today only on East Plana Cay in the southeast Bahamas (Alien 1892; dClough

1972). A cave recently excavated on Middle Caicos, with deep non-cultural

deposits, produced no hutia bone (R. Franz: personal communication). It

appears that the hutia never occupied either the Turks or Caicos banks.









Table 2: Vertebrate and Invertebrate Species List from the
Coralie Site, Grand Turk, with Scientific and Common Names.


SCIENTIFIC NAME


COMMON NAME


VERTEBRATES:
REPTILES:
Chelonia mydas Green turtle
Caretta caretta Loggerhead turtle
Geochelone sp. Tortoise
Cydura carinata Rock iguana
Leiocephalus psammadromus Curlytail lizard
Epicrates chrysogaster Boa
BIRDS:
Sula dactylatra Masked booby
Sula sula Red-footed booby
Ardea herodias Great blue heron
Egretta rufescens Reddish egret
Nyctannasa violacea Yellow-crowned night heron
Ardeidae Heron/egret
Eudodmus ruber Scarlet ibis
Phoenicopterus ruber Roseate flamingo
Dendrocygna arborea West Indian whistling duck
Pandion haliaetus Osprey
Haematopus palliatus American oystercatcher
Burdhinus bistriatus Double-striped thick-knee
Limnodromus griseus Short-billed dowitcher
Larus atric7la Laughing gull
Columba leucocephala White-crowned pigeon
Zenaida aurita Zenaida dove
Geotrygon chrysia Key West quail dove
Amazona leucocephala Cuban parrot
Amazona sp. UID Parrot
Tyrannus dominicensis Gray kingbird
Corvus nasicus Cuban crow
FISH:
Carchoarhinus sp. Shark
Dasyatis americana Southern stingray
Albula vulpes Bonefish
Holocentrus ascensionis Long-jaw squirrelfish
Epinephelus striatus Nassau grouper
Epinephelus sp. Grouper/hind
Mycteroperca sp. Grouper
Caranx crysos Blue runner
Caranx hippos Crevalle jack
Caranx ruber Bar jack
Caranx sp. Jack
Trachidnotus cf. falcatus Permit









Table 2-continued.


FISH (Con't):
Lutjanus cf. analis
Lutjanus cf. apodus
Lutjanus cf. griseus
Lutjanus sp.
Haemulon cf. album
Haemulon cf. flaviolineatum
Haemulon cf. plunmieri
Haemulon cf. sciurus
Haemulon sp.
Calamus sp.
Kyphosus sectatrix/incisor
Sphyraena barracuda
Bodianus rufus
Halichoeres radiatus
Scarus sp.
Sparisoma sp.
Acanthurus sp.
cf. Scomberomorus sp.
Bothus lunatus
Balistes vetula
Lactophrys sp.
Sphoeroides cf. testudineus
Diodon cf. hystrix


Y


Mutton snapper
Schoolmaster
Gray snapper
Snapper
Margate
French grunt
White grunt
Blue-striped grunt
Grunt
Porgy
Bermuda/yellow chub
Great barracuda
Spanish hogfish
Pudding wife
Parrotfish
Parrotfish
Surgeonfish/blue tang
Mackeral
Peacock flounder
Queen triggerfish
Boxfish/trunkfish
Checkered pufferfish
Porcupinefish


INVERTEBRATES:

CRUSTACEANS:
Callinectes sapidus Common blue crab
Coenobita dclypeatus Land hermit crab
Gecarcinidae Land crabs
Panularis argus Spiny lobster
GASTROPODS:
Fissurellidae Keyhole limpets
Diodora cayenensis Cayenne keyhole limpet
Astralium phoebium Long-spined starsnail
Lithopoma tectum West Indian starsnail
Cittarium pica West Indian topsnail
Nerita peloranta Bleeding tooth
Nerita tessellata Tessellated nerite
Nerita versicolor Four-toothed nerite
Nerita sp. Nerites
Cenchritis muricatus Beaded periwinkle
Littorina angulifera Mangrove periwinkle
Nodilittorina dilatata Prickly-winkle
Cerithium sp. Cerith
Strombus gigas Queen conch
Vermetidae Worm-snail
Hipponix antiquatus White hoof-snail









Table 2-continued.


GASTROPODS (Con't):
Macrocypraea zebra
Naticarius canrena
Polinices lacteus
Tectonatica pusilla
Cypraecassis tecticulus
Tonna pennata
Charordnia tritords variegata
Cymatium muricinum
Bursa sp.
Chicoreus pomum
Plicopurpura patula
Columbella mercatoria
Columbellidae
Fasciolaria tulipa
Oliva sp.
Olivella sp.
Tenebra dislocata
Bulla striata
BIVALVES:
Brachiodontes sp.
Anadara notabilis
Area zebra
Barbatia sp.
Pinctada imbricata
Lima scabra
Spondylus americanus
Anodontia alba
Codakia orbicularis
Divalinga quadrisulcata
Ludcna pensylvanica
Diplodonta sp.
Chama macerophylla
Chama sarda
Americardia media
Laevicardium laevigatum
Tellina fausta
Tellina listeri
Tellina radiata
Asaphis deflorata
Chione cancellata
Periqluypta listeri


Measled cowrie
Colorful Atlantic natica
Milk moon-snail
Miniature natica
Reticulated cowrie-helmet
Atlantic partridge tun
Atlantic trumpet triton
Knobbed triton
Frog-snail
Apple murex
Wide-mouthed rock-snail
West Indian dove-snail
Dove-snail
True tulip
Olive
Dwarf olive
Eastern auger
Striate bubble

Mussel
Eared ark
Turkey wing
Ark
Atlantic pearl-oyster
Rough fileclam
Atlantic thorny oyster
Buttercup lucine
Tiger lucine
Cross-hatched lucine
Pennsylvania lucine
Diplodon
Leafy jewelbox
Cherry jewelbox
Atlantic strawberry cockle
Eggcockle
Faust tellin
Speckled tellin
Sunrise tellin
Gaudy sanguin
Cross-barred venus
Princess venus


