I29.86:M57/V.

1
Clemson

University

:e

Cultural Resources

idies

3 1604 019 571 241

Willie ui rvicneulogy
and Historic Preservation

Evaluation

National Park Service

U.S. Department
of the Interior

Washington 1977

of the

Northern Gulf of Mexico

Continental Shelf
Volume

Prehistoric Cultural Resource Potential

Digitized by the Internet Archive

in

2012 with funding from

LYRASIS Members and Sloan Foundation

http://archive.org/details/culturalresourceOOgulf

Cultural Resource

Management Studies

CULTURAL RESOURCES EVALUATION

OF THE
NORTHERN GULF OF MEXICO
CONTINENTAL SHELF

Volume

Prehistoric Cultural Resource Potential

prepared for
Interagency Archeological Services
Office of Archeology and Historic Preservation
National Park Service
U.S. Department of the Interior
Washington, D.C.

by

COASTAL ENVIRONMENTS,
BATON. ROUGE.
504-383-7455

LA..

70802

1977

INC.

Foreword

One of the basic defects in the implementation of historic preservation

programs today is the absence of comprehensive plans which provide
frameworks for decisionmaking and conflict resolution.
Relatively few
efforts have been made to develop prototypes of such frameworks but they
are essential because they offer the only real solution to the seemingly
endless arguments over significance, inventory priorities and selections
of preservation and mitigation options. The Interagency Archeological
Services Division has been contributing to the development of such
prototypes through the funding of feasibility studies on various aspects
of historic preservation planning issues; some of these studies have
previously been published in this series.

In 1974, the Division began to deal with the very specialized problems of
protection of cultural resources on the outer continental shelf in
conjunction with the oil and gas leasing activities of the Bureau of Land
Management and the U.S. Geological Survey. Given the fact that lessess
were required by the conditions of their leases to conduct archeological
surveys, should these be done everwhere on the shelf? Our conclusion was
that surveys were not always necessary, especially in consideration of the
practical problems of submerged site identification, provided a large-scale
review of site occurence probabilities was undertaken.
Because the basic
problem was identical, although somewhat less complex, to that existing
on land, we launched the pilot study reported in this three-volume study
prepared by Dr. Sherwood M. Gagliano and his associates at Coastal
Environments, Incorporated of Baton Rouge, Louisiana.

The basic premise was that submerged archeological sites were not randomly
scattered about the sea bottom.
Instead, the prehistoric sites could be
expected to occur in a manner related to the paleogeography of the
continental shelf and shipwrecks could be expected to occur in relation
to present and past ports, sea routes, and hazards to navigation.
Dr.
Gagliano' s first task was to reconstruct the structure of this larger
setting and to block out the major regions in which archeological sites
could be expected to occur.
He then considered the types of sites likely
to be found and their discoverability.

This report goes far toward defining regions in which there is concern for
site presence and toward defining the objects of search. There should be
no illusions however, that this report represents the last word. A great
deal must still be accomplished in further refining our knowledge about
which areas have archeological potential and which should be the focus of
attentions for resource management. We also need much research and
development for site discovery methods. Tolerable techniques exist for
finding shipwrecks, provided they are used properly, but the available
techniques for finding aboriginal sites are primitive.

Nevertheless, we have in Dr. Gagliano's report a first generation

decisionmaking framework useful for triggering the implementation of
mineral lease archeological survey stipulations in those areas where this
seems to be needed and deleting this requirement from other areas resulting
It also
in substantial survey cost savings to the oil and gas industry.
is potentially useful when other types of land management decisions and
setting of priorities are made. Many of these concepts are transferable to
the terrestrial setting and studies funded by Interagency Archeological
Because the conduct
Services are presently exploring such applications.
of these studies is truly an investment in a significantly more costeffective future historic preservation program we encourage any comments or
suggestions on our efforts.

Rex L. Wilson
Departmental Consulting Archeologist
and Chief, Interagency Archeological
Services Division,
Office of Archeology and Historic
Preservation

ACKNOWLEDGMENTS

We wish to thank the following individuals for invaluable contributions

of information or constructive discussion which made the completion of
this volume possible:

Lawrence Aten, National Park Service, Washington, D.C.;

Michael Blake, Montrose, Alabama; Charles Bollich, Beaumont, Texas;

Jack

C.

Bonnin, Welsh, Louisiana; Marvin Buller, Rosenberg, Texas; the

late Ripley Bullen, Gainsville, Florida; George Carter, Texas A&M University;

Carl Clausen, North Port, Florida;

Wilburn Cockrell, Florida Division

of Archives, History, and Records, Tallahassee, Florida;

University of Alabama;

Cailup B. Curren,

Douglas Elvers, Bureau of Land Management,

New Orleans, Louisiana; Joe Frank, Lake Charles, Louisiana; Frank Fryman,
Florida Department of Archives and History, Tallahassee, Florida;
Jon Gibson, University of Southwestern Louisiana;

Steve Gluckman, Division

of Archives and History, Raleigh, N.C.; James Haisten, Panama City,

Florida;

J.

H.

Herbert, Houston, Texas;

Nick Holmes, Mobile, Alabama;

Mrs. Hubbard, Panama City, Florida; Jack Hudson, Houston, Texas; Kay Hudson,

Houston, Texas;
Florida;

Yulee Lazrus, Temple Mound Museum, Fort Walton Beach,

William McClure, Houston, Texas;

James Miller, Florida

Department of Archives and History, Tallahassee, Florida;

Charles Nelson,

Bureau of Land Management, New Orleans, Louisiana; Erwin Otvos, Ocean Springs,
Mississippi; George Percy, Florida Department of Archives and History,
Tallahassee, Florida;
W.

Dave Phelps, East Carolina State' University;

Armstrong Price, Corpus Christi, Texas; Reynold Ruppe

Arizona;
U.S.

Alan Saltus, Baton Rouge, Louisiana;

Geological Survey, New Orleans, Louisiana;

of Land Management, New Orleans, Louisiana;

of Engineers, Nashville, Tennessee;

Austin;

University of

Richard Scrivener,
Harold Sieverding, Bureau

Brent Smith, U.S. Corps

Dee Ann Story, University of Texas,

Wallace Stroud, Elton, Louisiana;

Mr.

and Mrs. Donald Totman,

Apalachicola, Florida;

Benward L. Treadaway, Jr., Violet, Louisiana;

J.L. van Beek, Coastal Environments, Inc., Baton Rouge, Louisiana;

Tom Watson, Panama City, Florida.

The director of the project and principal author of
Volume

was Sherwood M. Gagliano.

William

wrote the sections on Texas in Chapter

G.

Smith compiled and

and also most of Chapter

Eileen Burden and Richard Weinstein co-authored Chapter

related plates in Volume III.

and the

Rod Emmer assisted in the research of

the Alabama and Florida areas of Chapter

3.

Kathi Brooks, Lynne Hair, and

Peggy King helped with research and manuscript typing.

was done by Bunny Shannon.

4.

Final typing

Cartography was done by Curtis Latiolais.

Editing was done by Ava Leave 11 Haymon.

li

ABSTRACT FOR VOLUMES

I,

II, AND III

This evaluation of the cultural resource potential of the outer conti-

nental shelf, Northern Gulf of Mexico, was generated in response to a significant

increase in mineral extraction activities in the Gulf

and to a growing

awareness of the nonrenewable nature of these cultural resources.

The strategy developed for the study basically involved literary research
and synthesis of the archeological, historical, geological, and technical

parameters related to the identification and management of these resources.

The study and interpretation of Quaternary sea level fluctuations relating
to the geomorphology was a central concept utilized in making recommendations

for improved accuracy in identifying submerged archeological sites.

This

concept generated a testable model which helped to decrease the randomness

involved in predicting site occurrences on the continental shelf.

Archeological files of coastal states and knowledgeable amateurs in the

northern Gulf area were consulted in order to assess the nature and distribution of known sites in the present day coastal zone and to relate them to

formerly active processes.

This resulted in a graphic representation of these

sites by culture period and physiographic context.

A literary search was also conducted to document all known historic

shipping lanes, shipwrecks, and ship losses in the study area.

Shipping

lanes were mapped on four separate plates, reflecting the division of historic

shipping in the Gulf into four time periods.

Further formulation of recommendations for underwater site detection came

from an evaluation of the present level of development of geophysical surveying techniques, and the effectiveness of these techniques in locating various
types of sites.

iii

A summary product of the study was manifest in the development of

a zonation map, dividing the OCS into five areas ranked in priority for

probability of occurrence of shipwrecks and drowned terrestrial sites.

iv

ABSTRACT FOR VOLUME

This is a study of the predictability of drowned prehistoric habitation

sites in the continental shelf area, northern Gulf of Mexico, from the Rio

Grande River to the Florida Keys.

Because of the difficulties of obtaining

data concerning the location of a submerged site, an indirect approach was

formulated incorporating the limitations of the detection devices that are

available.

Geometric models for frequently occurring coastal physiographic features

are developed in order that forms on the Outer Continental Shelf (OCS) can

be identified and classified as relicts of specific, once-active physio-

graphic units.

The OCS is explored area by area, west to east, using the

published descriptions and maps that are available, in order to map the
shelf and to identify important relict forms with the past-active systems
that formed them before they were submerged.

Sea level in the Late

Quaternary Period is traced, considering the eustatic, isostatic, and

tectonic changes that make more land available for habitation at some
times than others.

The sea level determines the coastline - the sea-

ward limit of our study area for any given period of prehistoric time.

A geological history is developed of this fluctuating study area.

A method is presented of forming hypotheses about the nature of the

archeological possibilities of the OCS - hypotheses that can be tested with

the limited sort of data that can presently be gathered from the OCS.

method is this:

The

the OCS will be divided into Eastern, Central, and Western

Gulf areas, corresponding to the adjacent areas on land.

The archeological

literature of the land areas will be reviewed to identify major cultural

These can be predicted to have occurred

manifestations, by time and by type.

similarly on the OCS in the time periods when and where it was exposed concurrently.

These cultural manifestations are examined for the purpose of

making tables of index artifacts, environmental-use models, and particularly

landforms favored for habitation sites.

Then, addressing the problem of

increasing one's chances in site propecting on the OCS:

the landforms

(detectable, as relicts) that are most frequently favored at any period

are assigned a list of "signatures" - discrete site indicators that are

capable of being detected by the limited sensing tools and techniques

available for OCS survey.

An inventory is made of the known sites in the

Northern Gulf area that were occupied from 55,000 B.P. to 3,500 B.P.
Typical sites from three regions, Eastern, Central, and Western Gulf Coast,
are selected for Pre-projectile Point, Paleo-Indian, Archaic, and Poverty

Point Periods.

Age, ecofacts, artifacts, and associated landforms of these

typical sites are discussed.

The methodology developed in this study is

illustrated with a case study of the Mississippi Delta area.

Last, lists of signatures are presented for the types of frequently

occurring sites, the effectiveness of remote-sensing in identifying types

of sites is discussed, and the most effective sequential approach to pre-

historic site discovery is outlined.

VI

TABLE OF CONTENTS

ACKNOWLEDGMENTS

ABSTRACT FOR VOLUMES

ABSTRACT FOR VOLUME

I,

II, AND III

iii

LIST OF FIGURES

xi

LIST OF TABLES

xx

CHAPTER

I.

CHAPTER II.

INTRODUCTION

THE RELATIONSHIP OF PROCESS TO FORM IN THE

COASTAL ZONE

Distribution of Coastal Systems in the Shore Zone of

the Northern Gulf of Mexico

11

Mappable Geological Features

16

Uplands
Coastal and Riverine Terraces
Beaches and Barrier Complexes
Meander Plains
Deltaic Plains
Undifferentiated
Active Coastal Systems
Continental Shelf Features
Barrier Spit Complexes
Beach Ridge Trends
Barrier Accretion Forms
Shelf-Edge Bulges
Shore Trends
Escarpments
Tidal Scour Features
Banks, Shoals, and Shelf-Edge Knolls
Terraces
Karst Areas
Rocks
Channels
Entrenched River Axes
Delta Lobes

16
17
17
18
19
20
20
20
20
23
23
24
24
24
25
26
27
28
28
28
28
29

Surface Sediments of the Shelf

29

Changes in Levels of Land and Sea

30

CHAPTER III.

LATE QUATERNARY RELICT FORMS

38

South Texas Area

39

Central Texas Area

49

East Texas Area

83

vn

West Louisiana Area

95

East Louisiana Area

99

Mississippi-Alabama-West Florida Area

110

West-Central Florida Area

120

Central Florida Area

125

South Florida Area

131

CHAPTER IV.

SEA LEVEL IN THE LATE QUATERNARY PERIOD

148

Introduction

148

Fluctuations of Sea Level

148

Paleoclimatology and the Deep-Sea Period

150

Geological History

155

CHAPTER V.

Interval A
Interval B
Interval C
Interval D
Intervals E and F
Interval G
Interval H
Intervals I, J, and K

156
160
162
166
166
167
167
169

ARCHEOLOGICAL METHOD AND THEORY

170

Culture Areas

170

Early Man

173

Stage
Stage
Stage
Stage

Core Tool Tradition

Flake/Bone Tool Tradition
III: Blade, Burin, and Leaf-Point Tradition.
IV: Specialized Point Tradition
I:

II:

174
176
176
176

Landform Associations

177

Ecofacts

186

Archeological Sequence and Sea Level Fluctuations

189

CHAPTER VI.

SELECTED TYPICAL ARCHEOLOGICAL SITES OF THE

NORTHERN GULF

193

Site Inventory and Dating

193

Selected Sites

197

Vlll

Western
Western
Western
Central
Central
Central
Central
Eastern
Eastern
Eastern
Eastern

Gulf
Gulf
Gulf
Gulf
Gulf
Gulf
Gulf
Gulf
Gulf
Gulf
Gulf

Pre-Projectile Point
Paleo-Indian
Archaic
Pre-Projectile Point
Paleo-Indian
Archaic
Poverty Point
Pre-Projectile Point
Paleo-Indian
Archaic
Poverty Point

AN ILLUSTRATION OF METHODOLOGY:
DELTA AREA

CHAPTER VII.

,...,,....

THE MISSISSIPPI
300

...........
Mississippi River Delta ..,.,,.,

Introduction
The

300

301

Deltaic Plain
Marginal Plain
Marginal Basin
Alluvial Valley
Terraces and Uplands West of the Alluvial Valley
Terraces and Uplands East of the Alluvial Valley
,

Prehistoric Land Use

198
204
209
213
232
237
255
267
271
292
296

312

Lafayette Complex: Paleo-Indian to Early Archaic,

Maringouin Comples: Early to Middle Archaic
Sale-Cypremort Lobe of the Teche Complex: Middle to
Late Archaic
Metairie Lobe of the St. Bernard Complex: Poverty
Point and Tchefuncte Period
,

.....

Summary and Conclusions

307
309
310
310
311
312

314
324

326
327

329

CHAPTER VIII. CONCLUSIONS AND RECOMMENDATIONS FOR FURTHER

STUDY

332

Summary

331

Cultural Signatures

332

Quarry Sites
Salt Dome Sites
Spring Sites
Valley Margin Sites
Natural Levee Sites
Point Bar Sites
Bay Margin Sites
Coastal Dune Lake Sites
Shell Middens
Conical Earth Mounds
Crescentic and Circular Villages
ix

332
332
334
334
335
335
336
336
337
337

338

Remote-Sensing Techniques

339

Zone Map

3*1

Recommendations for Further Study

342

REFERENCES

344

LIST OF FIGURES
Figure No.
1-1.

2-1.

2-2.

2-3.

2-4.

2-5.

2-6.

2-7.

Page No

Physiographic divisions and bathymetry of the

northern Gulf of Mexico (Modified from Bergantino,
1971, and Brooks, 1974)

Typical distribution and surface relationships

of deltaic and related physiographic or environmental units.
Scale approximate (After Coleman
and Gagliano, 1965)

Idealized barrier spit and bay system showing

typical arrangement of physiographic units

Delta model indicating typical arrangement of

major components

10

Some major process parameters of the northern Gulf

of Mexico

13

Diagrammatic representation of shore zone environments

14

Idealized relationships between active and past

systems and shore zones

15

Pleistocene barrier island and associated strandplain sands, Smith Point area, Chambers County,
southeast Texas (After Fisher et_ al.
1973)

18

Pleistocene meander plain landscape in the vicinity of Devers, Beaumont-Port Arthur area, Texas
(After Fisher et al., 1973)

19

Features associated with tidal inlets between barrier islands (After LeBlanc, 1972)

22

Modern barrier island environments and facies,

Galveston Island (Cross-section after Bernard e_t
al
1972)
1970; After Fisher et al.

23

Features associated with low-cliff coast developed

in limestone

25

2-8.

2-9.

2-10.

2-11.

2-12.

2-13.

2-14.

Surface sediment distribution of the continental

shelf, northern Gulf of Mexico (After Curray, 1975,
1965)

31

Principal faults and areas of salt dome occurrences

in the northern Gulf of Mexico

33

Geologic cross-section showing major sedimentary

units and effects of normal faulting, southern Louisiana (After Jones, 1969)

34

XI

LIST OF FIGURES (Continued)

Figure No
2-15

2-16.

3-1.

3-2.

Page No

Generalized cross-section of Gulf Coast Geosyncline depicting components of apparent sea level
rise (After Kolb and van Lopik, 1958)

35

Major tectonic features and Late Quaternary movements in the northern Gulf of Mexico

36

Transverse cross-section from the southern Laguna

Madre near Port Mansfield illustrating the preHolocene surface as evidenced by shallow borings
(From Rusnak, 1960)

47

Cross-section of Padre Island, South Bird Island,

7.5-Minute Quadrangle (From Hunter and Dickinson,
1970)

3-3.

55

Section across Ingleside strandplain sand, Northern

Laguna Madre and Southern Mustang Island (From
Wilkinson et al.
1975)

64

Section across Ingleside strandplain sand, Mesquite

Bay and Southern Matagorda Island (From Wilkinson
et al. , 1975)

66

Pleistocene sedimentation in Matagorda Island area

(After Wilkinson, 1975)

73

One- cubic-inch, air-gun profile of the West Flower

Garden Bank (WFGB) (After Edwards, 1971)

87

Interpretive illustration of the West Flower Garden

Bank when sea level was about 121-134 m below
present level (After Edwards, 1971)

89

Rock outcrops off of Freeport, Texas, at depths

(After
ranging from -14 to -20 m below sea level
Winchester, 1971 )

92

3.5kHz sub-bottom profiles of Sweet Bank and Bank

3.
Major and minor terrace levels can be distinguished.
Gas seeps along the flanks are believed to be associated with structural activity
(After Poag, 1973 )

96

3-4.

3-5.

3-6.

3-7.

3-8.

3-9.

3-10.

Late Holocene deltaic area of the Mississippi River

(Modified from Gould and Morgan, 1962 ).
system
.

3-11.

Generalized cross-section through Late Holocene

deltaic plain of the Mississippi River
(After Fisk
and McFarlan, 1955 )

xii

100

101

LIST OF FIGURES (Continued)

Figure No
3-12.

3-13.

3-14.

3-15.

3-16.

3-17.

3-18.

3-19.

Page No

Downwarp of Prairie surface beneath Late Quaternary

Mississippi deltaic mass as determined from bore(After Fisk and McFarlan, 1955 )
hole data

103

Section through southeastern Louisiana illustrating

offlapping pools of coastal and deltaic sediment
(After Frazier, 1974 )

104

Depositional sequence shown by borings at South Pass

(After Morgan, Coleman, and Gagliano, 1968a and
1968b.)

105

Section through St. Bernard delta complex (After

Frazier, 1974 )

107

Pontchartrain Basin area cross-sections showing

weathered horizons (After Saucier, 1977)

108

North-south cross-section through the New Orleans East

area showing near-surface stratigraphy

109

Interpretation of relict topography in the Alabama-West

Florida area

Ill

Profiles and interpretation of relict topography in the

Alabama-West Florida area

112

3-20.

Bathymetric chart of lower pinnacle zone south of Mobile

Bay, Alabama (After Ludwick and Walton, 1957)
113

3-21.

Quaternary geological features of the Mississippi

Gulf Coast area

117

3-22.

Generalized cross-section through the central Mississippi

Gulf Coast area (Modified from Otvos, 1972 )
118

3-23.

Terraces of the Pascagoula River area (After Saucier, 1977)119

3-24.

Surface geology of the Cape San Bias to Alligator

Harbor area, Florida (After Schnabel and Goodell,1968

3-25.

121

Bathymetric chart of shelf-edge bulge and related

features south of Panama City, Florida (After Jordan, 195D 122
1

3-26.

3-27.

3-28.

Bathymetric profiles of shelf-edge bulges and related

features (After Jordan, 1951)

123

Bathymetric profiles of shelf-edge bulges and related

features (After Jordan, 1951 )

124

Cross-section from Cape San Bias to Cat Point, Florida

(After Schnabel and Goodell, 1968 )

126

Xlll

LIST OF FIGURES (Continued)

Figure No

Page No

3-29.

Major geomorphic features of the Central Florida area.

128

3-30.

Major geomorphic features of the south Florida area.

132

3-31.

Bathymetric chart of the Howell Hook area, south

Florida (After Jordan and Stewart, 1959 )

133

Idealized cross-section of coastal mangrove swamps in

the Ten Thousand Island area of the south Florida coast
(After Scholl and Stuvier, 1967 )

136

Cape Sable beach and relict shoreline features (After

1964)
Smith, 1968, and after Spademan et al.

138

3-32.

3-33.

3-34.

xhe Florida Keys and their environment (After Hoffmeister

140
et al.
1964 )
,

3-35.

Development of a fringing reef (From Smith, 1971.)

3-3-

(a)

141

Map of Florida
( B ) Cross-section of Florida along
the north-south line of (A) (Modified from Grabau,
f

1960)
3-37.
4-1.

The Lower Keys

143

(From Ginsburg,

1964.)

147

100 - 110 m terrace and drowned barrier reef on Campeche

165
(From Lindsay et al. , 1975 )

Shelf
5-1.

5-2.

Culture areas and subareas of the northern Gulf

region (After Willey, 1966 )

171

Chronology of early man sites and traditions (After

MacNeish, 1972; 1976 )

175

5-3.

Distribution of initial occupation sites in a prograding

beach sequence
179

5-4.

Distribution of initial occupation sites in a lobate

delta

180

Environmental succession of an idealized delta cycle

(Modified from Gagliano and van Beek, 1975)

182

Distribution of initial occupation sites associated

with a coastal plain stream system

183

Initial occupation sites associated with sequentially

developed, recurved spit complexes

184

Initial occupation sites and reworked material on

truncated shorelines

184

5--5

5-6.

5-7.

5-8.

xiv

LIST OF FIGURES (Continuted)

Pa fi G No.

Figure No.
5-9.

5-10.

5-11.

5-12.

Initial occupation sites and reworked material in a

transgressive delta

186

Changes in shell content of middens in a hypothetical

coastal estuary resulting from sea level fluctuations.

188

Shell mound on a subsided natural levee ridge enveloped by marsh mud (After Russell, 1967)

190

Hypothetical sequence of landform development during

a "ria cycle."

191

5-13.

Idealized sea level fluctuations of a "ria cycle."

6-1.

Steep, end-retouched artifact found beneath sabertooth cat at Friesenhahn Cave (After Sellards, 1952)

6-2.

6-3.

6-4.

6-5.

6-8.

6-9.

6-10.

192

200

202

Cross-section along south wall of Trench 1, Friesenhahn Cave showing stratigraphic relation of the several units of fill (After Evans, 1961)

203

Relationship of Berclair Terrace to older Tertiary

Goliad Formation, Pleistocene Lissie Formation, Beaumont Formation and the Late Holocene floodplain deposits (After Sellards, 1940)

205

Horizontal and vertical distribution of artifacts at

the Buckner Site (41 BE 2), showing stratified zones
of gravel, sand, and silt (After Sellards, 1940)
.

206

The McFaddin Beach Site

a beach deposit that has
produced Paleo-Indian projectile points and bones
from extinct Pleistocene vertebrates (Base map after
Fisher et al.
1973)

210

Location and physiography of the Jamison Site, 41

LB 2 (After Aten, 1967)

212

Idealized geologic section in vicinity of Natchez,

Mississippi, showing setting of Natchez Pelvis find
(Geology modified from Saucier, 1967)

215

Fossil locale on Tunica Bayou, West Feliciana Parish,

Louisiana

216

Loess deposits in roadcut along U.S. Highway 61 near

Vicksburg, Mississippi

216

6-7.

Floor plan of Friesenhahn Cave showing trenches and

articulated skeletons (After Evans, 1961)

6-6.

xv

LIST OF FIGURES (Continued)

Figure No
6-11.

Page No

Block diagram illustrating major physiographic features of south-central Louisiana (After van Lopik,
219

1955)
6-12.

6-13.

6-14.

6-15.

6-16.

6-17.

6-18.

6-19.

6-20.

6-21.

6-22.

6-23.

Four idealized phases of the geological history of

the Salt Mine Valley Site (16 IB 23) showing stream
cutting and valley filling (After Gagliano, 1970).

220

Salt Mine Valley (16 IB 23) showing locations of core

holes, excavations, and relative age of surface features

223

Idealized cross-section through Salt Mine Valley (16

IB 23) (After Gagliano, 1970)

224

Characteristics of bipolar cores from Pit V, Salt

Mine Valley (16 IB 23) (After Gagliano, 1967)

226

Bipolar artifacts from Pit V, Salt Mine Valley Site

(16 IB 23) (After Gagliano, 1967)

227

Steep, edge-chipped artifacts from the New Mine Shaft,

Salt Mine Valley

229

Steep, edge-chipped artifacts from the New Mine Shaft,

Salt Mine Valley

229

Fragment of split-cane basketry from New Mine Shaft,

-2.5 to -2.8 meters MGL, Salt Mine Valley Site (16
IB 23)

230

Pieces of three-strand cordage from New Mine Shaft,

-1.89 to -2.23 meters MGL, Salt Mine Valley Site ...