AMPHINEURA:
Acanthopleura granulata West Indian fuzzy chiton
ECHINODERMS:
Meoma ventricosa Red heart urchin
Tripneustes ventricosus West Indian sea egg
Clypeasteroida Sand dollar









Tables 3 and 4 are the quantified results of the analyses on all the

vertebrate and invertebrate remains. The most numerous animal in the site, by

MNI, was the rock iguana with 387 individuals. Fishes and reptiles

contributed nearly equal numbers of individuals to the total remains (ca. 450

each), but by usable meat weight, reptiles provided 84% of the diet, fishes just

15%. Although birds present a small portion of the diet, providing only 1% of

the meat consumed, a range of species was hunted and the species targeted

changed through time. Conch dominates the invertebrate remains. More

invertebrates were harvested than vertebrates (by total MNI), but by meat

weight, invertebrates contributed less than 3% of the total diet. Only 6% of the

diet, by estimated meat weight, came from terrestrial resources, yet there were

remains of many terrestrial species in the site, 49% by MNI (see Figure 12).

Crab is not included in this calculation. Tortoise, bird, and primarily iguana





100%-
90%-
80%-
70%-
60%
gAOL_.


WW fu ^^
40%-
30%
20%- r
10%
0%
fish


-I---


I -


iguana turtle bird tortoise small
reptiles


Figure 12: Percentage of Total MNI Provided by the Primary Vertebrate Species.







Table 3: Quantified Totals on Vertebrate Fauna from the Coralie Site.


SPECIES NISP NISP MNI MNI WEIGHT WEIGHT ESTIMATED E.M.W. AVERAGE % OF
BY % BY % (in g.) BY % MEAT WT. BY % INDIVIDUAL USABLE
REPTILES: ______________ _______ (in kg.)_____ WT. (in kg.) MEAT
Chelonia mydas 6545* 19.35% 50 5.20% 45340.50 80.40% 2385.90 73.00% 55.00 50%
Caretta caretta 1 < 1 0.10% 64.90 0.12% 226.80 6.90% 226.80 50%
Geochelone sp. 533 1.55% 18 1.90% 4179.20 7.40% 26.25 0.80% 2.50 70%
Cydclura carinata 12,538 37.00% 386 40.00% 2359.60 4.20% 111.35 3.40% 0.70 67%
Cycua sp. 1 < 1 0.10%/ 1.70 < 3.35 0.10% 5.00 67%
Leiocephaluspsammadromus 52 0.15% 12 1.20% 5.40 0.01% 0.84 0.02% 0.10 70%
Epicrates chrysogaster 4 0.01% 2 0.20% 0.30 < 0.42 0.02% 0.30 70%
Subtotal for reptiles: 19,674* 58.20% 470 49.00% 51,951.60 92.10% 2754.91 84.20%
BIRDS:
Sula dactylatra 20 0.06% 7 0.70% 15.30 0.03% 8.82 0.27% 1.80 70%
Sula sula 65 0.19% 7 0.70% 47.50 0.08% 4.90 0.15% 1.00 70%/
Ardea herodias 23 0.07% 2 0.20% 19.80 0.03% 1.80 0.03% 1.50 60%
Egrefta rufescens 1 < 1 0.10% 2.10 < 0.42 0.21% 0.70 60%
Nyctanmnasa violacea 1 < 1 0.10% 1.30 < 0.30 0.01% 0.50 60%
Ardeidae 1 0.20
Eudocimus ruber 8 0.02% 1 0.10% 1.90 < 0.52 0.02% 0.80 65%
Phoenicopterus ruber 3 0.01% 2 0.20% 2.50 < 1.80 0.03% 1.50 60%
Dendrocygna arborea 15 0.05% 4 0.40% 10.40 0.02% 2.80 0.09% 1.00 70%
Pandion haliaetus 3 0.01% 1 0.10% 3.20 < 0.50 0.02% 1.00 50%
Haematopus palliatus 1 < 1 0.10% 0.30 < 0.21 < 0.35 60%
Burhinus bistriatus 4 0.01% 2 0.20% 5.60 0.01% 0.30 0.01% 0.25 60%
Limnodromus griseus 1 < 1 0.10% 0.20 < 0.08 < 0.15 55%
Larus atricilla 2 0.01% 2 0.20% 0.70 < 0.48 0.02% 4.00 60%
Columba leucocephala 2 0.01% 2 0.20% 0.70 < 0.33 0.01% 0.25 65%
Zenaidaaurita 2 0.01% 1 0.10% 0.20 < 0.09 < 0.15 60%
Geotrygon chrysia 2 0.01% 2 0.20% 0.50 < 0.18 < 0.15 60%
Amazona leucocephala 6 0.02% 2 0.20% 0.80 < 0.33 0.01% 0.25 65%
Amazona sp. 2 0.01% 1 0.10% 0.30 < 0.26 0.01% 0.40 65%
Tyrannus dominicensis 1 < 1 0.10% 0.10 < 0.05 < 0.10 55%
Corvus nasicus 22 0.06% 6 0.60% 7.50 0.01% 1.26 0.04% 0.35 60%
Ayes 111 0.33% 128.40 0.05%_____ __
Subtotal for birds: 296 0.88% 47 4.90% 149.50 0.26% 25.43 .79%
FISH (con't next page): ---
Carcharhinus sp. 82 0.25% 4 0.40% 24.00 0.04% 98.40 3.00% 30.00 80%
Dasyatis americana 39 0.10% 7 0.70% 14.30 0.03% 11.90 0.36% 2.00 85%
*Some categories under NISP were not entirely counted and underestimate the results.







Table 3: continued.