230

Socketed bone projectile point from New Mine Shaft,

-2.5 to -2.8 meters MGL, Salt Mine Valley Site

....

231

Cut wood from New Mine Shaft, -2.5 to -2.8 meters MGL,
Salt Mine Valley Site

231

Cores, bladelets, and bifacial tools from Salt Mine

Valley (16 IB 23) (After Gagliano, 1967)

234

6-24.

Projectile points from Bayou Grand Louis (16 EV 4)

6-25.

Artifacts from the Palmer Site (16 EBR 13)

6-26.

Site distribution and morphological relationship in

the middle Amite River area (After Gagliano, 1963.

6-27.

236
236

238

Generalized cross-section of the Amite River Valley,

showing si+e- terrace relationships (After Gagliano,
239

1963)

xvi

LIST OF FIGURES (Continued)

Figure No

Page No

Stratigraphy of high floodplain exposed at Williams

Gravel Pit (16 EBR "A") (After Gagliano, 1963)

239

6-29.

Amite River Phase projectile points

244

6-30.

Amite River Phase artifacts

244

6-31.

Map of Monte Sano Mounds (16 EBR 17)

245

6-32.

Cross-section through Mound A of the Monte Sano Site

6-33.

The Copell Site (16 VM 102) on Pecan Island, Louisi-

6-28.

....

246

248

ana.

6-34.

Site distributions and morphological relationships in

the Pearl River mouth area (From Gagliano, 1963)
.

251

The Cedar land Plantation (22 HC 30) and Claiborne

(22 HC 35) Sites, showing midden concentrations, test
pits, and stratigraphy and radiocarbon dates as revealed by cross-sections and profile of the east wall
of Pit E (After Gagliano and Webb, 1970)

252

6-36.

Stratigraphic view of Cedarland Site

253

6-37.

Cross-sectional exposure of clay-lined hearth in black

earth midden in Late Archaic Cedarland Site (22 HC 30),
253
Hancock Co. , Mississippi

6-38.

Artifacts from the Cedarland Site. (After Gagliano and

Webb, 1970)

254

Poverty Point Objects from the Claiborne Site (22 HC

30), Hancock County, Mississippi

258

6-40.

Selected artifacts from the Claiborne Site

258

6-41.

Base of a Wheeler Punctated, fiber-tempered vessel

from Claiborne

260

Figure of the five steatite bowls, found as a group, at

the Claiborne Site

260

Location, borehole stratigraphy, and radiocarbon assay

of the Linsley Site (16 OR 40)
a site of the Bayou
Jasmine Phase of the Poverty Point Period

262

Bayou Jasmine Phase artifacts from the Linsley Site

(16 OR 40)

263

Microliths and other artifacts from the Garcia Site

(16 OR 34)

263

6-35.

6-39.

6-42.

6-43.

6-44.

6-45.

xvii

LIST OF FIGURES (Continued)

Pa e No-

figure Ho.

Debitage in cultivated field at Skelly Site near Dothan, Alabama

268

Stratified chipping floors sloping into old quarry

pits at Skelly Site near Dothan, Alabama

268

6-48.

Skelly Site artifacts

270

6-49.

Skelly Site artifacts

270

6-50.

Map of Vero area showing canal from which human fossil remains have been found (From Sellards, 1917a)

6-46.

6-47.

6-51.

6-52.

6-53.

6-54.

6-55.

6-56.

6-57.

6-58.

6-59.

7-1.

271

East-west cross-section through fossil deposits at

Vero (Modified from Weigel, 1962)

272

Map of Warm Mineral Springs, showing position of

Clausen's test pit in relation to surrounding sinkhole
walls and modern buildings (After Clausen, Brooks, and
Wesolowsky, 1975)

280

Generalized cross-section through the wall of Warm

Mineral Springs (After Clausen, Brooks, and Wesolowsky, 1975)

281

Paleo-Indian projectile point finds around Choctawhatchee Bay, Florida and vicinity

288

Artifacts from Point Washington,

31, Choctawha tehee Bay, Florida.)

WL "B" (near 8 WL

The Alligator Point Site (8BY "C")

entrance to St. Andrews Bay

289
,

located along the

291

Freshwater pond in coastal dune field between Choctawha tehee and St. Andrews Bays, Florida

291

Location of the Lake Kanapaha Site (8 AL 172) and surrounding lithic and ceramic sites (After Hemmings and
Kohler, 1974)

294

Elliot's Point Complex clay balls, similar to Poverty

Point Objects found in Louisiana and Mississippi.

298

6-60.

Paleo-Indian and Archaic projectile points from Choctawhatchee Bay area

298

Major features of south Louisiana and south Mississippi area and the regional setting of the Mississippi
Delta System

302

XVlll

LIST OF FIGURES (Continued)

Figure Ho .
7-2.

Page No.

Block diagrams illustrating progradation and transgression in a delta with a bifurcating branching habit, such as the Mississippi (After Frazier, 1967).
.

303

Diagrammatic representation of the relationship between major morphologic features and sedimentary fades in an advanced stage of delta building (After
Frazier, 1967)

304

Distribution of known archeological sites in coastal

Louisiana

306

7-5.

Cross-section of natural levee and backswamp

308

7-6.

Major delta complex (upper case) and lobes (lower case)

of the Mississippi River and prehistoric archeological
sites for the interval 12,000 to 2,100 years before
315
present (After Gagliano, Weinstein, and Burden, 1975)

7-3.

7-4.

7-7.

7-8.

7-9.

Major delta lobes of the Mississippi River and prehistoric archeological sites, for the interval from
2,500 to 1,700 years ago (After Gagliano, Weinstein,
and Burden, 1975)

316

Major delta complexes

(upper case) and lobes (lower
case) of the Mississippi River and archeological
sites for the interval from 1,700 years ago to present
(After Gagliano, Weinstein, and Burden, 1975)

317

Late Quaternary chronology of Mississippi Delta complexes and selected lobes

318

7-10.

Lafayette Delta Complex - 12,000 to 8,500 years B.P.

7-11.

Reconstructed paleogeography of a part of the Lafayette meander belt (circa 8,500 years B.P.) showing
related Paleo-Indian sites and vertebrate fossil locales

7-12.

Maringouin Delta Complex - 8,500 to 6,000 years B.P.

7-13.

Metairie lobe of the St. Bernard Delta Complex

to 2,000 years B.P

xix

319

321
.

325

4,000
328

LIST OF TABLES
Table No
6-1.

6-2.

Page No

Radiocarbon ages of Vicksburg, Mississippi, loess deposits (After Snowden and Priddy, 1968)

217

Possible relationships of the three zones, located

on the 13 meter ledge at Warm Mineral Springs, and
their age of deposition, period, and interval

284

8-1.

Occurrence of site types by culture period or stage.

8-2.

Effectiveness of remote-sensing survey and testing

tools

XX

333

340

CHAPTER

INTRODUCTION

Eighteen thousand years ago, during the time of the Woodfordian

glacial stage, sea level in the Gulf was 121 meters lower than it is today,

making the coastline of the Northern Gulf of Mexico as much as 100

kilometers farther out in some places.

At that time, sea level had been

making a long and uneven decline from the period of the Ingleside
Shoreline, 55,000 B.P., when it was a little higher than its present
stand.

During every period, oblivious to eustatic change, rivers and

streams wound their way to the Gulf, delivered their sediment loads and
built their deltas.

Cut-bank and swale formed, as did point bars and natural

levees, terraces and lagoons.

These geomorphic features gradually appeared

and gradually disintegrated, sometimes leaving relicts, sometimes not.

Sometime during this 55,000 years, prehistoric man and woman entered
the coastal zone.

From the landforms available to them, these people chose

the places where they lived and worked.

Skills gradually evolved.

moved about with the changing coastal zone.

They

Like the geomorphic features,

sometimes they left a record of their existence, and sometimes they did not.
If they lived on the area of the shelf that is outside the present coastline,

the rising water slowly drowned their habitation sites.

a stillstand at its present level at about 3,500 B.P.

Sea level reached

The possibilities for

drowned habitation sites on the Gulf shelf, then, is from 55,000 to 3,500 B.P.

When the Europeans arrived in ships in the 1500

approximately at its present stand.

s,

sea level was

The Gulf was quickly criss-crossed with

shipping lanes as the Spanish, French, English, and then the young United States

and other countries used the Gulf as an area of transportation.

The shipwrecks

that inevitably occurred are a kind of archeological byproduct of all this

commerce.

Shipwrecks litter the bottom of the Gulf, some of them buried by

sediment.

By a historical and geological coincidence caused by the rising

of sea water, the shipwrecks may overlie the buried and drowned habitation

sites.
In the wake of the recent energy crisis has come accelerated explora-

tion and development of the mineral resources of the outer continental shelf
(OCS)

in the Gulf of Mexico, as has been true along the California coast,

Atlantic coast, and the Alaskan coast.

This increased extraction of the

mineral resources has raised the specter of irreversible damage and loss to
the cultural resources of the OCS, the drowned habitation sites of indigenous

peoples and shipwrecks from the historical period.

In recognition of possible damage to environmental and cultural re-

sources resulting from the activities of the mineral extraction industries,

the U.S. Department of the Interior has initiated several study efforts to

inventory and evaluate the OCS environment, and in 1974 it implemented a requirement for marine archeological surveys of all leased drilling sites. The

requirement stipulated that all drilling sites and pipeline rights-of-way

be surveyed to determine absence or presence of submerged habitation sites

and/or shipwrecks.

Minimum requirements called for a geophysical survey

using a total field intensity magnetometer towed above the sea bed, dual

side-scan sonar coverage of the sea floor at a range width of about 500
feet per side, along with depth sounder and sub-bottom profiler runs.

It

was recommended that the profiler be capable of resolving the upper 50 feet

of sediment.
quired.

Navigation accuracy of + 50 feet at 200 miles was also re-

Recommended optional tools included cameras, divers, and cores.

It was advised that engineering soil borings be available for archeolo-

gists* inspection.
grid.

Survey line spacing was to follow a prescribed

line

Although these requirements represented an important initial step

in the protection of cultural resources, the surveys proved to be costly

and often resulted in delays in drilling and pipeline construction starts.

Furthermore, it was not clear whether the surveys would fulfill the desired objective of identifying endangered cultural resources.
It soon became apparent that a broad study was needed to more clearly

define the nature and extent of cultural resources on the continental shelf
and to outline approaches to their management.
evolved.

Thus, the present study

Its purpose was to determine the archeological potential, to

extablish guidelines for survey priorities and level of effort, and to

develop a rational framework for continental shelf archeology,

The area

of interest included the entire continental shelf of the northern Gulf

(Figure 1-1), from the Rio Grande to the Florida Keys, and from the mean

water line (MGL) on the Gulf beaches to -160 meters, with some preliminary
evaluation of bottom conditions and sedimentation patterns affecting dis-

coverability of shipwreck sites for the zone between water depths of -160
and -600 m MGL.

This order of study was followed:

first, in geology, a synthesis was

made of the literature and data pertaining to Late Quaternary geology of the
shelf and coastal zone.

The morphology of each coastal system, shelf sedi-

mentation, tectonics, and sea level fluctuations were considered.

a study of prehistoric archeology was undertaken.

Second,

A synthesis was essential

of the literature and data pertaining to coastal zone prehistory by culture

period for each culture area.

Emphasis was placed on site morphology and content,

o
a
H

c
CO
toO

u
0)

PQ

o
S-i

T3
CD

H
M-l

o
CJ

a
M-l

o
p

c
CD

O
d

CO

^5
13
CO

CO

C
o
H
CO

H
H>
X)

o
H
43
P.
CO

T3

M C
toO CO
O

H
W .H

>^N

J2 C^
PM i-H

index artifacts, and characteristic artifact assemblages.

Some

consideration has been given to the "Pleistocene megafauna" since the

remains of these large vertebrates are often associated with early man
and provide important environmental indicators.

Third came a study of shipwrecks.

A review was made of the

literature describing types of ships used in the Gulf from colonial

exploration and settlement through World War II (1500 - 1945)

Types

and characteristics of ships were determined by period as they relate

to discoverability.

A review was made of charts and maps, literature

and archival records, and other data pertaining to sailing routes,

along with a systematic analysis of historical reports of shipwrecks

for the entire period of interest.

Last in the order of study was the establishment of survey

priorities and the identification of pilot study areas.

Through

evaluation and ranking procedures, a map was developed showing culture

resource zones for the continental shelf within the study area.

These

zones are based on the probability of discoverability and recoverability

of shipwrecks and submerged habitation sites.

Potential pilot study areas

for detailed archeological exploration have been identified, and a review

of present survey techniques has been made to provide the basis for

recommendations for modification of search and recovery techniques.

This report is presented in three volumes.

geologic setting and prehistoric

The first concerns the

cultural resources.

The second deals

with historical cultural resources, which consist mainly of shipwrecks.

The third volume is an atlas of maps, figures, and tables.

The third

volume is designed to be used with the text, and in fact is necessary

to a reading of the first two volumes.

After the introduction, the first volume proceeds as follows:

in Chapter II, "The Relationship of Process to Form in the Coastal Zone,"

we attempt to develop geometric models for frequently occurring coastal

physiographic features so that forms on the OCS can be identified and

classified as relicts of specific, once-active physiographic units.

Chapter III, "Late

In

Quaternary Relict Forms," the OCS is explored area

by area, west to east, using the published descriptions and maps that are
available, in order to map the shelf and to identify important relict
forms with the past-active systems that formed them before they were

submerged.

In Chapter IV, "Sea Level in the Late Quaternary Period,"

we consider the eustatic, isostatic, and tectonic changes that make

more land available for habitation at some times than others.
level determines the coastline

the

The sea

seaward limit of our study area for

A geological history is developed

any given period of prehistoric time.

of this fluctuating study area.

Chapter V, "Archeological Method and Theory," presents a method of

forming hypotheses about the nature of the archeological possibilities of the OCS

hypotheses

that can be tested with the limited sort of data that can

presently be gathered from the OCS.

The method is this:

the OCS will

be divided into Eastern, Central, and Western Gulf areas, corresponding to

the adjacent areas on land.

The archeological literature of the land

areas will be reviewed to identify major cultural manifestations, by time

and by type.

These can be predicted to have occurred similarly on the OCS

in the time periods when and where it was exposed concurrently.

These

cultural manifestations will be examined for the purpose of developing

tables of index artifacts, environmental-use patterns, and particularly

landforms favored for habitation sites.

Then to the problem of increasing

one's chances in site prospecting on the OCS.

The landforms

(detectable, as relicts) that are most frequently favored at any period

can now be assigned a list of "signatures"

discrete

site indicators

that are capable of being detected by the limited sensing tools and

techniques available for OCS survey.

Chapter VI, "Selected Typical Archeological Sites of the Northern

Gulf," is an inventory of the known sites in the Northern Gulf area that

were occupied from 55,000 B.P. to 3,500 B.P.

Typical sites from three

regions, Eastern, Central, and Western Gulf Coast are selected for

Pre-projectile Point, Paleo- Indian, Archaic, and Poverty Point Periods.

Age, ecofacts, artifacts, and associated landforms of these sites are

discussed.

Delta Area."

Chapter VIII is "An Illustration of Methodology: the Mississippi

Ideally at this point we should summarize the relationship

between the prehistoric occupation sequence and the relict landforms

on the OCS.

But there have proved to be too many unknowns for this to be

possible at the present time.

The methodology developed in this study is

illustrated, however, with a case study of the Mississippi Delta area.

The last chapter, Chapter VIII, "Conclusions and Recommendation for

Future Study," presents the lists of signatures for the types of

frequently occurring sites, discusses the effectiveness of remote sensing

that is available in identifying types of sites, and recommends the

most effective sequential approach to prehistoric site discovery.

CHAPTER IT

THE RELATIONSHIP OF PROCESS TO FORM IN THE COASTAL ZONE

A major objective of the present study is to evaluate the potential

for occurrence of drowned habitation sites on the continental shelf.

Since

only a handful of sites have been identified, all of which are in shallow
water, an indirect approach must be taken.

The approach followed is based

on the assumption that in the coastal area there is a strong correlation be-

tween physiographic units, settlement patterns, and resource use.

Further-

more, the distinctive geomorphic form and character of sediment or material

of physiographic units may be preserved after the units themselves are no

longer active.

These relict forms may remain as evidence of past landscapes,

The physiographic unit is the product of the intensity and kinds of

processes active in coastal systems.

Process is reflected in such measur*

able variables as sediment characteristics, geomorphic form, vegetation,

soil, and human utilization.

The simple truth that the form of the feature

mirrors the process is the key to interpretation, and prediction depends

largely on our understanding of form-process relationships and our effective
use of natural analogs.
The physiographic unit, therefore, is the basic element in coastal

landscape analysis.

While each unit may have a wide distribution through-

out the coastal zone, it can be demonstrated that certain assemblages serve
to identify major natural systems and subsystems.

Figures 2-1 and 2-2

illustrate typical distribution and surface relationships of a deltaic and

a coastal bay and barrier system, respectively.

First inspection suggests that coastal systems are

hopelessly com-

plex mazes of waterways, bays, sand spits, and swamps; however, there is
an orderly arrangement of component parts.

Certain major components of

^C^^^^L
Figure 2-1.

Typical distribution and surface relationships of deltaic

Scale
and related physiographic or environmental units.
approximate (after Coleman and Gagliano, 1965).

Bar

Figure 2-2.

Active Bay Mouth Spits

Relict

Bay Mouth Spits

Relict

Dunes

Idealized barrier spit and bay system showing typical

arrangement of physiographic units.

most systems are repeated so frequently that it is useful to construct

simple models showing typical relationships of one part or component to
another.

A model showing major components of a delta system

in Figure 2-3.

is shown

Such models are not meant to imply that each type of

coastal system has the same components or the same proportion of components.

To the contrary, while many systems are superficially similar

in gross form, they may differ considerably in detail.

Still, the models

are useful in the interpretation of relict systems.

ACTIVE CHANNEL

ALLUVIAL VALLEY^

ABANDONED CHANNEL

UPPER
DELTAIC PLAIN

LOWER

OF EFFECTIVE
TIDAL INUNDATION

LIMIT

DELTAIC PLAIN

MARGINAL
DELTAIC PLAIN

ORE CURRENT

Figure 2-3.

Delta model indicating typical arrangement

of major components.

An active physiographic unit is one which is presently being acted

upon by the processes which shape it and result in its characteristic features.

Examples are active sand pits, point bars, and dunes.

Relict forms

(or features) are no longer being acted upon by the same processes which

gave them their distinctive features, but they retain some of the dis-

tinguishing characteristics of the physiographic unit.

10

In the same sense, a distinction can be made between active coastal

systems and relict coastal systems.

An active system is analogous to a

living organism, each physiographic unit constituting a distinctive part

of the whole.

While active, there are flows of energy and material through

the system; the energy flows take the form of moving water, wind, chemical

cycles, and food chains.

and others.

position)

The materials are sand, silt, clay, shells, peat,

Such systems may collapse (cease to function), move (shift

or evolve into other types of systems in response to changes

in process regimes.

Any of these changes may result in relict assemblages

of features which will identify the relict system.

Assemblages of relict features may be thought of as the skeletons

of formerly active systems.

When found, they tell us that the system

functioned in a certain place at some past time.

Following the approach

of the vertebrate paleontologist, it is not necessary to find all of the

bones in order to reconstruct what the body of the dead system was like
and how it functioned.

Drawing on conceptual models of coastal systems

and assemblages of physiographic units, the landscape of a relict system

can often be reconstructed from one or two distinctive features.

Distribution of Coastal Systems in the Shore Zone of the

Northern Gulf of Mexico
The shore zone can be defined as the band around the perimeter of a

large water body in which coastal systems operate.

The active shore zone

of the northern Gulf of Mexico varies in width from approximately 10 to 40

kilometers.

It reaches

maximum development in the vicinity of the Missis-

sippi River delta area and narrows along those segments of the coast where
long, straight, barrier beaches are developed.

11

As illustrated in Figure 1-1, the shore zone of the northern Gulf

can be divided into 13 segments based on the distribution of active systems

and subsystems.

Since the northern Gulf area extends over almost 18 degrees

of longitude, great variations in process parameters occur.

Variations in

climate, wave energy, vegetation, tectonics, and sediment type are readily

apparent.

Figure 2-4 illustrates variation in climate and wave energy.

Landward of the shore zone is the coastal plain.

This is a broad

area of sedimentary deposits accreted to the continental margin over a

long interval of geologic time.

The present study includes consideration

of the youngest part of the coastal plain, which is made up of Late Quater-

nary coastal and riverine terraces.

These terraces formed initially as

lagoons, depositional and erosional features in the shore zone, and adjacent bays, lagoons, and river valleys.

Their surfaces are characterized

by relict forms which provide clues to their origin.

The continental shelf lies seaward of the present shore zone, ex-

tending from approximately 30 to 220 km and to water depths of about 160

meters.

Like the Late Quaternary

coastal plain terraces, many areas of

the continental shelf contain relict terrestrial forms which are believed
to indicate positions of formerly active shore zone systems.

The present shore zone, then, can be depicted as a chain of inter-

locking coastal systems forming a band around the Gulf margin (Figure 2-5)

Geological data suggest that the present systems have been active more or
less in their same positions for about 3500 years, since sea level reached
its present stand.

The present shore zone is the latest in a sequence of

constantly changing positions that have occurred through geologic time since
the origin of the Gulf.

12

ON

ON

c
O

O
u
X

H
(1)

a
o
3

O
c
0)

.C
4-1
S-i

o
c
43
<4-i

o
en

U
CD

4J
cu

S
CO
CO

a
en
en
a)
CJ

O
M

a
o
CO

E
CU

B
o
C72

CM
CU

u
3
00
H
fa

13

PL AlN

CARBONATE KEYS

Figure 2-5.

'

oM

Diagrammatic representation of shore zone environments.

This study is interested primarily in the former positions of the

shore zone on the continental shelf during Late Quaternary times.
ever,

in order to interpret the forms on the shelf,

How-

it is also necessary

to consider the later Quaternary coastal and alluvial terraces.

Figure 2-6 is a simple model of the shore zone in the northern

Gulf area.

The present shore zone is in a quasi-equilibirum condition

resulting from the relatively stable sea level and climatic conditions
that have persisted for approximately 3500 years.

Geological data

indicate that in the past, the position of the shore zone has shifted
in response to changes in sea level, tectonic movements, changes in

sedimentation rates, and/or changes in marine energy conditions.

The geo-

logical data suggest further that the shore zone has marched back and forth
across the coastal plain and the continental shelf in the not - too - distant
past.
In many instances, the shifts in position of the shore zone have been

parallel (Figure 2-6 B,b and C,b).

It is clear, however,

14

that this has not

River

TERRESTRIAL

System

Coastal

SYSTEMS
CONTINENTAL

COASTAL
SYSTEMS

Sffgjj

System

MARINE

SYSTEMS

a-PAST SHORE ZONE

c-ACTIVE

SHORE ZONE

b-PAST SHORE ZONE

Figure 2-6.

Idealized relationships between active and past systems and shore

The present shore zone consists of a band of interA.
locking coastal systems around the Gulf margin.
B.
The present,
or active, shore zone is the last in a sequence of constantly
changing positions that have occurred through geologic time since
C.
Only remnants of past shore
the origin of the Gulf of Mexico.
zones, represented by relict features, are preserved.
zones.

15

always been the case.

In some instances, parts of past shore zones lie sea-

ward of the present shore zone; other parts lie landward and now stand raised
and drained as terraces on the coastal plain (Figure 2-6 B,a and C, a).

In

still other instances, the present active shore zone is coincidental with

segments of past shore zones (Figure 2-6 B, a and

c;

C,

a and c).

Mappable Geological Features

While each coastal system may consist of an assemblage of physiographic
units, the relict forms of these units may be so subtle or modified that they

cannot be readily identified.

Identification or interpretation is also based

to a large extent on the scale of maps, charts, or aerial photographs avail-

able for study and on availability of supplemental data, such as bottom sediment types.

Primary data in the form of maps, bathymetric charts, and aerial photographs were used at a variety of scales.

Maps at a scale of 1:250,000 were

found to be particularly suitable for the regional synthesis.

reduced to base maps at a scale of 1:500,000 or 1:1,000,000.

Features were
Final presen-

tations were prepared at a reduced scale of 1:1,90,4,762.

Secondary sources consisted of published geological, paleontological,

and archeological reports.

These often contained detailed descriptions,

borings, section radiocarbon dates, and interpretations.

Because of the great size of the area under consideration, only

prominent forms could be displayed on the summary maps.

Mappable forms

are shown in Volume III, Plate 2, and are described below.

Uplands
This is a catch-all category that includes all surfaces inland from
and above the lowest well-defined coastal and alluvial terraces.

16

The seaward

boundary of the upland surfaces is often marked by well-defined scarps and

abrupt changes in lithology.

A distinctive belt of Late Pleistocene and/or

Early Pleistocene deposits makes up the seaward end of the uplands through

much of the region from southwest Texas to southern Alabama.

The forma-

tions in this belt, known as Goliad, Williams, and Citronelle formations in

various areas, are characterized by sand and gravel deposits which apparently represent alluvial valley and upper deltaic plain facies of large river

systems.

In Florida,

these deposits grade into sandy, lime, and chert facies,

the surfaces of which often exhibit

well-developed karst topography.

Inland

from this sand and gravel belt, older Tertiary sedimentary deposits outcrop.
In contrast to the lower coastal and alluvial terraces, relief in the up-

lands is relatively high (up to several hundred feet), and the surfaces are

deeply weathered and indurated.

Coastal and Riverine Ter races

As the name implies, coastal and riverine terraces consist of raised

surfaces that were formerly flood plains and coastal zones active from Late

Pleistocene through Middle Holocene times (see Chronology, Volume III, Plate
1)

These surfaces are characterized by relict features related to formerly

active systems.

Subsequent to their abandonment by these systems, the sur-

faces have been tilted and stand as raised terraces.

They are often separat-

ed from the uplands above and the active coastal zone systems below by dis-

tinctive scarps.

Beaches and Beach Barrier Complexes - This category comprises some of the most

prominent and important mappable forms of the study area and consists generally
of linear sand bodies and complexes of sand bodies.

17

As the name implies, these

features formed along the Gulf shore and were previously shaped by wave and

longshore drift actions, producing well-sorted sand deposits.

In plan,

the

geometry is dominated by long, straight ridges which may occur singularly or in parallel to sub-parallel groups

(Figure 2-7).

The ridge systems

may break into fans and combs, marking former "ends" of the islands.

geometry may be accentuated by aligned depressions.

Ridge

Secondary features as-

sociated with the complexes may be aeolian (dunes) or tidal (tidal deltas)
in origin.

Dune features often accentuate the height of the ridges.

MAP VIEW
1

-r

X
4)

30

<

u.

tTLr^-T-~-^---^Tr?i^-^l' ':':'.v.'.

-V.

>-.bg rrier

strondpioin 5on^''

Blllllllll
t=

60-

i '-'.

; -'-''^! l

V-"-"-"-'S^_

*-_ri_"*:

j^^^bi

1/2

im.ie
1

CROSS SECTION

Figure 2-7.

Pleistocene barrier island and associated strandplain sands,

Smith Point area, Chambers County, southeast Texas (after
Fisher et al.
1973).
,

Meander Plains - The familiar scrolls of bar and swale meander topography serve
to distinguish relict meander plains on the surfaces of coastal and riverine

terraces (Figure 2-8)

While the original relief may be greatly subdued as

a result of sediment veneering,

colluviation, and erosion, the patterns

18

are readily mappable on aerial photographs and remote-sensing imagery.

Point bar complexes with accretion ridges, swales, and cutoff channels are

particularly characteristic.

Figure

2-i

Pleistocene meander plain landscape in the vicinity of Devers,

Beaumont-Port Arthur area, Texas (after Fisher et al., 1973).

Deltaic Plains - Because initial relief is usually very low and the range of
sediment sizes is also restricted, the distinguishing characteristics of relict
deltaic plains on coastal and riverine

erraces are subtle.

Deltaic plains may

be distinguished by relict stream scars and natural levees that branch to-

ward the coast.

Modern drainage is usually incised into former interdistri-

butary basin areas, and the abandoned distributary-channel levee systems

19

stand as straight ridges in the interfluves of the modern drainage deltaic

plains.
In some instances, old beach ridges, lake shores, and lacustrine

deltas can be identified on the terrace surfaces.

Notably absent from the

deltaic plain of the coastal terraces are transgressed features.

Undifferentiated - As the name implies, this is a catch-all category.

It may

include nondescript, relict coastal terrain, or features that have been so

greatly modified that their forms defy interpretation.

Active Coastal Systems

While we make reference to features in the active coastal zone systems,

the mapping of these features is beyond the scope of this report.

The reader

is referred to such standard works as Shepard and Wanless (1971)

Continental Shelf Features

The area of primary consideration in this study is defined as the

gently sloping, shallow-water platform that extends from the coast to the
shelf "break," or that point where a steep slope to deep ocean floor begins.
As emphasized in this report, the continental shelf of the northern

of Mexico is highly variable.

Gulf

The shelf ranges in width from about 30

kilometers in the Mississippi-Alabama-west Florida area to more than 120

kilometers in the south Florida, east Texas, and west Louisiana areas.
While the shelf break generally occurs at about 100 meters, features of
interest in this study extend to about 200-meter depths in some areas of
the Gulf.

Barrier Spit Complexes - Price (1968: 51) defines a barrier as

".

a partly

emergent barlike ridge of sand or coarser sediment lying off a shore or shoal and

20

usually sub-parallel to the shore, projecting from the flank of a headland

or connecting two headlands,

inlets [(Figure 2-9)

A barrier is usually cut by one or more tidal

forming a barrier chain

a succession of

barrier pen-

insulas and barrier islands of simple narrow beaches." Further, "barriers

are commonly connected below water by tidal deltas.

the makeup of a chain is changeable.

While inlets migrate,

The barrier (coastal) lagoon is a

succession of shallow troughs or barriers set off by widening of the islands.

There are bay-mouth, midbay, and bayhead barriers."

Barrier and spit complexes tend to be narrow near headlands.

The

downdrift ends, where they enter bays or migratory inlets, typically become complicated by the formation of recurved spit complexes.

While barriers

and spits begin as single, water- laid beach ridges, commonly vegetated, they

are usually elevated by aeolian sand and become beach-dune ridges.

Beach

ridges may be added to the seaward side if a sand surplus exists to form a

beach plain (Figure 2-10)

Washover fans may form on the lagoonal side and

are associated with washover flood channels, which often develop during storms,

These may be vegetated to become part of the marsh apron on the lagoon side
of the barrier.

Dune development may also become complex.

from blowout areas in the foredunes.

Blowout fans form downwind

Dunes may migrate across the island

and eventually enter the lagoon over tidal flats.

Freshwater ponds are

common features in complicated barrier dune areas.

Barrier and spit complexes are among the most common relict features
on the continental shelf.

This should not be surprising since they are pro-

minent around the Gulf margin in the present active coastal zone and possess
a very distinctive geometry.

21

TIDAL CHANNEL
SPIT ACCRETION RIDGES

BARRIER ISLAND
DEPOSITS

A'

MARINE

Figure 2-9.

TIDAL

DELTA

TIDAL

TIDAL

CHANNEL

DELTA
(BAR)

B'

Features associated with tidal inlets between barrier islands

(after Le Blanc, 1972)

>'._- "s-:

>v

-Beach
Upper sriorefoce

Lower shoreface

BARRIER ISLAND

TTT'
BAY

Figure 2-10. Modern

barrier island environments and facies, Galveston Island (Crosssection after Bernard
1970; after
et al
Fisher et al.
1972).

OPEN GULF

CROSS

SECTION

Beach Ridge Trends - These features appear to be similar to barriers except

that they lack the bulbous ends, or noses, that are believed to represent spit

complexes.

They may best be considered as poorly defined barriers.

Their

scale suggests multiple, sub-parallel, accretion beach and dune ridges rather
than single ridges.

Barrier Accretion Forms - These forms consist of accretion beach-dune ridges

developed on the downdrift side of a headland.

In plan,

the feature is fan-

shaped, with the apex of the form on the headland side of the complex and in-

dividual ridges pointing downdrift.

This type of feature should not be con-

fused with a distributary mouth accretion fan.

geometry, the origin is quite different.

Although similar in plan

Only one barrier accretion fan has

been identified on the northern Gulf of Mexico continental shelf, and it is

in the Mississppi-Alabama-west Florida area.

23

She If -Edge Bulges - This name has been given to a series of very prominent

protuberances along the west-central Florida shelf.

They are believed to be

relict cuspate forelands, but this interpretation remains to be tested.

Shore Trends - This is another catch-all category of linear features that have

been interpreted as relict shoreline trends.

interpreted as relict shoreline trends,

arid

In some places, the features are

in some places, the features are

clearly beach-dune ridges (usually single ridges as opposed to barrier spit

complexes and beach ridge trends).

On the south Florida shelf, some of the

trends are delineated by alignments of relatively small depressions.

Zones

of branching channels transverse to the shore trends are characteristic in

a number of areas.

Escarpments - Escarpments are indicated on the continental shelf by closely

spaced contours.

Although escarpments may be as much as 11 meters in height,

3.7 meters is nearer to the average.

Some of them are remarkably continous.

In several instances, they can be traced for more than 100 kilometers, and

on the south Florida shelf, one continues for more than 200 kilometers. The

most prominent escarpments occur on the outer shelf.

Most occur on rocky

shelves thinly veneered by unconsolidated sediments.

While some escarpments may be the surface expression of faults, the

trends and relationships with other features suggest that most are relict

shorelines.

Minor escarpments associated with "seacliffs" are not entirely unknown

along the northern Gulf.

Modern examples are usually found along the shores

of bays rather than on the open Gulf shore.

24

The Silver Bluff shoreline (Late

Pleistocene) of south Florida is characterized by an escarpment and its related features.

Figure 2-11 illustrates the cross-section of a low-cliff

coast typically developed in limestone coasts.

or freshwater limestone, aeolian calcarenite

sediment.

The limestone may be marine

or other calcium-cemented

These types of rocks are particularly common around the margin

of the northeastern and northwestern Gulf.

wave-cut notches.

Associated features include

In some cases, particularly where tides are developed,

a low-tide platform may develop and be bounded by a second escarpment.

En-

crustations of algae (chiefly Lithothamnion sp.) may be characteristic of

the rim of the low-tide platform (Bird, 1967),

Escarpment

Wave

Low

cut

tide

Platform

notch

Rim

m.h.w.

L.

W.

BEDROCK OR CONSOLIDATED SEDIMENT

Figure 2-11.

Features associated with low-cliff coast developed in limestone.

Tidal Scour Features - This is a group of features whose trend is transverse

to related relict shorelines.

They are believed to have been formed by tidal

scour and related deposition in tidal inlets and at the mouths of bays.

Tidal

colks, or scour holes, in inlets between barrier islands in the active

shore zone, may attain depths of 50 meters and have a distinctive bottom

topography (Figure 2-9)

Tidal deltas are often associated with such fea-

tures and may retain their form after the inlets no longer function.

25

While

a number of these features have been interpreted from bathymetric maps,

it

is not certain that they all formed when the associated relict shorelines

were active.

That is, some may represent submarine erosion and deposition

on the continental shelf.

W. Armstrong Price

(personal communication) re-

ports active bar and trough features in the south and central Texas shelf
areas.

It is also conceivable that while some of these features were ini-

tially associated with active shore zones, submarine erosion and deposition

may have continued even after the shore zones became inundated or submerged

because of the strong influence of the bottom topography on bottom currents.

Banks, Shoals , and Shelf-Edge Knolls - This category includes:

1)

low, shallow,

nearshore features of five-meters relief, in water depths of 15-30 meters, 20-40

kilometers offshore; 2) features of moderate relief commonly less than about
70-80 meters in height, but in places as much as 150-200 meters, in water 55-270

meters deep, 200-240 kilometers offshore;

3)

those which grade into a class of

even larger features, many of which are hundreds of meters in height, in water
depths of more than 270 meters, and more than 240 kilometers offshore on the

continental slope.
The features have probably originated in a number of different ways.
The shallow, nearshore banks may overlie relict coastal forms or salt domes.
The larger hills, particularly those near the shelf edge, were probably

initiated as diapiric salt and shale structures.

The possibility also

exists that some of the deeper features may be of igneous origin.

Most prominences on the shelf are distinguished by concentrations of

calcareous deposits in the form of shells, coral, algae, etc.

Thus, bottom

prominences, regardless of their mode of origin, become foundations for

bioherms.

The remains of the organisms in these reefs and banks depend on

26

specific environmental parameters controlling the growth of marine organisms

(depth, temperature, salinity, current, etc.) and the environmental changes

which have occurred through time.

Thus, the faunal remains from these banks

and reefs may record a complex sequence of changes.

The forms of shelf-edge knolls may range from low mounds to steeper
hills to slender pinnacles.

However, as Ludwick and Walton (1957) have

noted, it is the pinnacle that gives the varied topography found along the
shelf edge in many areas its distinctive character.
of interest to a number of workers.

The knolls have been

Important publications include those of

Edwards (1971), Neumann (1958), Parker and Curray (1956), Poag (1972), Poag
and Sweet (1972), Ludwick and Walton (1957), Rezak and Bryant (1973), and

Bryant

e_t

al.

(1969).

These factors are of interest to the present study for several reasons.

The most important is that some apparently have been exposed as

islands or were shallow reefs during intervals of lower sea stand.

Poag

(1973), for example, suggests that a series of wave-cut terraces, erosional

unconformities, and relict reefal assemblages indicates eight earlier sea

levels on banks near the edge of the east Texas and west Louisiana shelf
areas.

Changes in faunal assemblages also indicate changes in water depth

and environmental conditions.

Terraces - Drowned terrestrial terraces are probably very widespread on the continental shelf.
slopes.

They are characterized by relatively low relief and gentle

They are bounded by other relict, shore-zone forms which tend to identify

them as former terrace surfaces.

27

Karst Areas - These are areas on the shelf where limestone either crops out
or is covered only by a thin veneer of sediment.

The bottom topography is

characterized by relatively dense patterns of sinkholes.

itself

was

The limestone

probably originally deposited in bays and sounds, but was

later exposed to subaerial weathering.

It was during this subaerial ex-

posure that the sinkholes were formed.

Rocks - There are individual rock occurrences reported on bottom charts.

They may be caliche, beachrock, aeolian calcarenite, or freshwater limestone.

Rock occurrence has only been noted on the east Texas and west

Louisiana shelf areas where distribution seems to be related to a major

relict delta lobe and relict barrier-spit complexes.

Channels - There are systems of channels that are sub-parallel to the shelf
slope.

Many have a dendritic habit, suggesting tributary networks in a

shoreward direction.

Also they characteristically occur in zones and are

often associated with escarpments or shore trends.

Entrenched River Axes - These features have been identified by borings and
have been reported in the geological literature.

They are believed to repre-

sent former entrenched river valleys that have been filled by alluvium.

The

Mississippi River trench may be the best known of these features (Fisk, 1956;
Kolb and van Lopik, 1958; Frazier, 1974).

The Pearl River trench (Frazier,

1974), the Sabine (Nelson and Bray, 1970), Galveston (Rehkemper, 1969), and

Apalachicola

(Schnable and Goodell, 1968) trenches are also reasonably well

known.

28

Delta Lobes - Large, low-relief bulges with arcuate fronts mark the position
of relict delta lobes in the south Texas, east Texas, and west Louisiana shelf

areas.

Linear shoal areas are cutting across a number of these lobes and,

sub-parallel to the strike of the present shore zone, are believed to represent transgressive phases of the deltas:

the last traces of delta barrier

islands that formed during the deterioration stage of the delta cycle.

While

transgressive features are well developed on some delta lobes, they are apparently absent on others.

Perhaps those lobes without transgressive features

were formed during periods of falling sea level.

Surface Sediments of the Shelf

Clearly, deposition of marine sediment is presently occurring in many,
if not most, areas of the shelf.

In order to evaluate the discoverability

and recoverability of cultural resources, the kinds and rates of accumulation of these surficial marine sediments must be considered.

Although the

data in this area is considerable, much of it is in the unpublished files

of the petroleum and geophysical service companies.

As one might expect,

there are presently both erosional areas, where Pleistocene and Tertiary

deposits are either exposed or covered by very thin, modern sediment

veneers, and depositional areas (central Texas, west-central, central,
and south Florida shelf areas).

In depositional areas, such as the active

delta lobes of the Mississippi River delta, rates of accumulation of hundreds of feet per century have been documented (Frazier, 1974).

Rapid and

deep burial of either a habitation site or a shipwreck in such areas would

affect not only the discoverability and recoverability, but also the preservation.

For example, wooden artifacts, cordage, and other perishable

materials have recently been recovered from alluvially buried Tchefuncte

29

and Coles Creek Period sites (circa 500 B.C. to 1000 A.D.) in the now aban-

doned St. Bernard lobe of the Mississippi Delta.

In addition to variation in rates of accumulation, differences in

depositional processes and composition are also significant.

A map compiled

by Holmes (1973) represents a recent attempt to summarize sediment distri-

bution on the continental shelf and slope of the northern Gulf.

shows the distribution of five major sediment categories:

ternating sand, silt, and clay,

3)

1)

This map

sand, 2) al-

silt and clay, 4) veneer of sand, silt,

and clay over limestone, and 5) carbonate sand and clay.

Several important

sub-regional studies of sediment distribution patterns have been done based

primarily on size and sorting characteristics and mineral content.

Impor-

tant among these are Curray's work in the northwestern Gulf (1960),

Frazier's recent paper on the Quaternary stratigraphic framework of the

northwestern Gulf (1974), griffin's studies of clay mineral distributions

in the eastern Gulf (1962), and Ludwick's work on calcareous prominences
(1964).

A summary map compiled by Curray showing the distribution of surface

sediment of the continental shelf in the study area is shown in Figure 2-12.

Changes in Levels of Land and Sea

As our previous discussion has indicated, relict coastal forms marking

former shorelines are distributed both above and below the present level
of the sea.

Clearly, the emerged features owe their present position either

to an uplift of the land or a fall in the level of the sea, while submerged

coastal features indicate that in former times the sea stood at a lower level

relative to the land.

Thus, submerged relict shorelines have resulted from

either subsidence of the land, a rise of sea level, or a combination of the

two.

A part of our problem relates to the great difficulty in distinguishing

30

'

o
u
H

X
CU

IH

c
M
<u

42

U
H
O

(2
*->

y-4

-H
<u

42
CO
i-H

CO

C
cu
3

U
3
o
o
0)

42
U
M-l

O
f2

O
H

42
H

CT\

'H

m
\d
iH

U
CO

*s

H
X) r^
ON
4-1
rH

c
CU

#*

t^

H
T3

cO

0)
CO

u
3

c_>

CU
CJ
CO
M-i

M
CU

u U-l
3 CO
C/3 N

CN
CU

60
H

31

between tectonic uplift or depression of the land, and eustatic movements

related to the rising and lowering of the level of the sea.
The general low relief of the northern Gulf area and the absence of

igneous rocks or other indications of violent earth movements might lead

to the conclusion that the area is tectonically stable.

the case.

This is far from

The subsurface structure is very complex, and the region was

subjected to significant vertical movement during Quaternary times.

The study area lies within a great sedimentary basin that includes
the entire Gulf and Atlantic coastal provinces of North America (Figure
2-13)

Sedimentary deposits in the central and western part of this basin

(Louisiana - Texas coastal areas) have been accumulating since Jurassic or

Triassic times and are thousands of feet in thickness.

In some places, the

Tertiary and Quaternary section is over 30,000 feet in thickness (Figure

2-14).

In the Louisiana and Texas areas, the axis of this great sedimen-

tary basin (Gulf Coast Geosyncline) is sub-parallel to the present coast

extending through southeastern Louisiana,

In this area, there are great systems of growth faults, with down-

thrown blocks to the south.

to the present coast

The strike of these faults is also sub-parallel

(Figure 2-13).

Many of the major growth faults are

related to discrete episodes of rapid sediment deposition that occurred at

various times in the geologic past.

These episodes are generally represented

by great ladle-shaped pods of clastic sediment (depocenters) that represent

former deltaic areas.

Complicating the pattern even further are great

diapiric structures of salt and shale.

These are spines of relatively

light-density plastic material that have been squeezed upward in response

to sediment loading.

There is a clear relationship between depocenters and

zones of diapiric structures; spines occur in

gins of the depocenters.

32

arcs

around the seaward mar-

33

fnntr and Middl*

Sand ond

GEOLOGIC CROSS

Snol*

Mann*
Foams

SECTION

SHOWING

DISTRIBUTION

OF SEDIMENT

TYPES

AND

EFFECTS OF REGIONAL NORMAL FAULTS

SOUTHEAST LOUISIANA
ADAPTEO FROM

ATWA1ER. 1967

Figure 2-14.

Geologic cross-section showing major sedimentary units and

effects of normal faulting, southeastern Louisiana (after
Jones, 1969),

The parent beds from which the salt diapirs are derived are buried at

great depths in the central and western Gulf and are believed to have been

deposited in Jurassic or Triassic inland seas.

The salt structures are

common from central Texas to southeastern Louisiana, and a small cluster

has been identified on the Mississippi-Alabama shelf, but they do not occur
Thus, as shown in Figure 2-15,

in the eastern Gulf (see Figure 2-13).

vertical movements within the geosynclinal area may have a number of components and may be very difficult to interpret.
The structural character of the province changes along the coast and
shelf areas of Mississippi and Alabama.

Here, there is a northwest-southeast

trend of major structural elements marking

Coast Geosyncline.

the eastern margin of the Gulf

East of the De Soto Canyon - Florida Escarpment, the

character of the province is different, both from the standpoint

34

of structure and the character of the rocks that have filled the sedimen-

tary basin.

Q *0=
+

TRUESU BSIDENCE OF LAND

TRUE SEA LE VEL RISE

RECENT DEPOSITS

APPROXIMATE LOCATION OF HOUMA, LA

(a)

SEA LEVEL

CONSOLIDATION OF SEDIMENTS OF
GULF COAST GEOSYNCLINE

BASEMENT SINKING CAUSED

BY SEDIMENT LOAD ANO OR
SUBCRUSTAL FLOW

200
VERTICAL EXAGGERATION APPROXIMATELY X 20

APPARENT SEA LEVEL

Figure 2-15.

RISE

OR TOTAL SUBSIDENCE

(a)

(b)

(c\

(c~^\

(o^\

(o~\

(e)

Generalized cross-section of Gulf Coast Geosyncline depicting

components of apparent sea level rise (after Kolb and van
Lopik, 1958).

In the Florida area, the basin rocks are not as thick, and great depo-

centers with associated growth faults and diapiric structures are absent.

From the standpoint of sediment deposition and tectonics, the Florida area
is notably "quieter" than the Louisiana - Texas area.

characterized as an area of slow, gentle uplift.

It can best be

However, local areas of

subsidence or uplift do occur.

These general patterns have persisted into Late Quaternary times.

While indications of some of these movements are subtle, it is nevertheless possible to delineate areas of uplift, subsidence, or relative sta-

bility.

Data for the map presented in Figure 2-16 is derived in part from

the literature and in part from results of the present study.

35

o
u
H

X
0)

o
c
s-

<u

.c
u

O
c

3
dJ

e
a)

>
o
s

cO

2
i-i

0)

U
CO

3
o*
QJ

4J
cO

hJ

c
CO
to
0)
S-i

u
CO
0)

o
H

U
0)

o
CO

CM
0)
S-i

3
00
H

36

Eustatic changes in sea level are even more fundamental to our study
than tectonic movements.

They will be discussed at length in a later section,

37

CHAPTER III

LATE QUATERNARY RELICT FORMS

Many Late Quaternary relict forms have been identified on the Gulf
of Mexico Outer Continental Shelf.

While it is beyond the scope of this work to interpret the history of

every relict form, every form should be related to its presumed mode of origin.

From this, a framework can be built upon which an interpretive history

can be constructed.

In order to achieve this, many of the form interpretations

made in this study will require some local historical considerations.

In the

geological history section of this report (Chapter 6), a Gulf -wide synthesis
of these local historical indicators (i.e., the forms themselves) will be made.

We freely admit that there are many possibilities which could make the

interpretation of submerged relict forms erroneous.

For example, the interpre-

tation of channels as the result of terrestrial processes, when they may have

been formed by submarine processes, could negate many of the interpretations

to be presented.

However, our goal is the identification of areas which possibly were

favorable habitats for early indigenous people.

To accomplish this as ex-

haustively as possible, we should not exclude any area for which a reasonably
plausible interpretation can be given to indicate that such habitat existed.
That some interpretations may later be disproved by new information is in-

evitable, but in the present state of knowledge, it would be unwise to discount any area simply because the interpretations on which it is judged

suitable habitat for people are not definitive or completely documented.

There are simply very few, if any, places on the continental shelf, as the

following discussion will amply illustrate, for which sufficiently detailed

knowledge exists to make such definitive arguments.

38

The Late Quaternary relict forms will be discussed by area from west to
east, beginning in south Texas.

Volume III, Plate

A summary of this discussion is on a map in

2.

South Texas Area

The single most conspicuous form of the entire south Texas area is the

bulge in the shelf.

It is the product of long-term Rio Grande River deltaic

sedimentation (Vol. Ill, Plate 2).

On the northern side of the deltaic bulge,

out near the edge of the shelf, a series of channels of dendritic form is well

defined on recent bathymetric charts.

greater than

130 meters.

These channels terminate at depths no

If the -120 m contour is traced southward from this

area, it is seen to merge into the face of a marked break in slope in the cen-

tral area of the delta bulge.

For this reason, Plate

indicates an escarp-

ment running along this slope break and extending around the distal ends of
the dendritic channels on the northern side of the bulge.

This escarpment is

more or less well defined by the -120m contour, but is not distinctive enough
to be exactly determined.

for its depth.

Perhaps

120

to

135 m is a reasonable estimate

This escarpment is the deepest form that has been identified

in the south Texas area.

The deep feature at -120 m to -135 m is interpreted to be a shoreline.

It is basically a low escarpment, which suggests a short duration or short

period of standstill.

The dendritic channels which terminate at this level

further suggest that the standstill was preceded by an interval of falling

levels.

The channels probably remained active as sea level stabilized at

the -120 m to -135 m level.

Since there are no discernible features at

greater depth, we may infer that soon after this standstill, sea level began
to rise.

That the dendritic channels were not filled or erased by wave

planation during this rise suggests that sea level may have risen rapidly by
a large amount, submerging these forms to a sufficient degree that they have

39

been less intensively modified by marine processes.

As we shall describe

later, this rapid submergence may have been of great importance in preservation
of archeological areas that may have existed.

The channel forms in the southern

part of the area may have been covered by later deltaic sediments.

shoreline at the -120 to -135 m level would more than likely exhibit a markedly
different set of coastal process conditions than the present shoreline.