SPECIES

FISH (con't):
Albula vulpes
Holocentrus ascensionis
Epinephelus striatus
Epinephelus sp.
Mycteroperca sp.
Caranx crysos
Caranx hippos
Caranx rubber
Caranx sp.
Trachinotus cf. falcatus
Lutjanus cf. analis
Lutjanus cf. apodus
Lutjanus cf. griseus
Lutjanus cf. mahogoni
Lutjanus sp.
Haemulon cf. album
Haemulon cf. flauiolineatum
Haemulon cf. plumieri
Haemulon cf. sciurus
Haemulon sp.
Calamus sp.
Kyphosus sectatrix/ incisor
Sphyraena barracuda
Bodianus rufus
Halichoeres radiatus
Scarus sp.
Sparisoma sp.
Acanthurus sp.
cf. Scomberomorus sp.
Bothus lunatus
Baistes vetula
Lactophyrs sp.
Sphoeroides cf. testudineus
Diodon cf. hystrix
Osteichthyes
Subtotals for Fish:
VERTEBRATES TOTALS:


7 ~ r ~ I I r..................... 1 I -.


INiar


294
17
4
395
3
4
1
34
64
2
1
5
5
1
441
16
8
3
1
1083
30
5
21
13
115
380
83
6
1
10
56
398
3
199
10,035*


NISP
BY %


MNI MNI I WEIGHT I WEIGHT


BY% I (ing.)


BY %


E.M.W.
(in kg.)


E.M.W.
BY %


AVERAGE
WEIGHT
(in kg.)


MEAT
I I S 1 1. J.


0.85%
0.05%
0.01%
1.20%
0.01%
0.01%
0.10%
0.10%

0.20%
<
<

0.02%
0.02%
<
1.30%
0.05%
0.03%
0.01%
<
3.20%
0.09%/
0.02%
0.06%
0.04%
0.35%
1.15%
0.25%
0.02%
<
0.03%
0.15%
1.20%
0.01%
0.60%
29.50%/o


27
5
4
32
1
3
1
9
4
2
1
3
3
1
51
8
8
1
1
149
5
4
6
4
25
25
12
2
1
2
7
12
2
9


2.80%
0.50%0/
0.40%
3.30%
0.10%
0.30%
0.10%
0.90%
0.40%
0.20%
0.10%
0.30%
0.30%
0.10%
5.40%
0.80%
0.80%
0.10%
0.10%
15.50%
0.50%
0.40%
0.60%
0.40%
2.70%
2.70%
1.30%
0.20%
0.10%
0.20%
0.70%
1.30%
0.20%
0.90%


39.60
3.80
7.20
421.90
5.10
1.60
5.10
8.80
23.90
0.50
1.40
4.40
4.50
0.80
195.50
6.70
3.30
0.80
0.30
211.60
18.90
1.30
20.30
4.20
45.10
807.80
53.70
1.10
0.30
1.80
18.60
58.50
1.90
137.90
2157.80


0.07%
0.01%
0.01%
0.75%
0.01%
<
0.01%
0.02%
0.04%
<

0.01%
0.01%
0.01%
<
0.37%
0.01%
0.01%
<
<
0.40%
0.03%
<
0.04%
0.01%
0.08%
1.50%
0.10%
<
<
<
0.03%
0.11%
<
0.25%
3.80%


24.30
2.40
4.22
33.80
1.32
2.16
0.90
4.86
2.16
1.88
0.99
2.49
2.49
0.66
42.33
7.20
3.60
0.63
0.63
93.87
3.15
2.52
24.20
1.70
10.62
78.75
5.40
0.94
0.90
0.90
6.30
8.64
0.95
7.20


0.75%
0.07%
0.13%
1.00%
0.04%
0.07%
0.03%
0.15%
0.07%
0.06%
0.03%
0.07%
0.07%
0.02%
1.30%
0.22%
0.12%
0.02%
0.02%
2.90%
0.10%
0.07%
0.75%
0.06%
0.32%
2.40%
0.16%
0.03%
0.03%
0.03%
0.19%
0.27%
0.03%
0.22%


33,828* 100.00% 958 100.00% 56,415.70 100.00%


1.00
0.60
1.20
1.20
1.50
0.80
1.00
0.60
0.60
1.00
1.20
1.00
1.00
0.80
1.00
1.00
0.50
0.70
0.70
0.70
0.70
0.70
1.00
0.50
0.50
3.50
0.50
0.50
1.00
0.50
1.00
0.90
0.50
1.00


%
USABLE
MEAT


90%
80%
88%
88%
88%
90%
90%
90%
900/%
94%
83%
83%
83%
83%
83%
90%
90%
90%
90%/
90%
90%
90%
84%
85%
85%
90%
90%
94%
90%/
90%
90%
800/%
95%
80%


13,858 1 41.00% 441 46.00% 1 4314.60 7.70% 495.36 1 15.00%


3275.70


100.00%








Table 4: Quantified Totals on Invertebrate Fauna from the Coralie Site.


SPECIES

GASTROPODS:
Fissurellidae
Diodora cayenensis
Astralium phoebium
Lithopoma tectum
Cittarium pica
Nerita peloranta
Nerita tessellata
Nerita versicolor
Nerita sp.
Cenchritis muricatus
Littorina angulifera
Nodilittorina dilatata
Cerithium sp.
Strombus gigas
Vermetidae
Hipponix antiquatus
Macrocypraea zebra
Naticarius canrena
Polinices lacteus
Tectonatica pusilla
Cypraecassis tecticulus
Tonna pennata
Charonia variegata
Cymatium muricinum
Bursa sp.
Chicoreus pomum
Plicopurpura patula
Columbella mercatoria
Columbellidae
Fasciolaria tulipa
Oliva sp.
Olivella sp.
Tenebra dislocata
Bulla striata
Subtotal Gastropods:


NISP


NISP
BY %


MNI


MNI
BY%


WEIGHT
(in g.)


WEIGHT
BY %


ESTIMATED
MEAT WT.**
(in kg.)