While the present shoreline is exposed to a sea that overlaps a broad,

flat, shelf expanse, the paleo-shoreline at this low level must have directly

fronted a rapidly deepening ocean with a consequently greater impingement

of wave and tidal energy on the littoral zone.

Perhaps this shoreline had

the character of a low, cliffed coast with a fronting beach.

The low cliffs may have been cut into relatively unconsolidated,

regressive deposits of the falling stage preceding the stillstand and/or

into older deposits.

These same deposits may have been intensely gullied

and channeled inland from the shoreline by small, dendritic stream channels, some of which may have been integrated into the drainage net of the

major river system (the Rio Grande), but many of which may have been smaller
local basins reaching the coastline independently.

The Rio Grande mouth

position at this time was probably to the south in waters off of the northern Mexico coast; it was probably of small extent and influence since it

was building its delta into much more steeply deepening waters than those
into which the deltaic deposits of the present, broad Rio Grande delta are

deposited.

Sediments from this delta may have contributed to the bulge in

contours on the upper slope at about latitude 24.5 N.

The most suitable habitat afforded people who may have existed at the
time this shoreline was active was probably the shoreline, especially where

drainage systems intersected the coast.

Grande delta was most favorable of all.)

40

(Possibly the area of the Rio

The channels traversing the

coastal uplands may have also been suitable.

In the phase of sea level

rise over this landscape, the channel valleys must have become ria-type

embayments and afforded other valuable habitat locales.

that rising sea

It is possible

level stage could have preserved any dwelling sites through

protective burial below bay deposits.

If the rise were rapid, wave expos-

ure duration may have been brief, which would be favorable for preservation.

It seems quite possible that occupation sites may be discovered in

the vicinity of this shoreline or around the associated channels, espe-

cially where bay fill deposits thinly cover channel side areas.
cores would be of great importance

Shallow

in such areas in order to determine the

suitability of any particular place for early people.

Just inland from this form association is another form which is well

outlined by a relatively prominent bulge in the contours at approximately

"80 m.

This is interpreted as a possible lobe of the Rio Grande delta

formed during a stillstand of the sea at approximately that level.

This

form may be related to what seems to be a minor escarpment which is be-

lieved to exist in the northern part of this sector at a similar depth.

This feature will be more fully discussed in the following section, which
deals with the central Texas shelf.
The delta lobe expressed by the -80 m contour is crossed by a few

channels between about -76 m to -86 m. These are possibly equivalent to the

dendritic channels that extend to the -130 m depth just to the north, which

we described in connection with the -120 m to -135 m paleo-shoreline.

they

If

are equivalent, then there is a possible argument that the delta bulge

is older than the escarpment below it.

Next inland from this form is a bulge in the contours at -54 m to -60 m,

which is interpreted as another delta bulge.

ward across this area.

One long channel trends north-

Just to the south is a slightly higher bathymetric

feature, which is a bar form expressed prominently by the -50 m contour.

41

This bar trends NNW in an alignment which is the same as that of the southern

The bar form is interpreted as a beach barrier

flank of the delta bulge.

ridge which represents a stillstand of significant duration following a rising stage.

Whether or not it can be regarded as contemporaneous with the

somewhat lower delta bulge with which it is aligned is conjectural.

Since this is the first beach barrier ridge form we have described,

we will consider briefly the suitability of habitats that may have existed
for people at the time of its formation.

The beach itself, the dune areas,

and older, stranded ridges may have been desirable occupation sites.

The

lagoon shores behind the ridges and especially any areas of marshy character

were probably especially favorable for food gathering.

Where higher ad-

joining grounds bordered lagoon or marsh areas, the probability of occupation sites may be high.

In the area of the delta bulge, conditions for

occupation must also have been very good.

Sites may be related to beach

ridges or alluvial ridges if any can be identified in this area.

The long,

northward- trending channel might be a favorable location for exploration

for such sites.

A close depression is outlined below the

behind the barrier form.
deposits.

-50

m contour in the area

This area may have bay and deltaic sedimentary

Environmental conditions for early people may have been espe-

cially favorable in this area.

Proceeding further inland, there are suggestive closures of the-44 m

contour which are here interpreted as beach barrier forms, although the

interpretation of another deltaic bulge is equally tenable.

features outlined by the
outline.

-44

If these

m contour are connected, they form an arcuate

It is possible that each of these delta bulges showed such arcu-

ate form, as does the modern Rio Grande delta, and that each had barrier
forms fringing the arcuate shoreline.

42

Shoreward of this form, there is a flattening of the slope of the

shelf between the -44 m and -38 m contour, suggesting a terrace that may

partly have been a deltaic plain, contemporary with the beach barrier and/or
deltaic bulge form that we suggest is outlined by the -44 m contour closures.
Two channel features exist that cross this terrace extending from -38 to

-48 m.

Other evidence of channels is seen at about this same depth in

the northern part of the sout Texas shelf, where channel segments are seen

extending from -40 to -44 m depths.

Inland from this feature, there are at least two more barriers outlined by prolongated bulges in the contours at -28 to -30 m and -18 to -20 m.

The outermost of these two forms extends into the vicinity of the Sebree
Bank, where rocks of unknown type occur that are at least partially a result
of encrusting organisms (Mattison, 1948).

This relief barrier and the Sebree

Bank are not necessarily related, but it is possible that the bank represents an outcrop of shelly estuarine facies associated with this shoreline.
Some support is given this by a radiocarbon date reported by Curray (1960),

taken on samples of oyster shells from a depth of 30.5 m lying between the

barrier and the Sebree Bank.

The date obtained on this material was 9,530 +

270 radio carbon years before present (B.P.).

As the shoreline was submerged,

the shell was exposed and eroded partially by wave planation and became

colonized by encrusting organisms that have capped the outcrop of hard

substrate.

Likewise, the barrier was extensively eroded as it became sub-

merged in the wave zone.

The barrier outlined by the -18 to -20 m contour is a more distinct
feature.

If the -20 m contour is traced northward from this feature, it

indicates an area of rather irregular bottom topography about -15 to -20 km to

the north-northeast.

The irregular forms at this depth tend to be aligned

with similar, shallower features which trend south-southwestwardly in a

convex seaward arc toward the present shoreline.
43

This pattern of alignment has

led to the suggestion that they were active bottom features of material

being moved along the inner shelf floor.

Many workers (Swift

et,_aj,.

1971)

consider this to be the dominant bottom topography over wide areas of conti-

nental shelves (such as the Atlantic continental shelf off of the U.S.) which
are subject to more intense energy regimes than the Gulf of Mexico.
of such forms is referred to as a shoal retreat massif.

An area

This area of irreg-

ular bottom topography, with aligned elements sloping over the inner shelf,

may indeed be a shoal retreat massif.

Even so, since it begins with the -20 m

contour, it may originate with materials which outcrop at this depth and pro-

vide a source bed for the materials moving in over the shelf in a south-southwest

direction (and arising gently to the left to converge with the present shoreline)

These reworked materials could be derived from a f acies equivalent to

the - 18 to -20 m barrier.

This might have been a continuation of the form

itself or a shelly estuarine deposit.

Since the barrier diverges in angle with

respect to the present shoreline, our interpretation is that an earlier Rio

Grande delta lay to the northward at the time that the barrier was active.

A minor form, lying still farther toward shore and suggested by a northnorthwest oriented prolongation of the -16 m contour, is shown on Plate
a relief barrier.

as

This interpretation is readily subject to question since it

lies within, and has a common alignment with, the zone of aligned bottom topog-

raphy discussed just above as a possible shoal retreat massif.

We recognize also that if this small feature is actually part of a shoal

retreat massif, then it is also possible to interpret the nearby, similarly

oriented bar forms at -18 to -20 m and at -28 to -30 m in the same way.
they originated as shoal retreat massifs under submarine processes, then

44

If

their potential for archeological resources (shipwrecks aside) is nil or is

limited to scattered occurrences of reworked materials which may have become

incorporated into the shoal retreat massif.

Rusnak (1960) also recognized these last two types of features.

differentiated them into Type

I,

He

which were "low ridges approximately

parallel to the present coast line" and which resemble low barrier islands."
He further described the Type

ridge of -18 to -20 m as having "a narrow

shallow depression shoreward of this ridge (which) suggests a barrier lagoon

remnant"

(p.

Rusnak also states that the slight

159, parenthesis ours).

seaward deflection of this ridge is "attributed to the building of the

ridge on the flank of the Rio Grande delta."

This feature then, was inter-

preted by him in essentially the same manner that we have interpreted it,

including the notion that the deflection of the ridge indicated an adjacent
delta.

Type II was described by Rusnak as "parallel

ridges" extending out-

ward from Padre Island at a high angle (approximately 60 degrees)."

states that the ridges have a relief of about

1 to

3 m,

He

and that samples

from "limited outcrop sampling along-shore" indicate that they are made up
of "partially indurated sands or clayey sands

."

Rusnak interprets these

features as "distributary ridges of an earlier, more northerly extension

of the Rio Grande."

This information is interesting.

Partially indurated sands or clayey

sands do not sound like facies of shoal retreat massif.

They could, however,

be outcrops of scour areas within an overall shoal retreat massif.

Regardless

of that possibility, such outcrop areas of clayey facies may well indicate

some form predating the drift of the shoal retreat massif and associated

scour areas.

This could well have been a delta, as Rusnak believed.

45

We

feel these may be deltaic and/or estuarine beds, but that a delta is

nevertheless possibly expressed, if solely by the deflection of the ridges.

Rusnak also regards these deltaic deposits as present in outcrop

along the beach of South Padre Island, where he describes them as "in part,
of semi-consolidated, weathered, fluvial and fluvial marine clayey sands".
He regards their age as uncertain, but feels that "their altitude and

character indicate, however, that they are not Holocene and should be

tentatively assigned to the Beaumont."

Rusnak also presents a cross-section of Padre Island showing its

relationship to onshore strata logged in a water well boring at Port

Mansfield (Figure 3-1).

The Beaumont formation is shown to underlie

the

present beach at Padre Island at a depth of little more than -10 m and to

outcrop offshore at about -7 m.

From the mainland to the outcrop in the

shoreface, the Beaumont formation is shown to have a brown clay upper unit,

which all borings on the cross-section reached.

Two deeper borings pene-

trated this brown clay, showing it to be thin (-2 to -3 m) and overlying a

thicker unit (-7 to -10) of sand with silty clay.

At a depth of -11 to -13 m,

the two deeper borings encountered "hard shaly clay with shell".

Thus,

as the cross-section shows, these units extend very nearly horizontally

beneath the Laguna Madre and Padre Island, and perhaps all three outcrop

within the -20 m contour.

While recognizing this alternative possiblity, we feel that in light
of present limited knowledge, it is best to not reject the plausible ex-

planation that these bars are littoral barrier forms, and as such, afforded
potentially highly suitable habitat for people before they were flooded by
change of sea level.

Thus, all three relict bars, as well as potential estuarine

46

Fig. 3-1.

Transverse cross-section from the southern Laguna Madre

near Port Mansfield illustrating the pre-Holocene surface
as evidenced by shallow borings (From Rusnak, 1960).

or deltaic areas that may be associated with them, may well be excellent

potential areas of

search for remains of early people.

Having transected the shelf off south Texas up to the present

shoreline, let us briefly consider the shape of the present coast.

There

may be some important differences between the morphology of the present shore
zone and many of those that existed at differing levels in the past.

Many of

the earlier shorelines must have developed during relatively brief still-

stands of the sea, while it is clear from numerous lines of evidence dis-

cussed elsewhere in this report (seeChapter 4), that the present shoreline
has evolved during a very prolonged

tills tand.

For this reason alone, it

may be expected that present coastal forms may differ greatly in scale and

47

geometry when compared with most earlier features.

the modern Rio Grande deltaic plain, which is

Plate

2.

As an example, consider

a wide expanse as shown in

When it is understood that this plain has been developed over ap-

proximately 5,000 years, then it may be realized that a delta plain evolved
in only 1,000 years of relatively constant sea level might differ consider-

ably in overall geometry.

The Padre Island barrier likewise may be expected to differ in many ways

from earlier barriers for the same reason.

Even though it has accumulated over

several millenia of relatively constant sea level, its present width of subaerial

exposure (which is important in consideration of habitat availability) is generally less than four kilometers and often is only a few hundred meters

but

the overall form, including its subaqueous parts, is more than 4 km wide.

This

is probably much larger than previous barriers developed during shorter sea

level stillstands, although in some cases, barriers may have attained great
size in short periods.
factors as* 1)

Rapid barrier formation may have been favored by such

quantity of sand and other coarse fraction material available

for barrier growth, 2)

relation of factors such as slope of the transgressed

surface to varying rates of sea level change, and

3)

differences in offshore

hydrographic characteristics at the time of formation, which can lead to widely

different wave and tidal energy regimes, some sets of which may be ideally

favorable for barrier development.

The modern Padre Island has dunes that occasionally rise more than
10 m above present sea level.

Older barriers may certainly have differed

in respect to dune forms by wide magnitudes, depending on a complex suite

of factors such as past climatic characteristics or quantity and type of

sediment supply.
The wide Laguna Madre may also be
times.

atypical

of lagoons of eariler

The prolonged, relative constancy of sea level may be partly re-

48

sponsible for the great expanse of the present lagoon.

The long stlllstand

may also be reflected by the shallow nature of the lagoon due to filling
by local streams and other sediment supplies over a long expanse of time.
The wide lagoon grades into the mainland transitionally
over wide, saline, barren, wind/tidal flats.

in many areas,

Such flats when dry are

sources of wind-blown, clay-silt flakes, which aggregate to form the

remarkable clay dunes described by Price (1933).

At other points, the

lagoon abuts the mainland more abruptly, making low, wave-cut bluffs across

deltaic and clay dunes.

Much of the older Pleistocene terrain of the south Texas area is mantled by or is part of a terrain of aeolian sand dunes.

This terrain shows

both presently active, as well as essentially stabilized, dune forms.

This sand dune terrain and its relation to other coastal terrains is very

poorly understood.

In the southern part of the area, the sand features

may overlie old, weathered, alluvial sediments of a Pleistocene Rio Grande

Delta, which Price (1933) states is evident in the area of Raymondville

Texas.

Central Texas Area

The deepest indentations which can be identified on the central Texas

continental shelf are

a small number of channels

one possible channel fragment reaching -130 m.

are more numerous.

which reach -120 m, with

Above -110 m, channel cuts

These extend for the most part to about the -80 m contour,

A dendritic pattern is evident in some, and a zone of discontinuity seems to

exist between -70 and -80 m across

which few of the channels are continuous.

These channels are interpreted as having come into existence at a

time when sea level was at -120 m or below, through downcutting of stream

valleys.

Their present, scattered distribution reflects the fact that

49

this area has subsequently been mantled by marine clay and silt facies

which may locally exceed 10 m in thickness (Curray, 1960; Frazier, 1974).

The channel segments that are seen are areas where this sedimentary mantle
is thin or absent.

If the duts extended to a shoreline at or somewhat

below-120 m, then most of the lower parts of this drainage system were
masked by later sedimentation.

Many of these stream segments head in the vicinity of calcareous

banks at about ~70 or-80 m depths, which may have as much as 10 m of relief

above the surrounding bottom.

known.

The nature of most of these banks is poorly

They may be similar in some respects to banks at comparable depths

that have been described elsewhere, such as the West Flower Garden Bank,

(Edwards, 1971) and the shelf knolls

Ludwick and Walton (1957

of the northeastern Gulf described by

Such banks derive their carbonate sediment

cover from a number of reef-building organisms.

Calcareous algae of the

genus Lithothamnion may be especially important along with a variety of

other calcareous algae, encrusting foraminifera, and corals of numerous

kinds.

While many such calcareous banks in the east Texas and west Louisi-

ana shelf areas are known to be cappings of subsurface salt intrusives

this is not necessarily the case for all such banks.

The positions of the calcareous banks of the central Texas shelf are
in part determined by what was possibly an earlier shoreline or an earlier

generation of similar banks.

The present banks are developed wherever pre-

existing substrates were suitable for reef building organisms.

The earlier feature may be evidenced by the suggestive pattern of
the stream channels, which frequently appear to have their origin in the

area of the banks.

Perhaps such topography resulted from a fall in sea

level from somewhere above-70 m or so, giving rise to many local dendritic

channel patterns originating at about this elevation.

50

Inward from this

area,

the slope may have been less steep and unfavorable to widespread develop-

ment of such a channel pattern.

The zone of the calcareous prominences

may then have been an escarpment formed by a resistant unit separating an

inland terrace segment from a more steeply sloping and relatively intensely

channeled terrain element which extended to the depth at which the fall of
sea level reversed.

If the conditions during the higher sea level stand

(at or above 70 m)

were right, this escarpment might have come about in one

of two ways:

1)

it may be related to a shoreline form not entirely re-

moved by subsequent erosion, as evidenced by the dendritic channels; or

2)

alternatively it could have been a group of algal-coralline banks.

If a shoreline form is considered, what type might it have been?

There is no clear way to judge this.

It could have been a barrier

of high- carbonate type which may have been subject to cementation follow-

ing the fall of sea level which formed it.

Resistant calcarenites of

dune and beach origin may be the foundation of hard substrate on

which the present calcareous banks developed when the sea later transgressed this area.

It could just as well have been some other kind of

shoreline, which may well no longer exist, but which left some other

hard substrate areas, such as lags of sandy and gravelly units, caliche,
or estuarine shell units which developed along the eroding stranded

shoreline zone as sea level stood much lower.

With later transgression,

these lag deposits may have provided hard, current-swept prominences

which were favorable to calcareous reef development.

Such ideas can only be speculations without hard evidence from
the field concerning the nature of these banks and their stratigraphic

relationship to forms which may have preceded them.

If they did develop

as 1) lag units of gravel, caliche, sand, or shell; 2) older beach or

dunes (possibly cemented), or

3)

as an earlier generation of algal banks

51

that existed in the same positions as the present ones, then, for all of

these cases, they may have constituted relatively permeable prominences

at the time when the dendritic channel patterns were evolving.

These

coarser, permeable units very likely would have springs at their base,

which might explain the tendency of the dendritic channels to head near
the banks.

If any of the banks on the central Texas shelf are over

shallow salt dome prominences, then there may have been salt springs as
Any type of spring may have provided attractive habitat for early

well.

people.

The observation that few of the channel segments show continuity

between the

-70

and -80 m isobath is possibly an indication of a later

shoreline which may have occupied this position for an interval that was
so brief that its major effect was merely the erasure or masking of the

earlier topography.
edge

This supposition will require much greater knowl-

to detect any evidence of such a shoreline.

Perhaps there are other

features associated with the banks or buried by later sediments which

may confirm or deny such a speculation.

Toward shore from the calcareous bank and dendritic channel forms,
the central Texas shelf is remarkably featureless, presumably due to sedi-

ment cover.

A few channel segments exist between~28 and~42 m depths.

As mentioned earlier, these channels are comparable to those at similar

depths on the south Texas shelf.

Close to shore, a remarkable bottom feature has been described in

some detail by Tunnell and Causey (1969).

This feature is known as the

7-1/2 Fathom Bank, which consists of two small outcrops of indurated lime

marl with numerous freshwater snails that indicate it was formed in a

freshwater environment.

Remains of extinct megafauna have been found in

association with the rock of the bank, although these have been surficial
finds or finds within rock crevices rather than true in situ finds.

52

Tunnell and Causey compare this feature with the marl deposits which
are known to occur over the aeolian sand plain onshore.

These accumulations

collect in depressions in the sand plain, which may have the character of

ephemeral lakes.

Presumably, the rocks of the 7-1/2 Fathom Bank accumulated

in such a lake and became hardened.

by the sea,

Later, the old lake bed was transgressed

and active marine erosion has now exposed the rocks as a

bank on the sea floor.

Such an outcrop must have considerable potential

for occurrence of remains of early people, and conditions for bone preser-

vation are especially good in the calcareous marls.

There may be numerous

other such features in this and the south Texas continental shelf area.
The present shoreline is similar in character to that described in

south Texas, but without a deltaic area of any substantial size.

In the

central Texas area, there are five published cross-sections which detail
the geometry of the modern barriers (Fisk, 1959; Hunter and Dickinson, 1970;
and Wilkinson

e_t

a_l.

These provide an idea of the nature of

1975).

barrier deposits and may be useful as analogies in interpreting a submerged

barrier.

In the cross-section of Hunter and Dickinson (1970) the sand

mass of the barrier exceeds

km in width and is generally about 16 m thick.

As discussed previously, however, it is by no means clear if any of the

submerged barriers on the shelf may have become this large since this

modern barrier has developed during a stand of the sea of long duration.
An intriguing feature of the cross-section (Figure 3-2) is the Late

Pleistocene shell bed which underlies the barrier island sand.

A radio-

carbon date of this material has yielded an apparent age of 26,000 radio-

carbon years B.P.

The validity of dates in this range is subject to

question since even minor accounts of younger,

53

carbon-bearing materials

mixed with materials much older could result in an apparent age of

this magnitude.

The two cross- sections described by Fisk (1959) are a little more

than 30 km to the south of that just described.

body is somewhat wider (about

5 km)

Here, the barrier sand

but thinner (about 11 m)

Fisk re-

ports dates which suggest that the barrier sands have accumulated over
a period of at least 5000 years and that the associated lagoonal deposits

have accumulated over approximately 4000 years.

A boring made from the

modern beach penetrated beach and shoreface sands to about

present sea level and then penetrated about

m of

11m

below

a lighter-colored

unfossilif erous sand which Fisk interpreted as comparable to the aeolian plain.

Below this were estuarine deposits which rested on a pre-

sumed Pleistocene soil surface at -16 m.

this boring were dated at 23,400

Mulinia shells from -15 m in

+ 1800 years

B.P.

This sample was re-

garded as "anomalous with respect to regional information concerning the

level of the sea at that time" (Fisk, 1959, p. 123).

The apparent age

and the depth of this sample suggest that it may be related to the Late

Pleistocene shell bed of the profile shown in Figure 3-2, which lies at
about -15 to -17 m.
the literature.)

(The faunal composition has not been described in

It is possible that these occurrences are equivalent

and that a bed of estuarine shells underlies a large part of this area.
The old,

apparent radiocarbon ages are suggestive that it is a Pleisto-

cene unit.

If this is the case,

then there is nothing to preclude the

possibility that the lighter-colored, unfossilif erous sands (above the

estuarine bed)

described by Fisk is also a Pleistocene unit of beach

or dune origin.

The estuarine bed could be regarded as something of the nature of

a Mulinia-Anomalocardia biozone, which is perhaps a facies of the Beaumont

54

GULF

PADRE ISLAND

LAGUNA MADRE

FORE DUNES

YOUNGER DEFLATION

WIND TIDAL FLAT

BEACH

G.I.W.W
Spoil

Sea Level

Bank

p-

LAGOONAL MUD

-g

LATE PLEISTOCENE SHELL BED (lagoonal assemblage)

'-"

PLEISTOCENE CLAY.CLAYEY SAND. AND SANDY CLAY

Vj &^r*^T^ ~ +X"t."V O^f

?

'

VERTICAL EXAGGERATION *40

Figure 3-2.

formation.

Cross-section of Padre Island, South Bird Island, 7.5 Minute

Quadrangle (From Hunter and Dickinson, 1970)

In other areas, estuarine and shallow nearshore facies are

known in the Beaumont formation (see Pampe, 1971), and its presumed correlatives elsewhere, such as the Prairie formation of Louisiana (Jones

et al.,

1954) and the Biloxi formation in Mississippi as described to Otvos (1975a)

This unit is not described as indurated or weathered where encountered

in borings on Padre Island.

The mineralogy of the shells has not been

investigated in connection with the reported radiocarbon age dates, so it

is not clear to what extent recrystallization may have occurred.

Apart

from recrystallization, contamination could have occurred by admixture of

young and old shell during a time of redeposition by reworking of older
units.

The shell bed of Figure 3-2, for instance, may have resulted from

redeposition of older Mulinia and Anomalocardia during erosion of older

deposits fronting a lagoon shoreline.

new shell may have occurred.

In this way, a mixture of old and

Rusnak (1960) described "layered beds up

to 1 foot thick along the shoreline of Baffin Bay and the mainland side

of the northern lagoon"

(p.

166) and observed that "some of these shells

appear to be eroded from pre-existing deposits, but the bulk are wave con-

centrations of dead shell accumulating from living forms."

Perhaps in

similar facies in the stage of the incipient Laguna Madre, the reworked

55

shell formed a greater percentage of the lagoonal coquina sediment composition, perhaps predominating over the fraction contributed from new produc-

tion by living molluscs.

A third possibility is that contamination may have occurred during the

sampling process.

Fisk (1959) was explicit in pointing out that the estuar-

ine Mulinia sample was taken from bit cuttings.

of the Mulinia shells used in the

C^

If even a few specimens

assay were derived from the younger

beds above, the resulting age could be far lower than the true age of the

horizon from which the greater part of the shells were derived.
In the central Texas sector, the Laguna Madre area is similar to the

south Texas area, the description of which is given above.

The major dif-

ference is the existence on the mainland side of several large, ria-type

estuaries of hypersaline character due to the limited rainfall and runoff of

this area.

The Laguna Madre in the central sector is extremely shallow and

consists of extensive exposed flats across which sand and clay-silt flake
sediments can be moved by wind during times of exposure.