E.M.W.
BY %


-- 4. 1 4 4 I 4. I 1


10
2
2
2
1306
4
29
60
284
19
1
1
1
45,613*
5
1
6
3
1
2
1
5
2
3
1
31
1
2
1
5
4
11
1
32


47,453*


0.02%
<
<
<
2.49%
0.01%
0.05%
0.12%
0.54%
0.04%
<
<
<
86.76%
0.01%
0.01%


0.01%






0.06%



0.01%
<
<
<
<





0.01%





0.02%
0.06%
<
<
0.06%
<
<
<
0.01%
0.01%
0.02%
<
0.06%


0.46%
0.16%
0.16%
0.09%/
0.61%
0.31%
1.40%
2.70%
0.79%
1.48%
0.09%
0.09%
0.09%00
30.80%
0.09%o
0.09%
0.46%
0.23%
0.09%
0.16%
0.09%
0.16%
0.16%
0.23%
0.09%
1.15%
0.09%
0.16%
0.09%
0.55%
0.31%
0.85%/
0.09%
2.500/%


15.80
0.30
80.40
8.10
15,688.80
14.30
64.60
114.10
399.10
11.30
0.40
0.20
0.10
455,936.40
6.10
0.80
113.10
21.10
1.10
3.20
10.20
4.10
16.80
23.60
4.20
293.40
31.40
1.40
0.50
9.60
6.90
2.20
0.20
52.60


<
<
0.02%
<
3.28%
<
0.01%
0.02%
0.08%
<
<
<
<
95.38%
<
<
0.02%









0.06%
<
<
<
<
<
<
<
<
0.06%
<
<
<
<
<
<
<
0.01%


2.76
0.01
0.04
0.07
0.20




67.15












0.37



0.35


2.70%
<
<






64.70%












0.30%



0.30%


AVERAGE
INDIVIDUAL
MEAT WT. (kg.)


0.035
0.002
0.002
0.002
0.002




0.17


I 4. I 4. 1 1


90.26%


762


59.15%


472,936.40


98.94%


.J. I ________ I J. ________ .1 _______ .L


71.04


68.00%


"This category under NISP was not entirely counted and underestimates the results.
**Calculations completed only for primary edible species.







Table 4: continued.


BIVALVES:
Brachiodontes sp.
Anadara notabilis
Area zebra
Barbatia sp.
Pinctada imbricata
Lima scabra
Spondylus americanus
Anodontia alba
Codakia orbicularis
Divalinga quadrisculcata
Lucina pensylvanica
Diplodonta sp.
Chama macerophylla
Chama sarda
Americardia media
Laevicardium laevigatum
Tellina fausta
Tetlina listeri
Tellina radiata
Asaphis deflorata
Chione cancelUata
Periglypta listeria
Subtotal Bivalves:
AMPHINEURA & ECHINODERMS:
Acanthopleura granulata
Meoma ventricosa
Thpneustes ventricosus
Clypeasteroida
Subtotal Amphineura/Echin.:
CRUSTACEANS:
Callinectes sapidus
Coenobita clypeatus
Gecarcinidae
Panularis argus
Subtotal Crustaceans:
INVERTEBRATE TOTALS:


NISP


338
1
4
2
14
1
2
4
348
19
2
1
6
95
1
10
2
186
569
165
165
3


1938

100
1
13
2
116

7
102
2894
65
3068


NISP
BY %


0.63%
<
0.01%
<
0.03%
<

0.01%
0.66%
0.04%
<
<
0.01%
0.18%
<
0.02%
<
0.35%
1.08%
0.32%
0.32%


MNI


31
1
3
2
3
1
2
4
49
10
2
1
4
48
1
5
2
43
61
25
81
2


3.68%

5.50%

0.03%
<
5.53%

0.01%
0.19%
5.50%
0.12%
5.80%


=1=


MNI
BY%


2.50%
0.09%
0.23%
0.16%
0.23%
0.09%/
0.16%
0.31%
3.70%
0.79%
0.16%
0.09%
0.31%
3.70%
0.09%
0.40%
0.16%
3.30%
4.80%
1.90%
6.30%
0.16%


29.60%0

1.10%
0.09%
0.09%
0.09%
1.37%

0.16%
0.70%
6.50%
2.50%
10.00%


WEIGHT
(in g.)


I -. I-


113.60
2.20
12.20
3.90
6.60
1.90
56.20
13.10
1016.20
5.50
17.50
0.80
67.60
91.80
7.10
12.90
2.60
442.10
1056.10
605.50
130.10
31.30


3696.80

252.40
0.10
2.40
1.10
256.00

2.30
24.90
1025.40
63.90
1116.40


WEIGHT E.M.W.** E.M.W. AVERAGE
BY% (in kg.) BY % WT. (in kg.)


0.02%





0.01%
<



0.21%
<
<





0.01%
0.02%



0.09100
0.22%
0.13%
0.03%
<


0.49 0.50%/


0.43
0.61
0.25


0.40%
0.60%
0.30%


1.78 1.80%


0.07


7.00%



17.00
13.20
30.20%


16.40
12.60
29.00%


0.01









0.01
0.01
0.01





0.005



0.005



0.20
0.40


0.77%


0.05%
<
<
-C

0.05%



0.21%
0.01%
0.24%


= I I = I I -


52,575* 100.00% 1289 100.00% 478,005.60 100.00% I 103.09 1100.00%


I I I I


52,575*


100.00%


1289


100.00%


478,005.60


100.00%1 103.09 100.00%









contribute to these subsistence remains. These numbers indicate that the

inhabitants put considerable effort and time into harvesting terrestrial species.

From a computation made from the faunal remains of 13 Turks and

Caicos and Bahamian sites, I found an average of 4% (by MNI) terrestrial

resources. If I include crab in this tally, the number of terrestrial remains

jumps to 12.5% (using a six site sample). According to these results,

Bahamian islands provided little in the way of terrestrial resources to its

inhabitants.

The data from Coralie presents a different picture. The radiocarbon

dates and the pottery style found here show that this occupation occurred in

the early colonization period. Because island resources are often utilized

before any settlement occurs (Irwin 1992), it can be difficult to prove that a site

represents the first arrival of humans onto an island. The fact that there is no

evidence for human occupation in the Bahamian archipelago during the

ceramic period further supports the possibility that this settlement was the

first predation by humans upon this pristine population of animals. In this

situation, terrestrial species do contribute substantially to the Amerindian diet,

simply because they had not yet been eliminated from the environment.


Other Reptiles

Only one sea turtle bone belonged without a doubt to a species other

than green turtle. A very large maxillary fragment (upper jaw) was identified as

the remains of a loggerhead (Caretta caretta). This specimen had an estimated

live weight of at least 450 kg (1000 Ibs). Carr reported (1952) that loggerheads









commonly grow to 350 kg, but that since the 1960s, they had rarely been seen

over 150 kg. Because of its size, the giant loggerhead from Grand Turk must

have been caught while laying its eggs and its meat butchered in the field. This

suggests that both loggerheads and greens were nesting on Grand Turk

beaches in the past.

Leatherbacks (Dermochelys spp.) are the largest turtles known from the

West Indies, weighing an average of 450 kg (Carr 1952). They have not been

found archaeologically and apparently were never consumed. The other large

reptile missing from the Grand Turk assemblage is the crocodile. Crocodile has

been identified at the archaeological site of CK-14 on Crooked Island and AK-

14 on Acklins (deFrance 1991), legitimizing Spanish accounts of Bahamian

lakes inhabited by "serpents...seven palmos in length" (Dunn and Kelley

1989:107), which totals 1.7 m. Cuban crocodiles (Crocodylus rhombifer) have

been found in cave deposits on New Providence (Pregill 1982), San Salvador

(Olson et al. 1990), and Abaco (Franz et al. 1995). There is no evidence that

crocodiles ever inhabited Grand Turk or any of the Caicos islands.

The rock iguana (Cyclura carinata) is the most abundant single species at

Coralie, contributing 40% of the MNI. The total amount of fish eaten

outnumber the iguanas, and because of this, rock iguanas rank third in overall

contribution to the Taino diet. Iverson (1979) conducted an extensive study of

a large population of Cyclura carinata on the island of Pine Cay on the Caicos

bank. He found the average weight of an adult male to be .94 kg, an adult

female to be .48 kg, with average lengths (using snout-vent measurements-

SVL) of 28 cm (with tail about 60 cm) and 23 cm (with tail about 50 cm)









respectively. With a maximum length of 75 cm and a maximum weight of

about 2 kg, this species of rock iguana is the smallest in the West Indies.

The iguanas harvested at Coralie were 75% sub-adult (n=296), weighing

an estimated average of .3 kg each. For the 91 adults found, an average of .7

kg was used in calculations to take into account the sexual dimorphism found

in this species. Also identified from large vertebrae were three, probably male,

iguanas weighing 1 kg each, and another two individuals weighing

approximately 2 kg each. All these individuals fall into the known size range

for this species. A final individual exceeds this size range.

A specimen with a live weight estimated at 5 kg was found in the

deposits on Grand Turk. This very large iguana had an estimated SVL of 45

cm, and a total length of nearly 1 m. It approaches the size of rock iguanas

from Hispaniola and Cuba. The size was calculated from a maxilla that had a

tooth row length of at least 55 mm, yet this maxilla was broken at both ends,

so this figure may be an underestimation. This individual suggests that the

size range of Cyclura carinata was larger than it is known presently, due to its

evolution on an island with no predators.


Fishes

Fishes were an important resource from the very beginning of this

occupation, ranking second in the contribution to overall diet. Fish provided a

usable meat weight of 495 kg, 15% of the total meat intake at this site. There

was a variety of species identified from the nearly 14,000 bones, totaling 24

genera. Appendix A gives a systematic account of all the fish species identified









to date, and relevant data on availability, size, habitat preference, and

behavior. All of the fishes identified in the site are still present in the waters

surrounding Grand Turk today, although some are rare.

The most common fishes at Coralie were grunts (Haemulon sp.), with

167 individuals recovered. This reflects the situation in the water today, where

grunts provide the majority of the biomass on the reefs. They are almost three

times as numerous in this site as the second ranking species, snappers. Even

though grunts are a relatively small fish, they still contribute the greatest

amount of biomass to the diet because they are so abundant. In the Coralie

remains, they range in size from .3 to 1.8 kg in live weight. The average size

clustered quite strongly, with 74% of the atlas centrum measurements being

between 5 and 7 mm (equaling .4 to .75 kg live fish weight). Many of the

species within this genus school, so this size clustering may reflect a particular

procurement strategy, but it is more likely a natural reflection of the size

ranges within the population.

This assemblage contained a high percentage of large fish. Table 5 lists

the calculated sizes for identified species based on vertebra and atlas centrum

measurements. The largest vertebra found belonged to a grouper and

measured 26 mm across. Allometrically, this converts to a fish weighing nearly

11.5 kg. A huge jaw came from a barracuda estimated to have weighed 20 kg.

One fully adult shark was captured, weighing approximately 65 kg, along with

one juvenile shark (ca. 4 kg), and two mid-size sub-adults, weighing about 27

kg each. A few other groupers, snappers and parrotfish (Scarus sp.) fell into a

large size range, weighing between 5 and 10 kg each. The average size for the









Table 5: Fish Size Estimates from Vertebrae of Identified Species.