Fisk (1959) interpreted the history of development of the central

area of this part of Laguna Madre to follow this sequence:

1) a "clayey-

sand estuarine facies" of greenish-gray, sandy clay with "locally abundant

shallow water fossils" filled a pre-existing entrenched valley system;

2) at least part of this

estuarine unit was covered by the aeolian plain

sediment type; 3) gradually, conditions changed in the estuary with ac-

cumulation of bay deposits

grading laterally into sands with a shallow

marine fauna, which were suggested as being analogous to conditions in and

around the present-day Mississippi Sound; 4) more general, open lagoon

conditions characterized by clay-free sands with shallow marine shells at

the base grading to more restricted fauna upward; and 5) closed lagoon

conditions comparable to the present.

56

The interpretation given for this sequence is open to certain criticism.

The entrenched

valley system which Fisk contoured may be overly exaggerated, as

suggested by the sparse distribution of his control borings, particularly within

the supposed valley axes (see Fisk 1959, Figure

The clayey-

11, p. 147).

sand estuarine facies," which supposedly filled the entrenched valley

system, was dated (anomalously, according to Fisk) at 23,400 + 1800 years

B.P. by the Mulinia shell sample discussed above.

It would seem possible,

based on this date, that the entire clayey-sand estuarine facies is perhaps
a part of the Beaumont formation.

This would imply that where this unit

is present, the soil zone which typically marks the Pleistocene contact
is perhaps weakly developed or absent.

The greenish-gray color of this

unit accords well with similar descriptions given by Rusnak (1960) and
Shepard and Moore (1955) of stiff, greenish sands, clayey sands, and clays

which were thought to represent pre-Holocene nearshore beds due to occurrences of poorly preserved shell.

Absence of the soil zone could be

due co later removal and exposure of relatively unweathered estuarine

facies of the Beaumont formation perhaps at a time just prior to the

transgression represented by the latter part of the sequence.

Likewise,

the lighter-colored, unfossiliferous sands, which are interpreted as

aeolian plain deposits that in places overlie the "clayey-sand estuarine

facies," also could be a facies of the Beaumont formation.

clarification of some

Following

related features and stratigraphy of the main-

land, this problem will be considered further in a discussion of the cross-

sections of Wilkinson et al.

(1975).

Onshore from the Laguna Madre, there are important elements of the
present coastal landscape:

1)

in the south, the lagoon adjoins the aeolian

sand plain, and 2) in the north, the lagoon adjoins a stranded beach or

barrier complex.

57

The aeolian sand plain of the southern part is described by Fisk

(1959:

107) in this way:

Active and fixed dune fields of the aeolian plain give evidence of long-continued wind activity. The dune fields include both sand deposits and wind-scoured features and have
been termed banner complexes by Price (1958). Some dunes
in the active complexes rise 30 to 40 feet above windIn fixed dune areas, the blowout descoured blowouts.
pressions form extensive lowlands underlain by sandy clays
The fact that these deposits
and by calcareous pond marls.
10 feet shows how long
approximately
of
thicknesses
attain
stable
been
has
surface
local
the
.

On the mainland side of one of his profiles across the Laguna Madre

and Padre Island, Fisk shows a boring which penetrates about

At about

aeolian sand plain deposits.

mately

within this unit

m of the
(or approxi-

m below present sea level), pond marl material was dated at 11,490

+ 240 years

B.P.

Adjacent borings indicated that a weathered soil horizon

lay beneath the aeolian sand plain deposits at about 5 m below present sea
level.

From this information, we might guess that at about 11,500 radio-

carbon years B.P., the sea had not yet reached an elevation of -1 m with
respect to the present level.
of difficulties.

Such guesses are obviously subject to a number

The reasoning involves certain assumptions, such as that

the pond marls did not accumulate below sea level and that the date is not

somehow erroneous.

We elaborate on this here because it reveals the way

that many such isolated facts and similar lines of reasoning provide frag-

ments of information which,

after compilation, give insight into past sea

level changes.

From the description by Fisk cited earlier, the marl-forming ponds

are characteristic of the fixed dune areas of the aeolian plain.

If this

were true in the past, then fixed dune areas must have existed at the time
indicated by the date.

The appearance of fixed dunes could be construed

to be of importance in paleoclimatic interpretations, as could also the

onset of marl-forming conditions; this area, however, remains far too poorly

known to make any conclusive judgments.

58

Inland from the Laguna Madre, to the north of Baffin Bay, there is a

remarkable coastal sand ridge of arcuate form which extends discontinuously

to the south side of Matagorda Bay.

Ingleside Barrier by Price (1933).

This feature has been termed the

As we shall see, this is the first ap-

pearance of an inner barrier form that we have seen in our progression

around the Gulf margin, but by no means the last.

Such barriers are common

around the Gulf, but have been little investigated.

Indeed, the extent of

our ignorance about these coastal features, which outcrop on dry land where

they are readily accessible for geological study, makes it overwhelmingly

apparent how difficult geological interpretation of the submerged forms

can be.
Just north of the Baffin Bay mouth, the situation described by Rusnak
(1960) well illustrates the uncertainties of interpretation that exist.

A dense, brown "Pleistocene beach rock" is described cropping in front

of the Ingleside barrier

local area)

(which is known as the Flour Ridge in this

Rusnak is vague about composition of the coquina as well as

its relationship to the Flour Ridge.

He states

(pg.

190) that "the

attitude of the exposed marine beach rock which flanks the barrier suggests
that the nearshore area had a shallow gulf profile very similar to that of
the modern barrier."

From this, it is apparent that he regards the unit as

a nearshore deposit, presumably a beach deposit since he describes it also

as a marine beach rock.

Elsewhere, he describes the rock as a "coquina

beach rock" and a "dense Pleistocene limestone."

Limestone gravels derived

from this unit are said to consist of "well-cemented shell fragments commonly
so completely recrystallized to calcite that individual shells are unrecog-

nizable."
ic

Without any additional information, it is impossible to be

about environment of deposition.

specif-

While coquinas are characteristic beach

deposits, the judgment that this outcrop is a beach rock nevertheless deserves qualification.

In the same area today

59

(in fact,

in the exact area

of the old "beach rock" outcrop), deposition of a coquina is presently oc-

curring in a lagoonal beach environment.

This modern coquina is also loosely

cemented in places, forming what may truly be described as beach rock.

Beach rock occurrences similar to the above are also known, in asso-

ciation with features that are similar in setting and general character

to the

Ingleside forms, some 400 km to the south on the Tamaulipas coast of the

Laguna Madre near the small village of La Carbajal (Pearson

et al

1965).

Two dated samples of this material gave ages of roughly 25,000 and 35,000

years B.P.

The samples are described as recrystallized, so the results

are no doubt only minimal ages

If we recall the profile of Padre Island by Hunter and Dickinson

(1970), which was discussed earlier, the stratigraphic relationship of the

barrier to a Mulinia-Anomalocardia bed is striking.

The bed, however,

yields an old radiocarbon age and might have been exposed near the surface
for a prolonged time, then transgressed by the modern Padre Island sediments;

but it is not described as

weathered, cemented, or obviously recrystallized.

It could be assumed that the bed was covered by protective overlying strata,

which were removed just prior to the bed being

gressing barrier

overridden by the trans-

so that no prolonged period of conditions favorable

for weathering or other alterations ever existed.

However, no such pro-

tective strata are known above the bed, unless they are sands, such as

Fisk (1959) considered to be buried occurrences of the aeolian plain sediments.

Perhaps the most reasonable interpretation of the lagoonal shell bed

is that it is part of the transgressing sequence.

That is, it formed in

a lagoonal environment behind the early barrier form which evolved into the

modern Padre Island.

This early lagoon contained a living population of

Mulinia and Anomalocardia

but, in addition, was supplied with a great abund-

ance of fossil shells of these same

60

types which were derived from the Beaumont

formation, thus giving the apparent old radiocarbon age.

The fossil shells

were derived perhaps in part from equivalents of the old coquina outcropping
at the Baffin Bay mouth.
If the shell bed encountered beneath Padre Island is considered to be
a part of the transgressive sequence and if we then use the profile shown in

Figure 3-2 as a model possibly analogous to Flour Ridge, some interesting

suggestions result.

The old coquina outcropping in front of Flour Ridge could

well be analogous to the conquina beneath Padre Island rather than a marine
beach rock.

That is, they could both represent the same phase in the se-

quence of development of the two different barriers.

This would suggest

(but not necessarily require) that the old coquina may extend inland be-

neath the Flour Ridge.

But we have no knowledge of the validity of this

possibility.

Perhaps the following sequence of events is a plausible interpretation,

although we by no means suggest a conclusive one.

At some time in the past,

a transgression of the sea led to the formation of the Flour Ridge

barrier.

(Ingleside)

The geometry of this barrier, we have assumed, was analogous to the

Padre Island profile of Figure 3-2.

The sequence recorded by both trans-

gressions is estuarine deposits overlain by sandy beach, dune, and shoreface

deposits.

Since the older coquina unit is approximately at present sea level,

and if it were covered by sand as thick as that forming Padre Island, then it

would seem possible that sea level at the time that the Flour Ridge feature
was active could have been as much as 15 to 18 m above present sea level
(assuming there has been no tectonic uplift or subsidence).

elevation of the ridge is seldom greater than

9 m.

But, the present

The original relict of the

feature may have been partially reduced by post-deposit ional changes.

The

sands of the barrier may have shrunk in volume due to loss of carbonate con-

stituents and other readily weatherable mineral grains; sand may have been lost by

61

aeolian and other erosional processes; and the sand mass of the beach ridge
system may have settled by internal compaction and compaction of underlying
layers.

Following a long history of exposure in an area of active aeolian

processes that apparently have persisted over a long period continuing to
the present, the Flour Ridge stands now as a subdued remnant of its original

configuration.
differentiated.

A few ridge trends of the relict barrier-strandplain can be

These may represent late stage depositional topography just

prior to the time the complex became stranded, or they may be etched out reflecting internal structure.

During the long period of exposure to erosive and

weathering processes, carbonate and other unstable minerals were leached.

Some of this carbonate may have precipitated out within the form to give the

calcareous sandstone concretions which Rusnak (1960) decribes.

The recession of sea level which led to erosion and weathering of Flour

Ridge led to downcutting by the streams which enter Baffin Bay.

The southern

end of the ridge was exposed to strong erosive conditions due to the steeper

slope adjacent to the incised valley, which is now drowned beneath the bay.
At this point, the coquina became exposed, and its induration may be related
to a long period of exposure rather than

to formation as beach rock.

The

large sand body of Flour Ridge can then be envisioned at some indefinite time

period in the past as perhaps lying well above present base level on the

interfluves between coastal streams.

At such times, it probably acted as a

reservoir of local ground water with springs around its margin.

It was prom-

inently exposed to aeolian forces, perhaps for a significant part of this

period of exposure.

At this southern end, Flour Ridge is particularly modi-

fied by aeolian processes and aeolian deposition.

Some resistance to erosion

may have initially been offered by cementation effects due to carbonate constituents.

In time, however, much of the early cementation may have been

62

lost from prolonged exposure, bringing about dissolution of most carbonate

constituents.
However, we need look no further than the northern end of Flour Ridge
to find a very different interpretation of its development.

Wilkinson

et al (1975) have presented a cross-section of this area (known regionally

.

also by the name of the Encinal Peninsula) which is substantially different

from that of Hunter and Dickinson (1970), which lies only 29 km to the south.
This section is reproduced here as Figure 3-3.

The section is based

on the outcrop where the Ingleside trend is exposed along the shore of Oso

Bay and on 15 borings on a traverse extending to southern Mustang Island,

although only five of these borings are represented on the published crosssection.

The exposed section at Oso Bay is described as "up to 10 ft. of

aeolian sand overlying 12 ft. of very fine, yellowish-gray sand, which is

very muddy and slightly calichified in the lower

ft."

This

sand overlies a "bright-yellow oxidized mud of unknown thickness."

Wilkinson et al.

(1975) interpret that the Ingleside sand unit extends

beneath the Laguna Madre and Mustang Island.

In their own words (p.

349)

In the subsurface, Ingleside sand thickens from

12 ft. on the mainland to a maximum of 22 ft. unThis interval is
der southern Mustang Island.
fairly homogeneous consisting of very fine, firm,
These sands overlie a
muddy, well-sorted sand.
complex sequence of oxidized sand, greenish-gray
micaceous sand, and stiff plastic grayish mud.

These descriptions of the exposed and buried Ingleside sands differ in some
important respects:

1)

on outcrop, the Ingleside is termed yellowish-gray,

while in the subsurface, no color term is applied;

is "very muddy";
3)

2)

on outcrop, the sand

encountered in the subsurface, it is simply "muddy;

and

on outcrop, the lower zone is observed to be calichified, while no men-

tion of caliche is made in subsurface occurrences.

63

section

Mustang

PLEISTOCENE

Island

HOLOCENE

Undifferentiated Pleistocene

sand and mud

^r^

Bay-estuanne mud
Barrier island sand

Ingleside strandplain sand

Aeolian sand
'""/////,

Figure 3-3.

Soil

zone

Section across Ingleside strandplain sand, Northern Laguna Madre and Southern Mustang Island (From Wilkinson
et al., 1975).

If these differences reflect the real situation rather than merely lack

of descriptive

detail, then it follows that the presumed correlation of

the surface and subsurface sand units is questionable.

Fisk (1959) encountered

a sand beneath Padre Island and the Laguna Madre, as described previously, and

interpreted this sand to be of aeolian origin.

Note that a soil zone is indicated in Figure 3-3 at the top of the

"Ingleside strandplain sand."

figure was taken (Wilkinson

A close reading of the text from which the

et al.,1975) does not, however, mention any

soil zone in this position, so it is possible that the figure as published

contains a drafting error.

A soil zone which is frequently mentioned in

the text is not shown in Figure 3-3 and presumably belongs there instead of

64

This soil zone occurs at the top of the presumed

the one that is shown.

Pleistocene sands and muds which underlie the sand units.

error may admit a partial truth.

However, this

Since the Ingleside sand at outcrop is

described as yellowish-grey muddy sand, which is calichified in its lower

part, then it can also be said to show a soil zone.

The sand at outcrop

overlies a "bright yellow oxidized mud," which apparently represents

a widespread soil zone described in the article.

Numerous questions arise

concerning these bits of evidence, such as: 1) Are there two soil zones
or one that perhaps resulted from prolonged or intense soil-forming pro-

cesses and affected both the sand and the underlying clay? 2) Does the soil

affecting the Ingleside sand at outcrop exist in the sand beneath the
lagoon and barrier?
As we shall see, these questions will recur as we examine another

profile from the Wilkinson

Bay.

e_t

al.

(1975) article in the area of San Antonio

Other problems relating to soil zones will recur as we continue our

inspection of the evidence concerning the evolution of the Gulf continental

shelf.

Generally speaking, soil zones are inferred wherever oxidized and

somewhat compacted higher strength units are encountered.

Caliche and

ferruginous nodules may also be features of such soil zones.

As persons who

are familiar with soils are well aware, they may vary widely in character on
a local scale, depending on a wide range of circumstances, such as relief,

microrelief, parent materials, climate, etc.

For this reason, among others,

they are particularly untrustworthy for stratigraphic correlation purposes.

As was discussed earlier, Fisk's (1959) profiles, 60 km to the south of this
one, possibly illustrate how a discontinuous soil zone can be stratigraph-

ically misleading.

Wilkinson

e_t

al.

(1975) also present a profile from the San Antonio

Bay area which extends from the Ingleside shoreline feature across the lagoon
to Matagorda Island.

This profile is reproduced here as Figure 3-4.

65

The

sands outcropping on the mainland 20 - 40 km north of this profile line are

described as light-brown, tan, and brownish-gray sands of very fine size

which are regarded as

m or less in thickness and overlie an oxidized to

highly oxidized paleosoil

which at least in part is presumed to have

developed on the surface of deltaic interdistributary muds.

The paleosoil

is referred to as dessicated and calichified.

SECTION

PLEISTOCENE

HOLOCENE

Interdistributary
Soil

'llh

mud

Fluvial -deltaic

zone

Bay-estuanne mud
Lower shoreface mud

Ingleside strandplcin sand

20

12

sand

Barrier nucleus sand

3

SMILES
Flood delta sand

Figure 3-4.

Section across Ingleside strandplain sand, Mesquite Bay

and Southern Matagorda Island (From Wilkinson e_t ajL.
1975).
,

In subsurface occurrences along the profile of Figure 3-4

the sands

which are considered to be equivalent of the outcropping Ingleside sands

are described as light- brown to orange and are stated to be homogeneous.

66

Again, it seems possible to view these descriptions as indicative of

there being a paleosoil within the Ingleside sands at outcrop.

Other de-

tails suggest that the soil is somewhat different here than in the section
at Oso Bay.

In particular, the soil colors are described in darker hues and

the soils to the north were noted to be strongly bioturbated.

The

thin cover of sand (1 to 2 m) also would not preclude the possibility that
the oxidized, dessicated, firm, and sometimes calichified unit beneath is

not a distinct paleosoil, but is part of the same profile of weathering.

The presumed Ingleside sand equivalents beneath the Matagorda Island
are described as light-brown to orange in color.

In this

profile

then, it

may well be that a soil-forming episode has affected the sand unit, which
may be regarded as supportive of the hypothesis that it could be correlative

with the Ingleside sand at outcrop.

This sand unit beneath Matagorda Island

was said to have been encountered in five borings at elevations of 8 - 16 m.

The published section only shows two borings through Matagorda Island, and

one of these did not encounter the "Ingleside strandplain sand."

Neither of

the indicated borings is shown to have encountered the soil zone presumably

beneath the sand, nor is it indicated in the text of the report how many of
the other borings did encounter this soil zone, although it is stated that
the soil was encountered as much as

reported to be up to

5m

17 m

below sea level.

thick beneath Matagorda Island.

reported occurrence of this sand at

-8 m

The sands are

However, the

somewhere beneath the island would

suggest that, at least locally, it may be more than twice this thickness.

This profile again brings to mind the profiles of Fisk (1959) south
of Baffin Bay.

The tan sands Fisk encountered beneath Padre Island at

depths of -11 to -14 m were regarded as aeolian.

These differed from modern

barrier-island complex sediments, from Fisk's descriptions, solely in lacking

shell fragments.

Hunter and Dickinson (1970) did not differentiate sands

67

above the lagoonal shell bed which they encountered at 15 - 18 m beneath

Padre Island, but considered them to be
throughout.

barrier-island sands of Holocene age

If the "Ingleside strandplain sand" is present here,

it was

either not recognized by these authors or it lies below the lagoonal shell bed
That it may lie below is not precluded since the unit underlying the estuarine

shell bed is at least described partially as clayey sand (Figure 3-2).

If

it is truly absent, then possibly it was removed in a cycle of valley down-

cutting and widening as the sea withdrew from its level at the time the

"Ingleside strandplain sand" was deposited.

By this point, it should be obvious that interpretation of the Ingleside

shoreline features and their stratigraphic relation to adjoining coastal

features

is by no means a clear-cut and settled matter.

Before attempting a

preliminary interpretation of these features, let us consider a few more isolated bits of evidence known from the central Texas area.

Price (1958) reported the occurrence of lagoonal mud beneath the Ingleside
sand in a pit near the town of Ingleside, just to the north of Corpus Christi
Bay.

Although Wilkinson

e_t

al.

(1975)

deny that such a sequence is typical

of the area, they do not show that the sequence at this pit is not as de-

scribed by Price.

It is not to be expected that indicators of estuarine con-

ditions (estuarine shells and microfossils, in this case) would everywhere

survive the prolonged exposure above sea level which the Ingleside sands and
the underlying clayey units have both apparently endured.

Conditions for

preservation of these materials may have existed only rarely, and Price's
locale may well have been one of these rare sites.

Wilkinson

t_ al_.

(1975)

argue that the sediments beneath the Ingleside sand represent "older delta-

plain sediments" rather than lagoonal fill; however, in common usage at least,

"delta-plain sediments" may include lagoonal fill facies with which they may
be intergradational in space.

After all, the Nayarit littoral accumulative

68

form described by Curray (1961)

which these authors regard as a form anal-

ogous to the Ingleside strandplain sand (at least in the sense that they are

both viewed as progradational)

does itself contain intercalated estuarine

facies, as acknowledged by Wilkinson and his co-authors.

They also consider

the Chenier Plain area of southwestern Louisiana to be an analogous prograda-

tional form, but one with considerable admixture of muddy estuarine sediments.
The brown, Pleistocene coquina exposed to the north of the Baffin Bay

mouth may possibly be associated with the Ingleside shoreline feature, as

discussed earlier, but no relevant stratigraphic or paleoecologic information
to define the relationship has been published.

Rusnak (1960) was of the

opinion that it was a marine beach rock and part of the Flour Ridge feature.
This is implied in Rusnak 's statement that "the attitude of the exposed marine

beach rock which flanks the [Flour Ridge

barrier suggests that the nearshore

area had a shallow gulf profile very similar to that of the modern barrier"
(p.

190).

Rusnak apparently regards the beach rock as a nearshore or shore-

face deposit.

Modern shoreface deposits have attitudes in the order of 20 ft.

per mile of slope, which, it is important to note, is steeper than the base
of the Ingleside strandplain sand as projected by Wilkinson

t_ al.

(1975).

Fisk (1959) describes the shoreface sands of this area as gray sands with
abundant fossils and wood fragments.
These considerations could be taken to indicate that older sands under-

lying the lagoon and modern barrier could be at least

sand facies of the Ingleside beach deposits.

partially shoreface

When the sea level fell and the

Ingleside shoreline became stranded, these shoreface deposits may have been
reworked and modified in form by shoreline processes, and then affected by

prolonged weathering and erosion so that they presently resemble the Ingleside sands of the mainland.

So far, no fossiliferous evidence of the exist-

ence of such a nearshore facies has been reported in this area.

69

Although

Rusnak (1960) believed the indurated coquina to be such a nearshore

facies, this is not supported by paleo-environmental evidence.

The attitude

of the rock itself, which suggested to Rusnak a typical nearshore profile,

could possible be otherwise interpreted.

Wilkinson

e_t

al.

(1975) emphasize that the soil zone, or zone of weather-

ing which lies at the top of the deltaic interdistributary muds, slopes quite

steeply (10 ft. /mi.), as does their Ingleside strandplain sand.

their profiles, the slope is by no means uniform.

In both of

In Figure 3-4, a break in

slope of the weathered surface (the "soil zone") occurs at about 15 m (50 ft.)

below sea level; in Figure 3-3, a break exists at about 12 m (40 ft.) below.
The corresponding slope-break elevations of the tope surface of the unit, called
the Ingleside strandplain sand, are at about 12 m and

spectively.

m below sea level, re-

Seaward of these slope breaks, the Ingleside strandplain sand

unit slopes more gently.

The slopes beyond the shoreline shown in the cross-

sections are presumably projected since no mention is made of borings or other

information utilized beyond the beach.

According to Figure 3-3, the earlier

formed Ingleside strandplain sands underlie the modern beach at Mustang Island
at a depth of only about -7 m.

At this point, then, it would clearly seem

possible that this sand might outcrop only a short distance offshore.

No evi-

dence exists that this is the case, although there are some slightly anomalous

irregularities in the -12 to -14 m contours offshore from this area.

If these

represent the outcropping sand unit, then the projection in Figure 3-3 is

more or less accurate.

While considering such a projection, let us recall for a moment the

profiles of Fisk south of Baffin Bay.

If the Pleistocene soil zone Fisk

recognized (or the nearly coincident surface of the buried "aeolian plain"
deposits) is projected seaward from the beach, it also would outcrop at about
the -14 m contour.

In this case, we have a remarkable confirmation to this in

70

the shape of the 1\ Fathom Bank pond-marl bed which lies only about 14 m south

from these profiles.

Fisk considered pond marls to be typical aeolian plain

facies, but as shown by the pond deposits near Ingleside, these may have de-

veloped over littoral forms as well.

Fisk's sand, then, could be a littoral sand deposit, and the 1\ Fathom
Bank could be related to this deposit as at Ingleside.
would agree well with that of Wilkinson

st_ al_.

0-975)

This interpretation
.

On the other hand,

Fisk's interpretation could be applied to the data of Wilkinson and coworkers.

That is, the sands they encountered in the subsurface could be aeolian plain

deposits as Fisk envisioned to have been the case south of Baffin Bay.

How-

ever, these more northern locales lie outside of the area of greatest devel-

opment of aeolian forms.

Yet, we must confront the possibility that aeolian

forms may once have been more widely distributed.

After all, there are sta-

bilized dunes on the aeolian plain which indicate that there was formerly

perhaps more intense activity.

But, if any aeolian features did exist, they

did not extend significantly across the present subaerial outcrop of the

Ingleside shoreline trend in this area to the north of Baffin Bay.

Recall that in south Texas, Rusnak (1960) described "partially indurated"

sands and clayey sands outcropping in areas of irregular bottom topography
of narrow, ridge-like form, trending at a sharp angle to the coast.

He

thought that these ridge-like forms were deltaic distributaries, but it is

clear from current bathymetric charts that they may just as well be the result
of scour and drift creating what has been called a shoal retreat

massif.

If that were true,

then the indurated sands and clayey sands

could possibly be beds of the kind described by Wilkinson

et_ a_l

(1975)-

These ridges exist in a depth zone of -12 to -20 m, which is not unlike
the zones of outcrop projected to the north.

There are choices beyond these alternatives.

The diagrams of Figures

3-3 and 3-4 could also be viewed as two differently aged, littoral accumula-

71

tions, which would explain the slope breaks.

of the mainland may

The Ingleside shoreline feature

represent one shoreline, and the buried sands may be

another, laid down after a significant time interval.

could represent shoreface sands.

The zone of steeper slope

Since it appears from the descriptions of

the northernmost profile (Matagorda Island) that the buried sand is also

significantly weathered (orange colors are mentioned), then perhaps its age
is nevertheless an old one, yet younger than the Ingleside.