SPECIES Element NISP Size Range** Average Size Range Average
measured (centrum diam.- (in mm) (by weight- (in kg)
__________ ) inkg
Carcharhinus sp. vertebra 4 8.4 to 24.7 15.80 N/A N/A
Dasyatis americana vertebra 34 5.5 to 12.5 8.20 N/A N/A
Albula vulpes vertebra 67 6.0 to 10.5 7.95 .57 to 1.79 1.00
Holocentrus
ascensionis atlas 1 6.4 6.40 0.65 0.65
Epinephelus sp. atlas 32 4.9 to 22.0 8.40 .37 to 8.13 1.13
vertebra 87 4.1 to 26.0 13.40 .26 to 11.44 2.95
Caranxsp. atlas 5 4.9 to 7.8 6.10 .37 to .97 0.59
Trachinotus cf.
falcatus atlas 2 5.0 to 5.5 5.25 .39 to .47 0.43
Lutjanus sp. atlas 57 3.8 to 17.5 7.70 .22 to 5.09 0.95
vertebra 47 5.0 to 19.3 8.70 .39 to 6.22 1.22
Haemulon sp. atlas 132 4.5 to 10.6 6.60 .31 to 1.82 0.69
vertebra 97 3.5 to 11.2 5.70 .19 to 2.04 0.51
Calamus sp. atlas 2 5.2 to 6.0 5.60 .42 to .57 0.49
Kyphosus sp. atlas 1 8.0 8.00 1.03 1.03
Sphyraena barracuda atlas 1 7.6 7.60 0.92 0.92
vertebra 8 4.6 to 14.4 8.20 .33 to 3.41 1.08
Halichoeres radiatus atlas 6 5.3 to 6.3 5.75 .44 to .63 0.52
Scarussp. atlas 4 13.8 to 15.5 14.60 3.13 to 3.97 3.51
vertebra 23 12 to 20.5 16.10 2.35 to 7.25 4.29
Sparisoma sp. atlas 1 5.4 5.40 0.46 0.46
Acanthurus sp. vertebra 2 4.0 to 7.8 5.90 .25 to .97 0.55
Balistes vetula atlas 1 8.0 8.00 1.03 1.03
vertebra 3 6.7 to 11.3 9.00 .71 to 2.08 1.30
Lactophyrs sp. vertebra 24 4.3 to 11.0 7.65 .29 to 1.97 0.93

**Formula for allometric conversion from measurement to weight:
Log y = 2.047 (log x) + 1.162 (based on known dimensions and weights of 50
specimens), x = anterior diameter of the centrum (mm); y = body weight
(grams). (Wing and Scudder 1983:204)


plentiful snappers and groupers was between 1 and 2 kg each. Parrotfish

(Scarus sp.) weighed on average between 3.5 and 4 kg.

If the meat weights of just the three large sharks and the barracuda are

combined, they end up providing the majority of meat in the fish diet. Sharks









only rank 16th in terms of the number of individuals captured in this site, but

they rank second in meat contribution. The rankings for all these fish species,

by MNI and meat weight, are presented in Table 6. Reitz (1990) found a similar

situation at the site of El Azicar, Panama, where six cartilaginous fish equaled

the meat value of 68 bony fish. Similarly, Wing (1998) found in West Indian

sites that often a few large sharks or tunas provided the majority of the

estimated meat weight from fishing.

The 1/4' mesh assemblage of fish vertebrae were all measured and

provided an average size, site wide, of 6.03 mm (equaling .57 kg live fish

weight). The average size of the vertebrae recovered in the fine mesh samples



Table 6: Ranking of Fishes at Coralie by MNI and Estimated Meat Weight.


RANK SPECIES MNI SPECIES MEAT WEIGHT
____________________(in kg.)
1 Haemulon sp. 167 Haemulon sp. 105.93
2 Lutjanus sp. 59 Carcharhinus sp. 98.40
3 Epinephelus sp. 36 Scarus sp. 78.75
4 Albula vulpes 27 Lutjanus sp. 48.96
5 Halichoeres radiatus 25 Epinephelus sp. 38.02
6 Scarus sp. 25 Albula vulpes 24.30
7 Caranx sp. 17 Sphyraena barracuda 24.20
8 Spanrisoma sp. 12 Dasyatis americana 11.90
9 Lactophyrs sp. 12 Halichoeres radiatus 10.62
10 Diodon cf. hystrix 9 Caranx sp. 10.08
11 Dasyatis americana 7 Lactophyrs sp. 8.64
12 Balistes vetula 7 Diodon cf. hystrix 7.20
13 Sphraena barracuda 6 Balistes vetula 6.30
14 Holocentrus ascensionis 5 Sparisoma sp. 5.40
15 Calamus sp. 5 Calamus sp. 3.15
16 Carcharhinus sp. 4 Kyphosus sp. 2.52
17 Bodianus rufus 4 Holocentrus ascensionis 2.40
18 Kyphosus sp. 4 Tracdhinotus cf. falcatus 1.88
19 Sphoeroides sp. 2 Bodianus rufus 1.70
20 Acanthurus sp. 2 Mycteroperca sp. 1.32
21 Bothus lunatus 2 Sphoeroides sp. 0.95
22 Trachinotus cf. falbatus 2 Acanthurus sp. 0.94
23 cf. Scomberomorus sp. 1 cf. Scomberomorus sp. 0.90
24 Mycteroperca sp. 1 Bothus lunatus 0.90









(n=182) dropped to 4.2 mm (equaling .3 kg live fish weight). The smallest

vertebra recovered in the 1 mm mesh was 1.2 mm in diameter (a fish weighing

only 20 grams). These tiniest fishes may have been used as bait or added to

soup stocks. Table 7 is a listing of the remains found in the fine mesh

samples.


Invertebrates

There were 61 invertebrate species identified in this collection harvested

from various habitats (Colin 1978). Some of the smallest species are

associated with the mangrove fringe and are likely natural deposits in the site,

yet most represent subsistence activities. From a total of 1289 individuals, the

invertebrate contribution of meat to this diet is only 2.2%. With the exception

of Queen conch, gastropods and bivalves do not provide much meat per

individual. The fact that the small gastropods and bivalves were gathered at all

may seen odd when so many other food sources were readily available

(Shackleton 1983).

In looking again at the Miskito Indians of Nicaragua, Nietschmann

(1972) reported that they included small invertebrates as a semi-regular

addition to their diet. They gathered tiny coquinas (Donax sp.) in the leaner,

summer months when turtles were scarce. The Donax is a shell only 1 to 2.5

cm across, housing a very tiny creature. Yet, in two hours, one person can

gather enough Donax shells to yield almost 1 kg of meat and 750 calories.

Turtle hunting provides 2000 calories in the same amount of time, just three

times as much (Keegan 1992). Mollusks can be gathered at very little cost to










Table 7: Fine Mesh Samples of Fauna from the Coralie Site.