Fisk's buried

sand could be equally old, but also could be Holocene, as he believed, since

these localities are far removed and may, not surprisingly, show different

sequences.

If it is Holocene,

it very likely is an aeolian deposit, as Fisk

believed, since the bed contains no shell; if it is older, the lack of shell

does not preclude that it may be a littoral accumulative form.

At this point, this review has encompassed the important bits of in-

formation known to us relative to the Ingleside shoreline and related stratigraphic units, and the published interpretations have been examined and reexamined.

While this is not a large amount of information, it is far larger

than exists for almost all of the feature submerged on the continental shelf.
It is by no means a sufficient amount to yield a clear interpretation, as we

have seen.

Central Texas is only the first area in which presumed Ingleside

shoreline equivalents occur, however, and perhaps clarifying information will

yet be found to resolve some of these interpretative difficulties in other areas

Before concluding this discussion of the Ingleside shoreline features,

it

may be useful to attempt a larger view.

history implied by Wilkinson et al. (1975).

First, let us briefly outline the

Figure 3-5 was presented by these

authors as a diagrammatic representation of this history.

A mud-rich lobate

delta prograded across the area (Stage A) and was then left exposed by a lowering of sea level.

A soil zone formed over the delta mass (Stage B)

72

Sea

level rose back across the delta mass (Stage C) and again fell, leaving a
sand sheet of progradational, beach-ridge sands essentially draped over the
old delta mass.

Figure 3-5.

Interdistnbutary

mud
SEA LEVEL

Pleistocene sedimentation in Matagorda Island

area (After Wilkinson,
A - Deposition
1975).
of mud-rich lobate delta
at sea level near present
sea-level position, probably during Sangamon Interglacial.
B - Lowering
of sea level during renewed glaciation, with
subsequent oxidation of
interdeltiac muds.

An important element of this interpretation is that the slope of the

soil zone and the slope and configuration of the Ingleside sand itself are

essentially considered to be inherited from the slope of the pro-delta facies

of the original delta mass.

Such a prograding delta may have produced a con-

vex seaward bulge (it is described as lobate)

If the Ingleside

shoreline

were merely draped over a large delta lobe, its ridge forms would also be lobate
or convex seaward.

However, they are concave

seaward forms which parallel

rather closely the modern barrier -island trend.

Are the convex seaward, sub-parallel trends of the Ingleside shorelines
and the modern barrier significant?

73

They may

possibly indicate

somewhat similar sets of processes and components in the coastal dynamic

setting that existed when the two different shorelines were active.

As dis-

cussed previously, the concave seaward arc of the Ingleside shoreline features

would at least seem to deny that the supposed delta progradation produced a

convex seaward delta front, or even what we normally might refer to as a

lobate form.

If such a

convex seaward land-sea interface ever existed, by

the time the Ingleside shoreline came into

being, a curvature reversal

had taken place to produce a concave seaward shape.

Thus, if the delta front

were lobate, convex seaward, or even if it formed a straight shoreline originally, at some time before or during Ingleside shoreline formation, a con-

cave seaward arc was cut into it

In this central Texas area,

rather large

(9 - 20

m thick by

the condition today, briefly, consists of a

4 - 6

km wide) body of barrier island sand

forming the concave seaward, outer coastline; a sub-parallel,

410

km-wide

back-barrier lagoon; and embayed river mouths with deltas well inland from the
barrier coastline which, consequently, deliver very little sand to the coast.
This situation exists despite the fact that the record of fill in the lagoon
and construction of the barrier described by Fisk (1959) indicate that a sim-

ilar situation has existed for perhaps 4000 years, with a gradual shoaling of
the lagoon going on.
If it is assumed that conditions at the time of the Ingleside shoreline

were very much like those described for the present, then what facies distri-

bution might be seen?

Just as exist today, they would be shoreface, beach and

dune, estuarine, fluvial, deltaic, and aeolian facies.

imply in their work that in the Ingleside time

Wilkinson et al. (1975)

there was basically only a

single facies of a thin sand sheet deposited in a time of falling sea level

over an old delta mass that had previously been exposed and weathered.

As

shown in Figure 3-5, the lower extensions of this supposed thin sand sheet

would be overlying old pro-deltaic beds.

74

Although borings are shown in the

cross-sections to presumably penetrate these old marine units, it is not reported that any paleoecological evidence was encountered to indicate support
for its interpretation as a pro-deltaic marine deposit.

If such evidence did

not exist, it would not be surprising considering that these beds must have

been exposed to weathering for a considerable time.

The lack of evidence for

one paleoenvironmental interpretation, however, is no different than the lack

of evidence for another.

Thus, it would appear that Price's (1933, 1958) concept must still be

tenable, of the Ingleside shoreline as a form analogous to the present barrier.

The lack of estuarine fossils is no more of a serious objection than the lack
of nearshore marine fossils.

be rare.

If either exist anywhere in this area, they may

The fact that the Ingleside sand extends into the subsurface as a

thin unit could easily be due to later erosion, so this does not particularly

favor the concept of strandplain progradation during a falling sea level,

as

Wilkinson

et^

al.

Their argument was that a prograding

(1975) suggest.

strandplain in a time of stable sea level would necessarily thicken seaward.

Since the Ingleside sand of their profiles neither markedly thickened seaward

nor overlaid estuarine deposits, they argued that it must have developed in
a regressive or

progradational manner influenced by a falling sea level.

The presence of distinct ridges in the Ingleside shoreline features is,

of course, evidence of progradation just as it is evidence of progradation in
the modern barrier form.

The lack or rarity of estuarine fossils is, as we

have shown, not particularly strong evidence.

of the thickness of the Ingleside sand?

What, then, about the matter

First of all, we should note that a

field trip guidebook (Corpus Christi Geological Society, 1958) cites two profiles by C. E. Johnson which indicate that the Ingleside sand in the area of

Rockport, Texas, is up to 20 m in thickness.

Graf (1966) has also cited that

the Ingleside sands in segments of the old shoreline between St. Charles Bay

75

and Aransas Bay are 18 m thick, although he describes no borings or other

sources of this information.

Perhaps these reports of thicker sands represent

inlet fillings of greater original thickness.

Perhaps the thinner areas have

been substantially eroded in addition to undergoing some volume shrinkage

from loss of unstable mineral components.
The prograding of delta sediments in this section of the Texas coast

Wilkinson

deserves further consideration.

e_t

al.

(1975) believe this occurred

before the Ingleside shoreline features were formed.

The rivers of this

section of the coast (the San Antonio, Nueces, and Guadalupe) presently are
not large rivers and have not prograded substantially deltaic areas, for the

most part, even into the shallow bays.

made the prograded delta mass?

Was it these same small rivers that

Wilkinson

ejt

ajL.

(1975) state that with the

fall of seal level following the delta progradation, there was no further delta

building in the area of their study.

Perhaps then, the drainage basins of

the coastal streams entering this central Texas area underwent some change by

stream capture, or perhaps there has been no further delta building because
the right conditions for it have not since recurred.

A condition favorable for a period of delta progradation could be that

of a falling sea level.

Consider, for instance, the situation existing at

present in central Texas, and then imagine the changes that might occur if
sea level began to fall.

Suppose that the fall initially would be gradual.

The lagoons and bays would become shoaled.

over the fromer bays and lagoons.

Rivers would prograde forward

As these bodies of water became shoaled

from falling sea level, the growth rate of deltas would increase because the
same volume of added sediment would spread over progressively wider areas.

Even a moderate fall could possible occasion a rapid progradation of deltaic

deposits across the present lagoons and out to the barrier.

Another condition favorable for progradation of deltaic deposits would

be ample amount of time for progradation, such as during a prolonged still76

However, the prolonged relative stillstand of recent millenia has not

stand.

been ample enough for significant progradation of deltas by coastal streams,

despite the fact that deltaic sedimentation is taking place in the shelter of
a large

barrier island system.

The largest of the coastal rivers in this area,

the Guadalupe, has been slowly and steadily building a small, sheltered delta

within the bay, which as been detailed by the investigation of Donaldson

et al.

(1975).

These authors predict that given an indefinite continuation

of the current conditions into the future, the Guadalupe River will build

across the lagoon and barrier island to discharge into the open Gulf, but add
that, at that time, "the low load of the Guadalupe River probably would be

sufficiently dispersed by longshore currents to prevent appreciable delta

development in the open Gulf" (Donaldson et al., 1970: 130).
Imagine the situation that must have occurred if the diagram of Figure
3-5 is a correct one.

The progradation of a delta mass directly against the

opposing forces of the Gulf of Mexico, with no sheltering barrier, would mean
that a much larger supply of deltaic sediment was reaching that area than is
at present.

There are various ways to account for a greater amount of sedi-

ment reaching the coast, such as changes in vegetation cover in the watershed
or increase in the size of the catchment basin area; but a very large magnitude

change in the amount of sediment is difficult to visualize.

Let us suppose,

for the sake of discussion, that much of the modern Colorado River catchment

basin was connected to a basin which flowed to central Texas at the time of the
delta progradation.

This would have meant a greater sediment input, favoring

delta progradation, but it is doubtful that even this would be as strong a pro-

gradation as required in the hypothesis of Wilkinson et al.

(1975).

Within

the last century, the modern Colorado River has finally prograded to the barrier

shoreline (Kanes, 1970) with the assistance of artificial channel clearance.

This progradation occurred after a prolonged period of deposition into a shel-

tered

lagoonal situation.

Sediment from only a single large catchment is

77

possibly inadequate for any significant amount of delta progradation at the

open shoreline of the Gulf.

Climatic differences might have added somewhat

more sediment yield, but doubling of the amount might still have led to only

moderate progradation and then only locally.

High-energy Gulf shoreline con-

ditions would have prevented much sedimentation, considering how slow it has
been in recent millenia under sheltered conditions.
For these reasons, the delta progradation must have occurred either in
a sheltered situation or under conditions of slowly falling sea level.

shelter may have been a barrier, possibly the Ingleside shoreline.

The

The delta

may have prograded at the open Gulf shoreline while sea level was falling.

In

this case, the Ingleside shorelines may represent merely pauses in the fall of
sea level.

Perhaps there was a prolonged stage of river-mouth discharge at the Gulf

shoreline.

At such times, river-transported sands would have reached the coast

and may have contributed to the final staees of beach-ridee development, which
are the forms that we observe today.

In this special case, delta progradation

and beach-plain progradation may have been essentially the same thing for some

span of time.

During such a stage, the subtlety of the distinction between such

terms as barrier and strandplain reminds us of the arbitrariness of the terms

by which coastal features are described.

Price's

interpretation of the Ingleside barrier was no doubt influenced by

his well-known concern for the concept that coastal forms are products of equil-

ibrium relationships of particular sets of coastal conditions.

The large-scale

similarity of the Ingleside shoreline forms and the modern barrier forms clearly
suggests a similar equilibrium existed during the formation of both shorelines.
Not only is this suggested by the forms, but if we try to suggest that coastal

conditions were in some way markedly different in Ingleside times, then it is

difficult to demonstrate that any such differences could have been large enough to
matter.

This was shown earlier in our consideration of how difficult it is to

78

conceive of the small rivers of this central Texas area actively prograding
into the open Gulf.

Yet, the objection to the period of progradation is

diminished if we imagine that it could have taken place behind a sheltering

barrier or during a falling sea level period.

the delta mass

In either case,

and the Ingleside shorelines could well represent essentially time-coincident

units.
In the final analysis of the thesis of Wilkinson

e_t

al.

it is

(1975)

apparent that the crux of the matter lies in their interpretation of a distinct
soil zone beneath the Ingleside strandplain sand.

In their article, the soil

zone is specifically stated to exist only beneath their more northern profile
(Figure 3-5 X while in the southern profile, the corresponding horizon is not

described explicitly

as a soil zone, but only as oxidized mud and sand.

They

also recognize evidence of oxidation and calichif ication of the Ingleside

strandplain sand itself; therefore, it is apparent that either two soil zones
exist or that there is really only one soil zone affecting both units.

Since

their Ingleside strandplain sand as shown in their profiles is rarely more than
7

m in thickness, these soil zones are necessarily so close together in most

places that it could well be argued that they are one and the same.

If that

were the case, the argument that the Ingleside strandplain sand is significantly

younger than the underlying deposits would be questionable.

Unfortunately,

neither of these soil profiles has been described adequately

to judge

the possible relationships of these units.

With this problem in soil zones, we must let matters rest until further
clarification of stratigraphic evidence is made in the literature.
We choose, however, to interpret all of the evidences examined above con-

cerning the Ingleside shorelines in the following manner.

At some time in the

past, sea level stood higher that its present level, perhaps as high as 16 18 m above its present level or perhaps less than half of that.

79

The Ingleside

shoreline existed as an active littoral form at this time.

It was a prograda-

tional feature and may have had partly the character of a barrier island and

partly the form of a prograding strandplain, even though no definite estuarine

or shoreface deposits have been described from the central Texas area.

Perhaps the form was developed in a cycle of evolution of the following

kind:

1) As sea level

was rising to a peak level, river mouths were drowned,

and erosion of interfluve areas led to barrier-island formation through longshore drift of sand from interfluve headlands across drowned river mouths. 2)
continued, gradually diminishing transgression led to greater development of

barriers with greater continuity of lagoon segments, which were still filling
too slowly to keep pace with rising sea level.

3)

A period of stillstand or

near stillstand followed, and some filling of lagoons by both small coastal rivers
and some aeolian- and storm-washed sands
riers, occurred.

which were transported over the bar-

In many places, the lagoons and bays may have become virtually

entirely filled, as in the Laguna Madre Flats of today.

4)

A period of slight

fall of sea level, during which progradation and bay filling increased rapidly,
and progradation to the barrier shoreline probably occurred.

5)

A period of

occupancy of the barrier-strandplain shoreline followed, with continued accumulation of deltaic deposits in former bay and lagoon lowland areas behind the

barrier

which was terminated by 6) a period of continuing sea level fall,

abandonment of the barrier, exposure of the shoreface, and entrenchment of the

coastal streams across the former shoreline.
The Ingleside shoreline features probably began to take on their character
as we see them today during a time like that of Stage 5 in the discussion above.
If the situation at that time was anything like the present,

then this prograda-

tion must have been a slow process, possibly accelerated by a falling sea level.
The Ingleside features that we see could well be interfingered with deltaic and/
or estuarine beds.

Their relative prevalence would be a matter of the phase of

the transgressive-regressive cycle in which features that we see took ^iheir shape.
80

The streams may have prograded beyond the Ingleside shoreline as sea level fell.

This progradation would mainly have been from recession of the shoreline of the

falling sea, with advance occasioned by sedimentary progradation somewhat accelerated by falling sea level.

Deltaic progradation during a falling sea level may have led to a regressive sandy deposit, essentially of beach-ridge forms.

As the fall of sea

level reached the stage at which all coastal rivers began to deliver sands to
the mouth, beaches may have slowly accreted forward at progressively lower and

lower levels.

At some point during such a fall, rivers may have increased in

sand trasported to the coast as they began to scour and erode that sand pre-

viously stored in more inland deltaic and floodplain units.

The falling sea level may have been interrupted by relative stillstand
periods, as suggested by the slope breaks of Figures 3-3 and 3-4.

One or more

progradational beach-ridge forms at lower levels than the Ingleside features

outcropping today could have formed during such pauses.

The cross-section

could indicate two such separate pauses at perhaps about 10 m and 20 m below

present sea level.

The approximate 10 m below present sea level stand was

recorded by progradation of sand in the southern profile, but was an interval

of perhaps non-progradation, or even coastal erosion, in the area of the north-

ern cross section.

The approximate 20 m below present sea level stand is

possible evidenced by progradational sands in the cross section to the north

(Figure

3-4),

but these would lie

offshore of the southern cross-section, if

they existed at all.

Just to the north of Corpus Christi Bay, near the town of Ingleside,

an interesting occurrence of vertebrate fossils has been unearthed.

(1972), in a monograph on the Ingleside fauna, drew several important

81

Lundelius

conclusions:

1)

the fauna were preserved in freshwater pond marl and cal-

careous sand facies which must postdate the barrier-strandplain ridge; 2) "a

post-Sangamon age fits the faunal picture better than a Sangamon one"
3)

age

(p.

6);

the presence of two large tortoises "probably rules out a late Wisconsin
(p.

6);

and 4) certain elements of the fauna suggested that ecological

conditions at the time the fauna were living could have been those of a forested area perhaps more humid and with warmer or less extreme winters than at
present.

Lundelius was careful to stress the somewhat tentative nature of

some of these conclusions.

The pond marls of this site bring to mind the similar pond marls of the

aeolian plain and the submerged occurrence at 7-1/2 Fathom Bank.

As we have

cited, Fisk (1959) reported a radiocarbon date of 11,490 + 240 years B.P. on
a deposit of this kind in the aeolian plain.

These various pieces of information give credence to an interpretation

that perhaps a warmer and more humid climatic period developed at some time

after aeolian processes affecting the Ingleside ridge and adjoining terrains

began to diminish.

Stabilization of the aeolian sand terrains could have re-

sulted from a more humid local climate.

As Lundelius has suggested, the pond

marls may have accumulated in a pre-existing blowout depression formed at a

time of lower water table.

The rise of the water table to form the pond may

have been partly due to sea level change and partly due to climatic change.

Following this, marl deposition began, accompanied by the accumulation of remains of faunal more diverse than the present.

Lundelius describes

fauna with 17 kinds of medium- to large-sized

herbivores, as compared to only four in this category today.

These include

extinct species of camels, horses, elephants, bison, and other animals not so

readily analogous to living types, such as other proboscideans, the tapir, and
glyptodonts.

There were also canids, felids, and other smaller mammals.

small camel or llama-type animal, Tanupolama mirif ica , which was very well

82

represented in the deposit, showed distinct modes in the height of the third

molar of juveniles, suggesting to Lundelius that these animals bred seasonally

and were preserved in the pond deposit

due to its being intermittently dry

during a distinct seasonal dry period.

East Texas Area

The Outer Continental Shelf off the east Texas coast shows a remarkable

group of banks, some of which have reefal carbonate deposits.

Edwards

(1971), Poag (1973), Rezak and Bryant

Studies by

(1973), and several others have

revealed many interesting features and details of these banks which are of the
greatest importance in the overall interpretation of the continental shelf.

Many of the banks are known to result from intrusion of diapiric salt
into the subsurface

High-resolution, sub-bottom profiles show the structural

disturbance of strata in and around the banks.

(1973) and Edwards

Profiles illustrated by Poag

(1971) indicate that structural deformation is episodic in

time and shifting in space due to the intrusive salt rising in spine-like pro-

jections from the underlying salt plugs.

Because of these characteristics, it has many times been pointed out that

structural deformation may be so active that it may cause appreciable change

in the elevation of the banks even on a short-term scale.

Nevertheless, many

workers, even after noting these problems, have apparently regarded this as not
a particularly serious impediment for interpretation of features of the banks

produced during lower sea level stands.

Although Poag (1973) acknowledges this

potential instability as a problem, he nevertheless identifies various levels

and features on the banks, relates them to former sea level positions, and sug-

gests correlations with features at similar depths from bank to bank and elsewhere.

All other workers on the banks in the east Texas area have done the

same thing

83

Parker and Curray (1956) summarized the bathymetry of shelf banks in

a histogram which shows modes of their crests at about -3, -10, and-88 m.

Curray (1960) has pointed out that the modes in the bank tops which lie at
-3, -10,

and-88 m would seem to deny that a long stillstand at-76 m (as sug-

gested by McFarlan, 1961) could have occurred.

This ignores, however, the

fact that not all of the banks represented in the histogram are surrounded by

shelf terrain lower than -88 m.

Thus, it is not reasonable to argue that a

large proportion of banks should lie at this level when many of the banks in
the population described were actually hills rising from the coastal plain

rather than islands when sea level was that low.

The idea expressed by Curray (1960) that a very prolonged stillstand
(greater than 15,000 years, according to Curray) would lead to widespread

and pronounced truncation of banks at the level of the stillstand is not

necessarily sound.

Islands may persist for long periods of time, depending

on the relative balance of destructional and constructional processes that are

operative in the particular coastal setting in which they occur.

Edwards (1971) felt that reef-building constructional processes on the

West Flower Garden Bank (WFGB) were a late feature in the bank's history, with
no important reef facies below -30 m.

Further investigations by Rezak and

Bryant (1973) have shown that this is incorrect, and that drowned reefs are

known to exist at three lower levels:

-128 to -131 m.

-55 to -58 m, -88 to -89 m, and

This is direct evidence of constructional processes at

work, which must have compensated for destruction from wave attack.

The recognition of the three levels of drowned reefs in the WTGB leads
to many interesting questions about the internal structure of the bank.

Edwards' view that the bank has a core of Tertiary rock is somewhat question-

able in this light, and the possibility arises that its core is mainly reefal

material (coralline and algal).

This cannot be substantiated from published

descriptions, which are overly vague.

84

Edwards presented photographs of

presumed Tertiary rock outcrops at -79 to -82 m, but gave no description of their
character.

He described lithoclasts, which were a minor component of modern

sediment facies, from

-50 to -113 m.

These were described as "dark in color" and

usually with "numerous silt-sized shell fragments surrounded by a muddy maThe lithoclasts were said to be derived from biodegradation and ero-

trix."

sion of outcropping Cenozoic sediments, such as were illustrated by the photographs.

Without additional evidence, it is not possible to decide if the

bank is primarily reefal throughout or if it has an older rock core.

In either

case, reefs have existed in the past at four different levels, at least on the

WFGB.

Edwards (1971) concluded in a study of the WFGB that his evidence indicated former shoreline positions at several levels:

1) below-182 m, as indi-

cated by "buried, seaward-dipping erosional surfaces" which extend to this

depth, as seen in the sub-bottom profiles; 2) slope breaks and a small terrace
at -121 to -134 m; 3) an -89 to -90 m terrace and sub-horizontal, stratified re-

flectors in the sub-bottom, with some small-scale channel forms; 4) a -73 to

-82 m level marked by one wide terrace area and some smaller "erosion surfaces";
and 5) a -46 to -51 m level with terraces on the two main pinnacles.

The argument that Edwards (1971) used to hypothesize that deltaic sediments

were deposited in the area of WFGB at the time of his -89 to -90 m sea stand is
intriguing.

It is based on the sub-bottom profiles which show stratification

suggestive to him of topset, bottomset, and foreset structure, such as that

associated with deltaic masses. From these structural properties, he infers

that a delta existed to the north and east of the bank during the -89 to -90 m

shoreline occupancy.

This is a plausible argument.

Yet an alternative ex-

planation may be that this structural configuration is part of the local tectonic effect of the salt dome.

Specifically, it could well be evidence of a

rim syncline zone of deformation around the dome, which is an apparent feature

85

in the deeper sub-bottom as shown also by Edwards' one-cubic-inch, air-gun

profile (reproduced here as Figure 3-6).

Resolution of these problems will require both samples and, possibly,

cores to learn the character of the sediments and a better three-dimensional

concept of

the structure of the dome from more reflection profiles.

With

samples and cores, the age and environment of deposition of the sediments can
be established, and our paleogeographic information will increase.

This, added

to increasing structural knowledge from more sub-bottom, structural data ac-

quisition (and increasing skills in storage, retrieval, and display of these

data)

should permit more definite interpretation of paleogeography in the

near future.
Poag (1973) stated that "wave cut terraces, erosional unconformities,
and relict reef al

assemblages" on the east Texas banks indicated eight past

sea level stands at depths of -60, -78, -82

-100, -126, -135, "194, and "223 meters.

Poag states that features at these levels are present "at equivalent depths
on several banks," but actually presents evidence only in the form of four

sub-bottom profiles over four banks.

Some of the features shown on these

profiles are relatively minor and are not particularly convincing (such as
the-223 m feature), but others (such as a shelf at
and

-78 to -85 m) seem real

significant.
Clearly, detailed interpretations of this group of banks, where tectonic

effects are probably significant, will require highly detailed investigation

of each separate bank.

For our purposes in this report, however, the banks

clearly show evidences (as cited in the works reviewed above) of having existed
at various times anddifferent positions of sea level as submerged banks, reefs,

islands, headlands, and hills rising from the coastal plain.

For all cases of

subaerial exposure, their suitability as habitat for early people was likely
to have been excellent.

86

Z 5
co b
on ^

"^^

w'f^'

'

:i):fi

'>3&

''

''

' 5
'

*.

.."..*>.
\-x>.

u*P-"

P3

--.\v

:-.>

^!

'

-''

.'

'''?

CO

CO

fc\ftH? ';',.-

'

C
v

'*w*aa'. n
\*K*v ^'s'

'rtS^

'

v.-

.-.-*

S,

'ASF

',"'':

<-.-

cu

>.

"

u
CO

whSSt* <Wu

'

<*-

*"- *-'

'-.-.",v*y

-.

u *

'

a)

&
o

PSNiv: &&& '':

.* -

'->

v'

w
(U

0)

a)

4-1

M-l

O
..,.*

..

(U

H
H
oJ

MH

O
l->

"

.V

v '.'"-

iH

r-l

CT\

CO

r-l

&

0)

CO

v|f

W
QJ

8$

JH'.r. j

>

/'

'

O
r>-

W/f (.-.*
li/-;.;.

"it-;
..

'''

'hJ-JK.n

in

4J

r*-

CM
co
in
o
co

CM CO

87

a)
V-i

3
60
H

Edwards has presented an illustration of this interpretation of the ap-

pearance of the WFGB at the time sea level was about 121 to 134 m below the
present level (with which we concur in general)

(Figure 3-7)

The illustration

is an oblique aerial view looking landward (northward) of the WFGB dome struc-

ture, rising as a hill at the coast much as other salt domes (such as High

Island) do today.

The various potential habitats can be readily visualized.

Barrier froms are interpreted to have existed flanking the dome in the foreground

Besides the banks, there are other indications of relict forms on the
Outer Continental Shelf.

Volume

3,

Minor excarpments or slope breaks are indicated on

Plate 2, in water depths of -130 to -140 m and -80 to -90 m.

These

compare with similar features on the banks.

One of the most prominent features of the east Texas shelf sector lies
on the far western side, where the -80 to -100 m contour reveals a large, arcuate

bulge which we interpret here as a submerged delta.

There is a steepening of

the slope associated with this bulge ar about 84 m.

At present, this is still

an area of limited bathymetric detail, but generally it seems that contours

The regularity

are rather regular, without indications of complex topography.

of the contours probably reflects the smoothing over of relief features by

Holocene marine sedimentation which, by obscuring the details, makes interpretations difficult at this time.

Curray (1960) referred to this form as the

Colorado delta, but did not elaborate on this interpretation.

He did show that

surface sediments in the area of the bulge were of a subarkosic nature similar
to those of the present Colorado-Brazos and Rio Grande deltaic sediments.

In the eastern half of the east Texas sector of the shelf, there are two

ridge-like forms which both trend into the area of Stetson Bank.

On the land-

ward side of each, there is a depression of channel-like configuration.

Curray

(1960) termed the channel on the western side the Outer Brazos-Colorado channel,

88

3
O
>
CO

CO
rt

s
!-H

CU

>
CU

rH
CO
CU
CO

c
CU

x:

^
a

CO

PQ

CU

s-\

T3

iH

>-i

l~~

CO

ON

O
M
cu
&
o

H
P>H

rH
ri

CO

id
)-i

cO

5
-a

4-1

CO
CU

12

u
CU
4-4

CU

X,

LW

<

4-

s-'

4-1

iH

O
C
o

CU

>
CU

H
4J
CO

4-1

CO

cu
CO
cu

>-i

j-i

iH
tH
H
0)

D,

&
o

>

<-i
Q)

ja

4-1

CU
>-i

E
<r

a, ro

CU
4-

M
,

1^
1

CO
CU
1-1

3
00
Pn

89

rH
1

iH
CN
rH

which he believed to have formed as an actual alluvial channel of these rivers.

The ridge south of the channel he regarded as a barrier spit, which developed
at the mouth of this river during a falling stage of sea level.

This interpre-

tation reflects the fact that the ridge is not a sub-horizontal terrain element,
but slopes across the shelf from depths of -53 m to -64 m.

The more eastern ridge form and associated channel was interpreted in an

analogous manner.

In this case,

the channel was termed by Curray the "Louisi-

ana Channel," and the "barrier spit "-shaped ridge slopes from about -57 m
to -65 m.

These interpretations seem plausible, but are not completely acceptable.

The geometry certainly does suggest coastal barriers or strandplain forms.

Their

sloping across the shelf conceivably does represent a regression, but can also
be a tectonic product or partly a product of erosion.

When the two are

viewed together, they define an arcuate outline which may well represent a

delta lobe with which the forms are contemporaneous.

This arcuate lobe is of large scale, even larger than the present Brazos-

Colorado delta area.

The meaning of this scale difference is intriguing.

can be interpreted many different ways.

It

The delta perhaps represents the in-

tegrated discharge of a much greater catchment area than that of the BrazosColorado catchment area of today.

Since the coastal rivers and streams which

formed the delta bulge flowed across

at the time when sea level was-60

100 to

150 km of additional exposed terrain

m or so below its present level, many of the

streams which now flow separately to the sea may have then been integrated into
one or more major trunk streams.

Thus, conceivably, the delta bulge represents

the combined flow and sedimentation not just of the Brazos-Colorado, as Curray'

name would imply, but also of the Trinity, Sabine, Calcasieu, and several other,
lesser, coastal rivers.

90

The size of the delta lobe could partly represent either an abundant

supply of sediment or a long duration of sediment accumulation.

Sediment

supply during a falling stage could well have been higher because of rivers

adjusting to a decreasing base level and may have been higher at other times
for other reasons.

The Stetson Bank, which is in the area of the forms just described, is
a salt-dome structure which reportedly (Edwards, 1971) has Tertiary rock out-

crops at the surface.

It is also the area from which one of the most inter-

esting radiocarbon dates reported from the entire Gulf continental shelf comes
This is the articulated Rangia cuneata sample, first reported by Neuman (1958),

which gave close age results on replicated samples.

-58 m,

The sample elevation was

and the mean of the age determinations was about 12,900 years B.P.

This date is perhaps the best estimate of the age of any of the shoreline

features submerged on the shelf that exists.

that age.

The delta may very well be of

On the other hand, it cannot be denied that the estuarine Rangia-

bearing deposits are possibly not contemporaneous with the delta bulge.
Similar forms sloping across the shelf exist inland from the form as-

sociation just described.

One of these, shown on Volume 3, Plate. 2, was also

described by Curray as a relict barrier spit, flanked on the landward side by

a channel which he termed the "inner Colorado-Brazos channel."

extends from about -26 to -40 m.

This feature

Smaller, less-extended forms exist in the

same depth range in the eastern part of the east Texas sector.
Again, Curray 's interpretations of the nature of the larger of these

forms is regarded here as plausible.

as barrier-strandplain features.

detail do

The smaller ones

are also interpreted

Poor bathymetric control and lack of other

not permit further interpretation.

The areas known as the Freeport Bank and the Freeport Rocks just off-

shore of the Brazos-Colorado Delta are part of one of the most interesting

associations known on the Outer Continental Shelf.

91

The maps presented

by Curray (1960) and Winchester (1971) show that indurated rock outcrops occur
at numerous localities which form linear trends.

Winchester has described the rocks which outcrop off of Freeport, Texas,
in a depth range of -14 to -20 m. These rocks form a linear trend over a distance
of 31 km (Figure 3-8).

Four lithologies predominate in samples:

1)

ortho-

quartzite, 2) fine- to medium-grained calcarenite, 3) coarse calcarenite, and

4)

caliche nodules.

cement.

All four rock types are cemeted by low-magnesium, calcite

Shell materials are affected by recrystallization.

Much of the shell

material may have been derived from the Beaumont formation, which is said to
underly the rocks.

28'42

Nautical Miles

/
95 - 24-

Figure 3-8.

^ZVW

J-

J-

Rock outcrops off of Freeport, Texas, at depths ranging

from -14 to -20 m below sea level.
(After Winchester, 1971)

Winchester makes a comparison of the environment of accumulation of

the beds forming the Freeport Rocks with the present-day beach near High

Island, Texas.

At the High Island beach, much of the coarse, shell fraction

92

is reworked

older shell with Rangia cuneata and Crassostrea virginlca as common

Winchester obtained dates ranging from 3800 to 28,000

occurrences.

years B.P. on shells of these forms from High Island beach.

He feels that

much of this shell is reworked from the Beaumont formation, which outcrops
at a shallow depth near High Island beach.

For the older specimens, this

seems quite probable, although some which give younger ages may be reworked

from Holocene deposits as well.

Caliche nodules are also abundant at

High Island beach, and the stratified beach deposits show the same major
lithographic types as at Freeport Rocks.
The preponderance of reworked shells in such deposits makes shell

dates suspect.

The recrystallization of the shell also casts doubt on

the meaning of radiocarbon dates of such shells.

Winchester points out,

therefore, that the actual age of the Crassostrea virginica shell which

Curray (1961) dated from the Freeport Rocks area is possibly older than

50,000 years B.P., although the indicated date was 26,900

+ 1800 years

B.P.,

which may reflect contaminating "new" carbon added at the time of recrystallization.

Winchester goes on to argue that a date of 22,886 + 431 years B.P.

obtained by him on the low-magnesium, calcite cement of the orthoquartzite
rock type is probably older than the time of formation of the Freeport Rocks
since the source of the cement carbonate must have been shells, many of which

were much older.

Winchester also dates a caliche nodule and argues that since

it formed originally as low-magnesium calcite, it has possibly undergone little

or no subsequent recrystallization or uptake of contaminating carbon.

graphic study of the nodule also revealed no recrystallization.

was dated at 15,857

+ 268 years

B.P.

Winchester states

Petro-

The nodule

that "because the caliche

formed on the Pleistocene Beaumont surface prior to its incorporation into the

sediments of the Freeport Rocks, the age of the Freeport Rocks must be less
than 15,857

+ 268 years B.P."

(p.

220).

93

Nelson and Bray (1970) have described and interpreted the sediments of
the Sabine and Heald Banks offshore from the easternmost part of the Texas
coast.

They interpret a history of sea-level rise over an area previously

entrenched by the Sabine and Calcasieu Rivers.

Cores and marine sonoprobe

records allowed reconstruction of paleotopography and paleoenvironments.

The Ingleside shoreline features which were discussed at length in the

central Texas sector are believed by many to correlate with similar, old,
stranded beach ridges which exist in the east Texas sector.
is not accepted by all, however (Wilkinson

e_t

al.

1975).

This correlation
The Ingleside

features of central Texas terminate in outcrop at the western side of Matagorda

Bay.

From this point, there is no evidence of such stranded beach ridges for

a distance of about 130 km along the coast.

The first occurrence of ridges

in the east Texas area is near the Hoskins Mound, a salt dome just northwest
of the southwestern tip of Galveston Island.

From this point, there are inter-

mittent areas of outcrop of the old littoral features extending all the way
into Louisiana.
It is of interest to note that while in the central Texas sector the

Igleside trend is concave seaward and nearly parallel to the present shoreline, in east Texas and Louisiana it departs from this pattern.

Near Hoskins

Mound the trend diverges in orientation from the present shoreline and
actually shows a slightly convex seaward curvature.

This partly reflects the

effect of the Brazos-Colorado River system, which has influenced this area

throughout the later Quaternary.

Following the old stranded shorelines to

the northeast, the concave seaward curvature again becomes apparent.

94

Graf (1966) has presented borings and a general interpretation of these

In general, he concludes that they were developed

relict shoreline features.

during a phase of deposition of the Beaumont formation.

His borings and

analysis reveal intense weathering and post-depositional alteration of the

shoreline sediments.

This and other evidence led him to suppose that an

interval of relative aridity followed the period of their formation.

West Loui siana Area

A zone of banks, shoals, and knolls extending across the shelf edge of
this area (Volume III, Plate 2) is a continuation of the trend previously dis-

cussed in the East Texas area.

depths of -80 to -200 meters.

The most prominent of these features lie

at

A smaller group, including Phleger and Sweet

Banks, lies seaward of the -200 meter contour.

Another cluster occurs in the

central part of the West Louisiana shelf at shallower depths, between the -50
and -70 meter contours.

Still a fourth group is found around the margins of

the Mississippi Trough, between the -50 and -150 meter contours.

Submerged banks in this area have been studied by oceanographers and marine
geologists from Texas A&M University.

They found that the banks, like their

counterparts in the East Texas shelf area, exhibit features interpreted as submerged, wave-cut terraces and escarpments.

Figure 3-9 (After Poag, 1973) il-

lustrates 3.5kHz sub-bottom profiles of Sweet Bank and Bank

Plate

2.

Bank

(See Volume III,

lies between the -80 and -100 meter contours immediately east

of the East Texas- West Louisiana boundary line).

can be distinguished on the profiles.

deposits

3.

Several well-defined terraces

While they may be capped with reef-like

most if not all of the knolls had their origin as shale or salt

diapirs.

95

The Mississippi Trough, a major feature of this area, is believed to have

formed during intervals of low sea stand and, as the name implies, was cut by the

Mississippi River.

When formed, the river was not building a subaerial delta,

but rather was discharging directly into the head of the trough and a great

submarine fan formed in the deep waters of the continental rise.

This in-

terpretation was developed by Fisk (1956) and has been generally accepted by
later researchers.

The situation at the mouth of the Mississippi during these

low sea stand intervals was comparable to that found at the mouth of the

present-day Congo River.

'gas

seep

Figure 3-9.

3.5kHz sub-bottom profiles of Sweet Bank and Bank 3. Major

Gas seeps along
and minor terrace levels can be distinguished.
the flanks are believed to be associated with structural activity.
(After Poag, 1973.)

96

There is some evidence that the trough functioned during the interval from

25,000 to 15,000 years B.P.

Supporting data come

vicinity of the modern birdfoot delta.

from deep borings in the

Radiocarbon dates from these borings

indicate that during this interval clay minerals characteristic of the

Mississippi, Alabama, Tfest Florida shelf areas were sweeping to the west
(Morgan, Coleman, and Gagliano, 1963).

The best explanation for this condi-

tion is that Mississippi River sediment was being funneled into the deep Gulf.
The middle and inner areas of the very broad West Louisiana shelf are

dominated by six large delta bulges or lobes.

One of the lobes is located

south of Cameron, Louisiana, and lies largely landward of the -20 meter
contour.

At least three prominent, transgressive sand complexes help to

distinguish this lobe.

A pair of smaller, but very well defined delta lobes lies inside of the
-20 meter contour south of Vermilion and Cote Blanche Bays.

They are capped

by transgressive sand complexes, the most distinctive of which is Trinity

Shoal.

Tiger Shoal, which is also part of this lobe, is a curious "Y"-shaped

feature which may represent the bifurcated, relict natural levee ridges of a

major distriburary channel.

Two overlapping lobes lie south of Houma, Louisiana.

They are landward

of the -20 meter contour and are capped by multiple transgressive complexes.

The transgressive complexes on the inner lobe are known collectively as Ship
Shoal.

The sixth lobe is found between the -20 and -40 meter contour south of

Grand Lake and White Lake.

Not only does it occur in deeper water than the

other five, but it also lacks the transgressive complexes.

Numerous reports

of "rocks" from this area of the shelf may be calcium carbonate nodules, ferrous

nodules, or possibly slabs of cemented beach sand.

97

The first two instances

could indicate a subaerially weathered surface; the third, erosion of trans-

gressive sands.
It is interesting to note that relict delta lobes on the low coastal

terraces of Texas and Louisiana formed by the Brazos, Trinity, Sabine, and
other coastal plain rivers also lack transgressive sand complexes.

Such lobes

may be characteristic of intervals of falling sea level, while those with

transgressive sand complexes form during stillstands or periods of rising
sea level.

Thus, we may speculate that the first five deltas discussed above

formed during relative stillstands or periods of slow rise, while the sixth
formed under falling sea level conditions.
Even though this area of the shelf has been extensively investigated by
coring, geophysical, and remote-sensing techniques, the lobes are not well

described in the literature.

Fisk (1955) showed in illustrations how deltas

formed south of Houma, Louisiana, after the Mississippi River abandoned the
trough (following the Wisconsin glacial maximum).

have followed this interpretation.

Curray (1960) and others

Curray (1960) shows the lower of this

pair of lobes forming around 12,000 years B.P. and remaining active until at
least 9,000 years B.P.

According to Curray, the inner delta, the ship shoal

delta, became active around 8,000 years B.P.

The lobes south of Vermilion and Cote Blanche Bays have been referred
to as the Maringouin Delta Complex.

8,000 to 6,000 years B.P.

Dates for this complex are given as

(Coleman, 1966; Saucier, 1974; Frazier,

1974).

Several authors (Jones et 1., 1974; Saucier, 1977) have shown that relict

Mississippi River channels are associated with the Prairie Terrace of south-

western Louisiana.

Jones interprets Mississippi River scars as far west as

Mud Lake (west of Cameron, Louisiana).

At least by inference then, the two

western lobes may be related to the Mississippi River and their age equivalent
to the onshore Prairie Terrace.

98

The West Louisiana shelf area has been subjected to both rapid subsidence
and estensive diapiric movements as a result of sediment loading.

Major

fault zones occur, in predominately east-west trends. The boundary between

this unit and the East Texas unit, marked by an alignment of large diapiric

structures, may also be a major zone of structural weakness.

Uplifted diapiric

structures have formed both true islands and prominent relief features in the

otherwise flat coastal terrain.

Onshore salt-dome prominences are known to

have important archeological sites.

We can assume that the same is true for

at least some of those on the continental shelf.

East Louisiana Area

This area encompasses the Late Holocene deltaic plain and subaqueous

deltaic area of the Mississippi River system (Figure 3-10).

Surface features

and surficial sediments are generally less than 4,000 years old.
The Holocene history is very complex but has been well described in the

literature.

For summaries the reader is referred to the works of Fisk and

McFarlan, 1955; Kolb and van Lopik, 1958; Bernard and LeBlanc, 1965; Gould,
1970; and Frazier, 1967,

1974

Deltaic processes and the earlier sequence

of delta lobes are discussed in Chapter

of this volume.

As shown in Figure 3-10, the landmass of southeastern Louisiana is com-

posed of overlapping delta lobes.

The distal ends of the older lobes have

eroded and subsided, so that former land areas now lie below the shallow inshore waters of the continental shelf.
St.

For example, the landmass of the

Bernard delta (Figure 3-10) once extended beyond the present position of

the Chandeleur Island.

Marksville Period archeological sites (circa 2,000

years B.P.) are known to have existed along distributaries of this former

delta (Mclntire, 1959; Saucier, 1963).

times.

Occupation continued through historic

Thus, prehistoric terrestrial sites inhabited during active

99

buildout

100

of the old deltas and during early stages of their deterioration can be antici-

pated in shallow shelf areas.

The Modern Birdfoot Delta (Figure 3-10) has been constructed by sedimen-

tation in the vicinity of the active outlets of the Mississippi largely during

historic times.

Old light houses, port towns, and other historic buildings

and archeological sites are known in the Modern delta area.

these, such as the port town of Balize

A number of

(1734-1888 A.D.) have subsided and been

covered with silts and clays, to be incorporated within the sedimentary deposits
of the delta mass.

We can reasonably anticipate that a large number of prehistoric terrestrial

sites have also been encapsulated in the alluvial deposits of older delta lobes.

Figure 3-11 illustrates a geological section through southeastern Louisiana and

shows the great thicknesses of sediment that has accumulated during late

Quaternary times.

Because of the combination of high subsidence rates and

rapid deposition, surface features of older delta are typically buried.

This

situation contrasts with conditions found in more stable shelf areas, where
relict terrestrial features may retain their form on the bottom or be reworked
by marine processes after they have been drowned.

TERRACED

DELTAIC PLAIN

PLAIN

GULF OF MEXICO

^-BIRDFOOT DELTA

SEA LEVEL-

-100
500'

DOWNWARPED WEATHERED SURFACE

(T0P0F PRAIRIE

1000'

t-300
50

DELTAIC SAND. SILT,

"-Z

-200

FM.)

& CLAY

100
8. MARINE SILT
& CLAY WITH LOCAL

DELTAIC

SAND LENSES

Figure 3-11.

150 MILES

FLUVIAL & STRAND-PLAIN

SAND & GRAVEL

Generalized cross-section through Late Holocene deltaic plain of

the Mississippi River.
Note thick wedge of deltaic sediments
and buried surface of the Prairie Formation.
Location of section
shown in Figure 3-10 (After Fisk and McFarlan, 1955).

101

Figure 3-12 shows downwarping of the Prairie Terrace surface which has
resulted primarily from rapid Holocene sediment deposition in this area.

It

should be noted that the map datum is a buried weathered surface of soil zone
that has been traced through innumerable borings in the deltaic plain and

continental shelf areas.

Until recently most geologists have correlated the

upper buried soil horizon with the surface of the Prairie Terrace and considered it to be the top of the Pleistocene.

As illustrated in the section

shown in Figure 3-13, this simple model is no longer valid.

dicate

Boring data in-

that there are several distinctive buried soil zones within the Late

Pleistocene-Middle Holocene section.

These are related to the complex inter-

play of depositional events and sea level fluctuations (see Frazier, 1974;
Saucier, 1977).

Another feature of special interest in the section shown in Figure 3-13

is the filled and buried trench of the Pearl River.

The section

suggests that

the trench was cut some considerable time after 32,000 years ago and was in
the process of being filled 17,000 years ago.

By about 9,000 years B.P. the

trench had been completely filled and was no longer an active feature at the

place where it is crossed by the line of section.

In fact, Frazier (1974)

indicates that there is a significant hiatal surface across the top of the

trench surface and at the base of the Recent (i.e., post 8,400 years B.P.).
Part of this same section has been encountered in borings at South Pass
in the Birdfoot Delta area.

The modern delta has extended seaward beyond the

former edge of the continental shelf and provides a natural platform for

drilling into now-buried shelf deposits.

Detailed studies of samples from

borings at South Pass have been made (Morgan, Coleman, and Gagliano, 1968).
Figure 3-14 illustrates the reconstructed section.

Sediments associated with

the modern delta extend to depths of -76 to -107 meters.

102

These are directly

'

CO
'J)

a
B

o
H
o3

/""

cu

U"|

m
o>
H

ft
ex
H

T3

U)
0)

#.

03

J-i

CO

0)

CO

Fn

f)

>-l

03

03

S
> Td
c
fi

^5

0)

4J

n)

P^

&3

J-l

OJ

OJ

4J

<4-<

03
H-l

<J
S

^u
nj
<u

03

03

01

T3

,>

0)
cu

rH

a O
o3
14-1

.C
1

!-i

01

^
O

CO

XI
(1)

E
n o

v-i

o3

M-l

U
ft id
eu

IW

o
ft
S-i

en
!5

3
H

a
QJ

U
QJ

T3

a
CM
:->

OJ

u
3

60
H
p4

103

Cfi
[fl

B'

s
I

BORINGS

SEA
LEVEL
100 -

100

200

SCALE

CLAYEY PEAT.
PEATY CLAY.
ORGANIC MUCK

PEAT

-=1 PRODELTA SILTY CLAY

BAY

SILT.

CLAY. & SHELL

MILES

INORGANIC CLAY

DELTA FRONT
"=r-. .-r-

SILTY SAND & SILTY CLAY

_-_---_

NATURAL LEVEE
SILTY CLAY

r^^^l DISTRIBUTARY MOUTH

BAR SILTY SAND

BARRIER. TIDAL DELTA.

CE3 OR

WEATHERED & ERODED

PLEISTOCENE SURFACE

Figure 3-13.

IN

STRANDLINE SAND

q RADIOCARBON

AGE, YEARS

BEFORE PRESENT

SIGNIFICANT HIATAL SURFACE

AT BASE OF RECENT DEPOSITS

Section through southeastern Louisiana illustrating offlapping

pools of coastal and deltaic sediment. Note multiple weathered
zones and hiatal surfaces.
Location of section shown in Figure
3-10.
(After Frazier, 1974.)

104

MGL-

i. \

STRATKiRAPHIC
UNIT
ORGANIC CLAY

ENVIRONMENT
BAY & MARSH

LITHOLOGY
f* ;*_|X_

PLANT ROOTS HXTO

CHANNEL &
DISTRIBUTARY

"

>?

CLAY

DELTA FRONT

PRODELTA

V)

BAR "SANDS"

MOUTH BAR

'/

SHELF

, .

LAMINATED
& CLAYS

SILTS

CLAY UNIT

100

SILT

&

SANDCZD

BURROWS

4)

E
v

'

> 37,000
CLAY UNIT

200

26.500 + 1350
>37,000, > 37,000

DELTAIC &

SHALLOW

MARINE

Figure 3-14.

HORIZON
CLAY UNIT

SHELL

algae \nzm

II

*
s

25,250 + 1450

-u

IT

15,175 +

15,175+575
500

ALGAL

REEF

SHELL
III

ZONE

STRAND PLAIN
SANDS

PEAT &

WOOD

FRAGMENTS
CARBON -14

C*J

DATES

Shallow
Depositional sequence shown by borings at South Pass.
approximately
shelf
deposits
dated
at
15,000-15,500
continental
years B.P. are overlain by Late Holocene Mississippi River deltaic
(After Morgan, Coleman,
deposits.
Location shown in Figure 3-10.
Gagliano,
and
1968a and 1968b).

underlain by older shelf deposits introduced by the St. Bernard delta system.
At depths of approximately -119 to -122 meters is a very distinctive shell

horizon representing a hiatus at the top of the Pleistocene.

horizon, dated at 15,175

This shell

575 and 15,575 + 500 years B.P.^has been encountered

in numerous other borings throughout the Birdfoot Delta area.

Within the shell

horizon, and below to a depth of approximately -141 meters, is a distinctive

clay mineral suite.

The clays in this part of the section (Clay Unit II) are

higher in kaolonite than the montmorolonite-illite-rich Mississippi River

clays of the sections above (Clay Unit I) and below (Clay Unit

III)

These

kaolonite rich clays represent an eastern. Gulf suite and are believed to have
accumulated during a time when Mississippi sediments were being diverted into
deep waters through the trough.

105

A minor hiatus is represented by a thin shell bed and a change in clay

mineralogy at a depth of approximately -145 meters.
this shell bed are 25,250

+ 1,450 and

>

Radiocarbon dates from

37,000 years B.P.

Still lower in the section (-177 meters) is a third hiatus represented

by an algae reef zone consisting largely of the genus Lithothamnion overlying

sandplain
>

Radiocarbon dates from this zone are 26,500 + 1,350 and

sands.

37,000 years B.P.

The three hiatal surfaces in the lower part of this section are the

most significant to the present study.

Figure 3-15 shows a section through the St. Bernard delta complex.

The

slope of the weathered surface of the Pleistocene has been mapped in this
area, as has the trench of the Pearl River.

Of particular interest is a sand

body associated with a relict shoreline at a depth of approximately -76 meters.

Saucier (1977) and Kolb

e_t

al.

(1975)have identified two weathered zones

within the upper Pleistocene of the Lake Pontchartrain area(Figure 3-16

).