MESH SIZE: 4 mm 2 mm 1 mm
SPECIES NISP MNI WEIGHT NISP WEIGHT NISP WEIGHT
VERTEBRATES:
Reptiles:
Chelonia mydas 237* 73.30 71 4.30
Cyclura carinata 100 3 13.20 122 4.30
Leiocephahlus psammadromus 1 0.10
Iguanidae 2 18 0.10
Typhlops ricordi 1
Fish:
Dasyatis americana 1 0.10 1
Lutjanus sp. 2 0.20 5 0.20 3
Haemulon sp. 11 1 11.10 10 0.60
Sphyraena barracuda 1 1 0.20 1 0.10
Halichoeres sp. 2 0.30
Labridae 3 0.10 1
Scarus sp. 1 0.20
Sparisoma sp. 2
Scaridae 2
Lactophyrs sp. 2 0.10 38 0.80 6 0.10
Diodon sp. 1
Osteichthyes fragments 84 7.80 203 6.20 90 0.70
Osteichthyes vertebrae 86 7.10 105 3.60 6
Vertebrata 93 8.10 750 23.70 80 0.80
Subtotal for Vertebrates: 620* 5 121.70 1315 44.00 207 1.70
INVERTEBRATES:
Crusteceans:
Gecarcinidae 25 5.10 19 0.90 10 0.10
Gastropods:
Strombus gigas 1000* 1011.30 many 250.10
Olwvasp. 1 1 2.50
OliUvellasp. 2 2 0.30
Land Gastropod:
Truncatella putchella many many
Bivalves:
Asaphis deflorata 1 1 6.30
Echinoderms:
Echinoidea 1 0.20
Subtotal for Invertebrates: 1049* 4 1025.50 20* 251.20 10* 0.10

FAUNAL FINE MESH TOTALS: 1669* 9 1147.20 1335* 295.20 217* 1.80

*Certain remains occurred in abundance in very small fragments and
were not individually counted.









its consumer (Armstrong 1980). They require no long travel expenditures to

harvest and there are no risks involved. Invertebrates are available year-round

and can provide food when hunting, fishing, or weather is unfavorable. Women

and children often collect invertebrates when men are engaged in other

subsistence activities (Meehan 1983). It is for these reasons that mollusks will

make a contribution to even the richest diets. At Coralie, it is a very small part

of the diet, used either to supplement calories in a lean time, or to add some

variety to the menu. The nerites, in particular, seemed to cluster in the

deposits, with each grouping being the refuse from a single meal.

Even though invertebrate remains are small in quantity at Coralie, the

collection is diverse. Gastropods outnumber bivalves, but if conch is

eliminated from consideration, the MNI from small gastropods and bivalves is

equal. Conch is the most important invertebrate species, with an MNI of 395.

Each of these conchs provided .17 kg of meat, totaling 67 kg. This is

undoubtedly a small percentage of the total conch meat consumed by these

settlers, for very rich conch beds surround the Turks and Caicos and have

been harvested for centuries (Sadler 1997).

The presence of conch in this site represents both a dietary resource and

a functional material resource, for conch has many secondary uses. Here, it

was fashioned into tools and ornaments, and used in the construction of

cooking hearths. It is this last use that accounts for the majority of the

remains inside the confines of the settlement. Heavy conch shells would not be

carried back to a site except for secondary uses such as these; meat would be

extracted in the field.









This fact is reiterated in some ceramic period shell middens, which

contain a few targeted species, usually bivalves, and already fashioned conch

tools (Nodine 1987). Conch refuse does not show up in these middens either,

except from a secondary use. The whole conchs recovered at Coralie are large

and many detached, full grown lips were found. The larger conchs were

preferred in hearth construction and may have been better suited for tool

manufacture as well. The length of 10 whole, adult conchs ranged between 19

cm and 25 cm from spire to base, averaging 22 cm. Only two whole, sub-adult

rollerss" approximately half the size of the adults (11 cm long) and three whole

juveniles (3 to 5 cm long) were recovered inside the site area. Most of the

conch refuse was broken up from secondary uses.

Nerites (Nerita sp.), small marine snails, are the second most common

mollusk by MNI, with 159 individuals. Three different species of nerites were

picked off the rocks south of the site on the leeward shore. This same habitat

supports the West Indian topsnail (Cittarium pica), periwinkles, and chitons.

The topsnail also had a secondary use here as a raw material for tool making.

The most common edible bivalves were two species of tellin (Tellina radiata and

Tellina listen), tiger lucines (Codakia orbicularis), and the gaudy sanguin

(Asaphis deflorata). This last species is new in Turks and Caicos archaeological

sites.

The contribution of sea urchins to human diets is rarely recognized,

although they are consistently present in small numbers in archaeological

sites. The red heart urchin (Meoma ventricosa) and the West Indian sea egg

(Tripneustes ventricosus) occur at Coralie. The eggs of the female sea egg are









quite palatable. Spiny lobsters (Panularis argus) are common with 33

individuals recovered. Blue crabs (Callinectes sapidus), which occupy the

mangroves and shallow, tidal flats, are present but rare in the site.

The only terrestrial invertebrates that were consumed at Coralie were

crabs. Land snails did not contribute to the diet. Land crabs (Gecarcinidae)

dominate the collection. There are two species living on Grand Turk today.

The giant white crab (Cardisoma guanhumi), also called blue crab because of

their purple, juvenile phase, reach a maximum body length of 9 cm. They are

coastal inhabitants whose burrows must reach the water table. Black crabs

(Gecarcinus lateralis) are smaller, tastier, and live more inland in cliffs and hills

(Chace and Hobbs 1969). Depending on the location of the site and the habitat

surrounding it, different crabs will be exploited. On St. Kitts, the black crab

predominates, but on Antigua, Crooked Island and the site of Maisabel, Puerto

Rico, all remains are of the white crab (DeFrance 1989, 1991; Jones 1985;

Goodwin 1979). The amount of crab remains in some Saladoid sites can be

staggering. At the earliest site on St. Kitts (the Cayon site), there was an

average of 2500 crab claws in each 1 m unit. At the Coralie site, it took 217 1

m units to produce an equivalent amount of crab claws. Both species of land

crab were harvested and some crab remains were found burned inside cooking

hearths. Crab ranked eighth in meat provided to this diet.


Biomeographically Important Species


The species just described comprises the primary subsistence base of

the early occupants of Grand Turk. Some of the other animals identified in the









remains provide information on past environments, past species distributions,

speciation on islands, and human interference with these processes. The most

unusual find at Coralie is the discovery of a large bodied, thin-shelled tortoise.


Tortoise

All native species of West Indian tortoise are extinct. They are only

known today from paleontological sites in the Bahamas, Hispaniola, Cuba,

Barbuda, Mona, Navassa, and Sombrero Island (Auffenberg 1967; Franz and

Woods 1983; Pregill 1982; Steadman et al. 1994; E. Williams 1952). All are of

the genus Geochelone. Even though tortoises did not make a large contribution

to the diet at Coralie, they were consumed. The discovery of tortoise bones of

this genus in Amerindian cooking hearths on Grand Turk introduces a

previously unknown element to West Indian biogeography. This tortoise has a

unique morphology and is a new species endemic to Grand Turk. The

systematic associations between this specimen and other extinct varieties of

West Indian tortoises are presented in Appendix C.

All Caribbean islands that currently support tortoise populations do so

because of human introductions of South American species (Pritchard and

Trebbau 1984). Today, some of the Virgin Islands and Lesser Antilles sustain

populations of Geochelone carbonaria (South American yellow-footed tortoise),

which is still a food source for some of these local populations (Censky 1988).

It is uncertain whether Amerindians or early European colonists did this

introduction. Watters et al. (1984) and Schwartz and Henderson (1991) report

that it was supposed that the Saladoid people transported the tortoise, as they









did small mammals, to the Lesser Antilles. This assumption is unsupported

because the tortoises' remains have never been found in archaeological sites.

Due to its island evolution, the Grand Turk tortoise is three times larger than

the South American species.

Analyses at Coralie identified 18 separate individuals, including

hatchlings, juveniles and large adults. These tortoises were harvested from a

local breeding population. The Taino did not introduce this animal to this

region. Numerous remains of a tortoise, structurally similar, though not

identical, to the one on Grand Turk, were found recently in a cave on Middle

Caicos that contained deep, pre-human deposits (R. Franz: personal

communication). Tortoises lived in the Turks and Caicos Islands for a long

time, adapting themselves precisely to this habitat, long before the first people

arrived.

Certain inferences can be drawn from the morphology of the shell, which

reveal how this species adapted to a small, dry, island environment. On

oceanic islands with no large predators, tortoises were free to grow large

without bearing the weight of a thick protective covering. The Grand Turk

tortoise's shell is very thin and correspondingly high domed to provide the

structure with some strength. This species did not have to dig burrows for

protection, like its thin-shelled North American counterpart the gopher tortoise

(Gopherus polyphemus). The ability to retract its neck and limbs under its shell

is another tortoise defense mechanism. Gopherus can fully retract its neck and

rotate its forelimbs in front of its face for further protection (Bury and Germano

1994). From reconstruction of the anterior peripherals, nuchal and bridge









elements of the Grand Turk tortoise, it appears retraction of the head would

not have been possible inside this narrow, slot-like opening.

Tortoises are herbivorous, live in temperate and tropical zones only, and

lay a few eggs at a time, which they bury in shallow sand burrows (Auffenberg

1976). They are very slow growing, but if undisturbed they can reach great

proportions and live extremely long lives. It takes 20 years, under optimal

conditions, for tortoises to grow to sexual maturity (P. Pritchard: personal

communication). The two largest and oldest Grand Turk specimens (with a

C.L. of 75 cm, or 2.5 ft) show unusual evidence of wear on the plastral surfaces

nearest the bridge. This area had literally been flattened by these large, heavy

animals scraping themselves across this rocky landscape. This seems to be a

function of age and weight, for the smaller specimens do not display this

characteristic.

Morphologically, and perhaps behaviorally, the Grand Turk tortoise is

similar to another species long adapted to dry, small, and isolated oceanic

islands-the famed Galapagos tortoise (Geochelone elephantopus). The

Galapagos tortoise is larger, reaching 130 cm in total length (Fowler 1983), but

the two species share other characteristics, most notably thin, high domed

shells. Neither animal has any adaptation against predation. Galapagos

tortoises use their long necks to reach under cactuses and spiny desert shrubs

to obtain fruits (Fowler 1983). Similar habitats on Grand Turk would suggest a

similar dietary adaptation. This collection of Geochelone remains from Grand

Turk has contributed greatly to our knowledge of the structure of this

phylogenetically unique species.








Birds

Birds make up a small portion of the Taino subsistence at Coralie, yet

this assemblage is quite extensive and diverse when compared with bird bones

found at other Amerindian sites excavated to date (deFrance 1989; Righter

n.d.; Versteeg and Schinkel 1992; Wing and Scudder 1980, 1983). The total

number of bird bones recovered was 296, yielding 47 individuals; 62% of these

remains could be identified to species. Out of these 20 species, 15 are

represented by just one or two individuals. Dominating the assemblage are two

species of booby (Sula sula and Sula dactylatra), one species of duck

(Dendrocygna arborea), and one species of crow (Corvus nasicus), which

together make up 50% of the individuals recovered. The collection includes 12

species of seabird (birds that feed over open ocean) and eight of landbird. A

complete description of the avifauna found at Coralie is provided in Appendix

B. This includes species systematics, material recovered, and remarks

concerning habitat preference, size, and present geographic distributions. A

few of the birds identified in this collection were not previously known from this

region, and inform us about varying past distributions.

The thick-knee is a common Old World genus with only two species

occupying the New World; one lives in the Peruvian Andes and the other in

northern South American, Central America and on Hispaniola (Blake 1977).

This latter species, the double-striped thick-knee (Burfhinus bistriatus), was

identified on Grand Turk. There are no other Late Holocene period records for

this species in the West Indies outside of Hispaniola. Pregill and Olson (1981)

have presented paleontological records for thick-knee in the Bahamas from San