Saucier

indicates that radiocarbon assays of marine deposits between the two

weathered horizons include dates of 27,000 + 1,200

B.P., plus at least one date

>

30,000 years B.P.

and 29,300 + 2,000 years

While conceding that dates

in this time range are questionable because of possible contamination, Saucier

believes that designation of the deposits between the two weathered horizons as

Farmdalian is justifiable on the basis of stratigraphic position.

Kolb suggests

the same interpretation but is more cautious and concludes that many more

radiocarbon dates are needed and much more boring data must be analyzed before
the matter is resolved.

The two weathered Pleistocene horizons from the Lake

Pontchartrain area seem consistent with Frazier's section illustrated in

Figure 3-13.

Figure 3-17
Pontchartrain.

shows a section through an area on the south side of Lake

Here a Late Holocene barrier-island complex rests on a weathered

106

saaiaw

X)

CD

cO

3
CD

H&0

CO
4-1

Pn

H c
3

CO

13
0)

-C

CD

CD

Xi
4J

CO
CD

o
4-1

4-1

CD

an

4-1

CO

c
3 o

J-i

Xi

^O
H
X

4-1

CO

o
o
hJ

4-1

0)
,3
4-1

CD
4-1

u
3
QJ
M

4-1

>

u
0)

X
<d
T-\

H
Pi

(X iH

o
o

J-i

CO
CD

cd
4-1

CD

iH

CD

4-1

^V

XI

X)

4-1

<r
r-~

)-l

CO

c
u

u
0)

c^
^H

4-1

u o

CQ

CO

l-i

CD

CO

4-1

X.

C/>

CO

CD

Fn

U
42
60

>

3 H h
cu
O
U u
A c 4-1
4-1

4-)

4-)

3
o

<

4-i

CO

H 13

o
H

4-1

ro

w X
<D

m
I

ro
CU

H
3
DO
H
Fn

107

CD

4-)

^y

SCALE
10

400+00

800+00
CENTER-LINE STATIONS, FEET

E'

1-10

^*
Vertical

900+00

1000+00

BEACH

'\S~\
Scale, Feet

1200+00

LAKE PONTCHARTRAIN BRIDGE- ROUTE

Hon*

15

DEPOSITS'"
Exagg =

50 1
1

100+00

CENTER- LINE STATIONS, FEET

Figure 3-16

Pont char train Basin area cross-sections showing weathered

horizons (After Saucier, 1977).

108

IViSVODVaiNI

rt
14-1

u
CO
1

U
cO
cu

c
00

C
&
O

QYOailVd N *

J=
co
CO
CU
>-i

cO

aovAnvs noAva

4-1

CO
CO

w
CO

C
cO

CU

rH

NOOOVl

QNIIfl

y)

CO

0)

cu
S-i

1VNY3 3NH3dld
svo
ivanivN N3Hinos

0)

X.
4-1

bo
H
Ph

-C
oc

)-l

cj

J=

4-)

j=

c
H

en

4-1

a
cu
CO

1VNVD

CO
CO

1VNVD

H
4-1

CO

O
O
iJ

^
AVMHOIH

X. Xi

Oil

4-)

cO
S-i

CO
1

x;
4-1

oc
H
4-1

CO

u M

2
avoaiiva N83Hinos
l^

tH
1

NIYaiayH3iNOd 3*V1

CU
)-l

3
00
Pm

133d

109

4-1

CO

and eroded surface.

5,000 years B.P.

The barrier island is believed to have been active circa

This feature was subsequently covered by sediment as the

early St. Bernard Delta built out into the area.

Mississippi

Alabama

West Florida Area

This shelf area displays remarkably well-preserved relict topography.

There are several factors related to its occurrence.

relatively high wave energy.

The area is one of

Through much of Late Quaternary history it has

been characterized by sandy coasts with relatively low input of fine-grained

sediment.

The combination of high energy and a good sand supply has resulted

in the development of prominent barrier beach complexes.

feature is a major re-entrant in the shelf.

Another distinctive

In the vicinity of Choctawhatchee

Bay the shelf break lies only 40 kilometers from the present shore.

During the course of the present study, newly compiled bathymetric maps
(Outer Continental Shelf Resource Management Maps) at a scale of 1:125,000

were published by the National Oceanic and Atmospheric Administration, National

Ocean Survey.

Using these maps, an interpretation of the large-scale relict

topography of the shelf was made.

The interpretative map along with profiles

is presented in Figures 3-18 and 3-19.

There are several prominent deep-water features.

The most important of

these is the DeSoto Canyon, which is evident from the bowing of contours from

depths of approximately -1,000 m to the canyon head at approximately -450 m.

Immediately west of the DeSoto, three other canyon-like features occur in

water depths from -1,250 to approximately -300 m.

to the present study is a trough,

Of more immediate interest

or canyon-like feature, in shallower water

depths near the edge of the continental shelf extending from approximately
-250 m to -80 m.

Lying due south of the present city of Pensacola, it is

intimately related to a pronounced zone of carbonate knolls, relict shelf-

110

BOTTOM CONTOURS

IN

METERS

RELICT SHELF FEATURES

BARRIER-SPIT COMPLEXES

m
,

TIDAL PASSES

BANKS, SHOALS, &

SHELF EDGE KNOLLS

TERRACES

CHANNELS

Figure 3-18.

^sr^ BEACH RIDGE TRENDS

Interpretation of relict topography in the Alabama-West

Florida area.

Ill

kilometers

A
o

10

20

40

30

A'

50

60

70

TERRACE
(barriof i
lagoon complij)

Oi-

20

40

60

"

PINNACLE

80

ZONES

100

120

140

30

kilometers
160

Figure 3-19.

Profiles and interpretation of relict topography in

Profile locations
the Alabama-West Florida area.
shown in Figure 3-17.

112

edge channels, and escarpments.

three distinctive sub-zones.

Within this broad zone there are at least

The seaward terminus of the deepest channels

lie in depths of approximately -120 to -130 m.

to be branched or dendritic,

Some of these channels tend

their upper ends extending to about -80 m.

In between the channels are carbonate knolls, or pinnacles.

These features

have been previously described by Ludwick (1964) and Ludwick and Walton (1957).
The reefs are clustered in zones approximately 1.5 km wide with an average

relict of

9 m.

Ludwick and Walton studied the pinnacles in the area from 85

to 88 west longitude.

They found that where the pinnacles were present they

were found usually at one or both of two depth zones, -68 m to -84 m, and -97
to -110 m (Figure 3-20).

Figure 3-20.

In the eastern part of the study area pinnacles were

Bathymetric chart of lower pinnacle zone south of

Mobile Bay, Alabama. Depth contours in fathoms.
Supplemental notations in meters. (After Ludwick
and Walton, 1957.)

113

found to be well developed and almost continuous.

To the east, pinnacle de-

velopment was found to be poor, replaced by humps and domes.

Samples taken

from the reef were found to be composed predominantly of calcareous organic

structures such as worm tubes, encrusting Bryozoa and calcareous algae, solitary corals and Foraminifera.

Among the most common constituents were cal-

careous algae of the genus Lithothamnion

The reefs were found to be neither thriving or living reefs, nor wholly
fossil, but rather an intermediate stage.

They are believed to have become

established as reefs during a period of lowered sea level.

Ludwick and Walton

(1957:2097) state that the depth of the water at the living reefs probably

did not exceed -46 m and the shoreline was near the present day -55 m contour
or about 15 km north of the reef.

Reef growth in this area occurred simul-

taneously with reef development at similar depths found elsewhere in the Gulf
of Mexico.

The non-living "West Indian" foraminiferal fauna that occur

in

significant concentrations in the reef area are believed to represent conditions not unlike those existing at present in shallow areas off the coast
of south Florida or in the Florida reefs.

The second sub-zone of escarpments, dendritic channels, and reef -like

structures occurs in depths from -80 m to approximately -60 m.

is at approximately -80 m,

The escarpment

and the dendritic channels extend inland from this.

There are fewer knolls in this zone; their tops are at approximately -60 m.

A third sub-zone of branching, dendritic channels extends inland from

about the -50 m contour to approximately the -40 m contour.

South of Destin,

barrier-spit complexes appear to be associated with this sub-zone and occur

at depths of -46 to -48 m.
is the first in a series of continuous

At depths between -30 and -36 m

and very well defined barrier complexes.

configuration of a double barrier.

Traced laterally, this has the

There is a barrier with a dune complex on

114

the landward side that rises to elevations of approximately -27 m, giving the

overall feature from 6 to 10 m of relief, falling off to depressions which

probably represent an old bay.

It then rises again to a second barrier and

dune complex behind which is a second depression suggesting a former lagoon.

Along strike, the smooth, well-defined form of the barrier is broken periodically by what are believed to be tidal scour features representing old tidal

passes or inlets.

Here the trend of the ridge and trough topography is

transverse to that of the general shoreline.

No doubt, submarine erosion has

continued in these depressions or tidal scour areas even after inundation of

the barrier lagoon shoreline.

Perhaps the most remarkable feature in this area, and among the most

striking on the shelf of the northern Gulf, is a very large relict, barrier

complex lying south of Bonsecour Bay.

crescent or barchan shaped.
is some 45 km.

The barrier is more than 50 km long,

The distance between the horns of the crescent

Bathymetric contours indicate a more or less continuous ridge

extending around the crescent with a crest at depths of -30 to -38 m.

These

are probably remnants of coastal dunes that were formed when the feature was

active and that have been subsequently modified by submarine erosion.

Lying

landward of this "dune ridge" is an area of highly irregular bottom topography

characterized by hills and troughs.

ranging from -42 to -36 m.

Some enclosed depressions have depths

This may have once been an extensive coastal dune

field.

Connected to the southwestern horn of the crescent-shaped barrier is an

accretion fan of relict beach ridges.

The ridges and bars of the fan branch

out toward the west.

Inland from the double barrier, at depths from -26 to -36 m, is another

barrier-lagoon complex at depths between -25 and -27 m.

Another set of escarp-

ments and barriers occurs at a depth of approximately -22 m.

115

A final set lies

The -18, -22, and -27 m shorelines trend

at a depth of approximately -18 m.

into the present shoreline and intersect the present shoreline at an acute
angle.
It appears that the shoreline area east of Destin has been coincident or

reoccupied at several times during the Late Quaternary.

A number of high-prob-

ability areas for site occurrence should be found in this area of the conti-

nental shelf related to the relict features just discussed.

Hyne and Goodell (1967) have previously described the innermost submerged

barrier complex off Choc tawhatchee Bay.

In their paper, they noted the trends

of submerged barrier complexes at -18 m and -27 m.

Still another feature that should be noted in this area is the reported

occurrence of a submerged pine forest west of Panama City.

found in situ tree stumps at a depth of -18 m.
and

Here, divers have

Radiocarbon dates of 36,500

35,700 years B.P. have been obtained from wood samples, while a peat

>

sample after thorough leaching of humic acids yielded a date of

years B.P.

(Shumway et al.

>

40,000

1962).

Relict shorelines trend at an angle across Mississippi Sound and are

truncated and reworked by the present active barriers forming the outer margin
of Mississippi Sound.

The sands of these modern barriers have been derived

from the reworking of older, now-drowned trends.

can be seen on Ship Island.

A remnant of one such trend

This reworking of barrier trends of different age

also accounts for the notable difference in heavy mineral suites found on each
of the Mississippi islands as reported by

Foxworth et al.

1962.

In the onshore

area of Mississippi Sound the present coast is characterized by a major relict

barrier-island and lagoon system.

by Otvos (1972)
3-22).

These have been described in some detail

who believes that they are Sangamon in age (Figures 3-21 and

Similar barrier lagoon features extend along the eastern Alabama and

west Florida coasts.

A series of pronounced hooks or recurve spit complexes

116

cO
CU
>-i

cO

U
CO
CtJ

U
O3
rl

a,
H
CO
CO
rl
CO
CO

CU

&
4J
4-1

O
CO
CU

U
4-1
ft)

cu

o
H
bO

O
O

.H
CU

u
CO

cu

4J
CO

CN
I

en
cu

3
60
H

117

a>

4)

(0

sangamon

k-

prairie surface

3
O

higher
pleistocene
surface

-Q

ridge

complex

eg
ka

10

beach

o
(0

big
ridge

scarp

CITRONELLE FM.
BILOXI FM.

GULFPORT FM.
UNDIFFERENTIATED PRAIRIE

Figure 3-22.

Generalized cross-section through the central Mississippi Gulf

Coast area.
Location of section shown in Figure 3-20.
(Modified
from Otvos, 1972.)

on these old barriers in Choctawhatchee Bay suggests that this water level has

been at this shoreline at least three times.

To the east of the present floodplain of the Pascagoula River is a well

defined relict floodplain of the same river.

is morphologically well defined.

As shown in Figure 3-22, the area

From oldest to youngest they consist of the

Prairie Terrace, a slightly lower surface distinguished by relict floodplain

features called the Pamlico Terrace (Harvey and Nichols, 1960), the Deweyville
Terrace, distinguished by relict meander and scars of large radius; and the

Holocene floodplain and coastal marshes.

Saucier (1977) suggests that the

relict Pascagoula floodplain is a Farmdalian feature (Figure 3-23)

While

there are no radiocarbon dates from the terrace deposits of this area, the

morphologic relationships of the various terraces identify it as a key for

unraveling the Late Quaternary sequence.

118

>

35'

T
88" 25'

"*)

3
I

Prairie terrace

Deweyville terrace

Pamlico terrace

Holocene deposits

Abandoned Pascagoula

River

meander

belt

nd

SCALE

12

4
I

Miles

Figure 3-23.

Terraces of the Pascagoula River area.

1977.)

119

(After Saucier,

West-Central Florida Area

Beginning at water depths of -145 and -135 m

is the first of a series of

bulges that characterize the shelf edge in west-central Florida.

At least

six of these shelf -edge bulges occur; the uppermost lies at a depth of

about -70 m.

Jordan describes these features in a 1951 article and illustrates

fathometer profiles showing surface form.

From the configuration of the con-

tour lines and the fathometer profiles it is not too difficult to interpret

these features as a special type of barrier- spit complex.

They are interpreted

here as relict cuspate-foreland islands similar to the present-day complex

which exists in the vicinity of the Apalachicola Delta (St. Joseph Spit, Cape
San Bias, St. Vincent Island, Cape St. George, St. George Island, and Dog

Island - see Figure 3-24.

and 3-27,

Figure 3-25 is a location map for Figures 3-26

Profiles A and B of Figure 3-26 provide examples of barrier islands

separated from an inner shore by a shallow relict bay or sound. On Profiles A

the ridge and swale topography above -95 m probably represents an accretion

ridge set on the shore side of the former bay.

Profiles C and D of Figure 3-26 appear to represent a beach or barrier

ridge directly against the shore (bay-sound absent).

Examples of barriers,

accretion topography and pronounced escarpments can also be seen in Profiles

E-G of Figure
follows:

3-27.

Depths of sea levels suggested by the profiles are as

Profile A, -95 to -105 m; Profile B, -117; Profiles C and D, -120 m;

Profile E, -134 m; Profiles F and H, -75 m; and Profile G, -70 m.

Schnable and Goodell (1968) have interpreted the Late Quaternary history
of the area in the vicinity of the Apalachicola River

sedimentation has resulted in pulses of progradation.

(Figure

3-24).

Deltaic

Because of an abundant

supply of sand and relatively high wave-energy conditions, major cuspate barrier complexes have formed in the vicinity of the river.

The deltaic sedi-

mentation has created the fortunate situation where the sandy deposits of
older Pleistocene beach complexes are preserved and separated by fine-grained

120

<u

(-

z
u
U
u

??

\;

z
Ul
u
o
H
X
ui
-1
a.

ui

z
Ul
u
o

5 t

li

<

J
s

SE

^
t

io
CO

[
5-1

Ul

?
i

co

CU

* 5

DS

C3]
CO

H
rl

O
rH
En

CO
CD

U
CO

o
u

x>
CO

PC

u
o

u
CO

60
H
<3

o
CO
CO

rH
pq

en
CU

o.
co

o
Q)

.d
14-1

oo
vO

>%<^
60 rH

O
O

rH
CU

60

>

rH
rH
CU

X)
CO
}-4

CO

CN
I

co
CU
5-i

60
H
Ph

121

CJ

Figure 3-25.

29

28

Bathymetric chart of shelf-edge bulge and related

features south of Panama City, Florida. Depth
Supplemental notations in
contours in fathoms.
meters, (After Jordan, 1951).

122

r-A;

--,

4- 82
92

O
X

X
0HI

O
o
CO
I

in

o
LU

<
111

o
o
<

X
LU
LU
-J

<
o
CO
-I

<

o
IQC
LU

>
ilometers

n~- rrrrfgjfi

n~"n~Tr~ri

Figure 3-26.

vt

Bathymetric profiles of shelf-edge bulges and related

(After
For locations see Figure 3-24.
features.
Jordan, 1951.)

123

O
X
<

Q.
LU

O
o
CO
Q

CO

LU
\-

<
LU

o
o

<
X
UJ
LJJ

<

o
<
o
Icc
LU

>

Figure 3-27.

related
Bathymetric profiles of shelf-edge bulges and
(After
3-24.
For locations see Figure
features.
1951.)
Jordan,

124

deposits.

As shown in Figure

plexes have been identified.

through the sequence.

beach com-

3-24, at least three sequences of

Numerous undisturbed cores have been taken

Radiocarbon dates, stratigraphic relationships, and

environmental interpretation suggest that the large

"middle" complex of

beaches was formed during a relatively high stand of the sea that was near,
or slightly higher than,

years B.P.

the present sea level between 24,000 and 40,000

Schnable and Goodell believe that the middle beaches correspond

to the Silver Bluff shoreline of Florida and Georgia.

Saucier

(1977) cites

these features in support of a relatively high sea level stand during the

mid-Wisconsin Farmdalian substage.

this volume

However, as discussed in Chapter

4 of

radiocarbon dates from this time interval have been questioned

by other researchers.
The borings also reveal the position of a filled trench of the Apalachicola
River.

As shown in Figure 3-28,

limestone of Miocene age.

the base of the trench is cut into lithified

Rang i a cuneata shell samples collected just above

the Miocene-Recent unconformity at the base of the beach (-22 m) yielded a

radiocarbon date of 9,950 + 180 years B.P.

Landward of the Farmdalian beach complex are remnants of one or more
older complexes of beach ridges (Figure 3-24).

These are considered to be

Pleistocene by Schnable and Goodell.

Central Florida Area

This area forms the northern part of the Florida Plateau, a broad, flat
area underlain by relatively stable limestone.

The outer shelf consists of

what Price (1954) has termed the downwarped shelf (Figure 3-29).

This is a

terrace-like feature lying between the -200 and -100 m contours.

According

to Gould and Stewart

(1955), the landward margin of this outer shelf is marked

by a set of terraces at -100 to -120 m.

6 m.

Local relief on the terrace is

to

Apparently the -120 m terrace is wave-cut and marks the lowest Pleistocene

125

11

MR

mi

JO 5r

MQ

MK

5 75,

MN

10

10

20

30

40

-15

50

60

70 -2
0J

-O
V)
0)

HO-

4^

a>

-c

a
90

&
-30

100

no

120

130

140

-45
150

Figure

3-28.

Cross-section from Cape San Bias to Cat Point, Florida.

(After Schnabel and Goodell, 1968.)

126

Figure

3-29.

Major geomorphic features of the Central Florida area,

127

sea level (Bergantino, 1971).

In the central

Florida area, the outer shelf is

relatively narrow, but it broadens and has more complex surface topography to
the south.

Tight spacing between the -100 and -85 m contours defines a major escarp-

ment which separates

the downwarped

bottom topography (Price, 1954).

contours.

shelf from a zone characterized by hilly

This zone lies between the -80 and -40 m

There is at least one nose-like protuberance at approximately -60 m,

which may represent a relict barrier-spit complex.

Gould and Stewart (1955)

have also described a set of terraces at the -65 to -85 m level.

The -40 m contour approximates the outer edge of the Middle Ground Uplift

area (Brooks, 1973).

On the edge of this uplift area is the Florida Middle

Ground, an 800 -square km reef with local relief up to 10 m.

This feature is

believed to be a relict reef from Pleistocene time upon which two or more
meters of worm

algal and coral growth has developed during the past 7,000

years (Brooks, 1974:302).

Reefs have been described by Jordan (1953) and

others in the Florida Middle Ground area.

Many isolated reefs rise from general

bottom depths of -37 to -44 m to a common depth of -26 m.

Shoreward of the -40 m contour, the shelf can be described as a gently
sloping, submerged young karst plain.

This part of the shelf, for the most

part, is smooth with relief of as little as 5 to 6 m being rare.

Karst

topography characterizes the bottom within 50 km of shore, and it is here

that maximum local relief occurs.

The coast in the vicinity of Homosassa Bay, about 80 km north of Tarpon

Springs, Florida, consists of flat limestone, extensively pitted with sinkholes

and lacking sandy barriers.

The sinkholes apparently developed during intervals

of lower sea level (Shepard and Wanless, 1971).

To the northwest the Suwanee River and the Fenholloway River also pass

through flat limestone terrain and discharge onto a very gently sloping limestone shelf with exceptionally low wave- energy conditions.

128

Tanner (1960) has

classified this as a zero energy coast.

oyster reefs.

This general area has well-developed

The streams are all entrenched into the carbonate rocks; some

can be traced for about 6.5 km, where it is about 5 m deep.

At this distance

offshore, water depths on either side of the channel are only about 1.2 m deep.

Erosion of the shallow marine bottom is so negligible that even micro-scale

tidal current channels in the more shallow (less than

m) embayments have

been maintained as enduring features of the bottom (Shepard and Wanless, 1970).
There are no areas in the marine environment of the karst where sedi-

mentation has occurred during Late Holocene times.

sult of the sediment starvation of the area.

This of course is the re-

First, there is a lack of

quartzose rock or clayey deposits within the drainage basins which empty into
the area.

Second, although there is a relatively high level of precipitation,

the gradient of the streams is very slight, minimizing flow velocities and

erosion capacity.

Finally, the predominant rock formations, composed of lime-

stone, are soluble in water.

Erosion of the shoreline itself is also negligible.

The limestone forma-

tion presents a hard surface to any physical effects, and there is an absence
of sand and sediment to be transported.

The shoreline is eroded, however, during

drastic changes such as a hurricane storm surge which both raises the water
level in the Gulf and creates higher- energy waves to impinge

upon the shore.

This increased level of wave energy can cross the shelf and be focused upon
a level of the shoreline seldom exposed to wave energy.

The coastal terraces in this area are simply an onshore continuation of

the of f shore, low -relief young karst plain.

with innumerable sinkholes.

These terraces are virtually flat

A few hummocks represent inactive dunes.

On

the east side of the area there is a distinctive boundary between the low

coastal terrace and older uplands (some of which are undifferentiated

Pleistocene coastal terraces).

The boundary is marked by the Pamlico Sand

129

Belt, interpreted to be a beach-dune shoreline feature possibly equivalent

to the Igleside shoreline.

In the central part of the area, between the Waccassa and Fenholloway

Rivers, this old shoreline loses definition where it merges into a riverine
coast.

In this segment a relict delta of the Suwannee River has been identi-

fied by previous researchers.

The Uplands surface is also characterized by numerous well developed

sinkholes.

Sinkholes are of three general types:

sinkhole, and the aquifer sinkhole.

the collapse sinkhole, the solution

The collapse sinkhole (Jon

e_t

al.

1972)

occurs where the limestone roff has caved into a void created by the solution
of limestone.

A solution sinkhole (or doline) occurs where

a soil

mantle set-

tles into the sinkhole at the same rate that the limestone is dissolved away at
the bottom.

The most prevalent in Florida is the aquifer sinkhole where the

water table is high and erodes the overlying sediment and surrounding soil
into the sinkhole.

This results in sediment becoming trapped in the aquifer

system, and none being supplied to the Gulf for marine deposition.

Paleo-Indian projectile points and bones of extinct Pleistocene vertebrates have been found in many of the sinkholes and at numerous locales within
the bottoms of rivers and from river margins both within the uplands and

coastal terrace of this area (Neil, 1964).

North of Tampa there are many sand dunes which occur as isolated features,
in groups, and as extensive dune fields.

The more isolated forms are usually

giant U-shaped or parabolic dunes, which reach widths of as much as

for the most part lie seaward of the Pamlico scarp (approximately 8 m)

km and
.

are partially drowned, being wholly or partially surrounded by sea water.

Examples of drowned dunes are the Cedar Keys in Levy County and Horseshoe

Beach in Dixie County.

130

Some

The dunes are now largely Inactive.

beach-derived sands.

All seem to have been nourished by

Beaches are generally absent from this part of the

Florida coast at present, and it is characterized by a marshy, poorly defined

shoreline developed over shallow carbonate rocks.

The submerged dunes were

probably supplied by a pre-modern beach built at a lower sea level.

These drowned parabolic dunes must have formed after the Silver Bluff
Had they been present during Silver Bluff times, they would have

shoreline.

been destroyed by marine erosion or covered by marine sediment during that

period.

Dunes on the coastal terrace obsecure the Pamlico scarp north of Tarpon
Springs and are believed to have formed during the time of a shore lower
and later than the Pamlico.

Orientation of the dunes suggests that they

were formed by a southwest wind.

South Florida Area

The outer shelf of South Florida consists of an extensive terrace-like
area, which is believed to be a downwarped part of the continental shelf

(Figure 3-30)

The downwarped shelf lies between the -200 and -100 m

contours.

Beginning at about -120 m, it has a series of intermittent ridges along

its outermost edge.

These ridges may have a relief of as much as 18 m.

Per-

haps the best known of the ridges is Howell Hook, which has been described
by Jordan and Stewart (1959) as a residual barrier-spit complex formed when

Pleistocene sea level was at its lowest stand (Figure 3-31)

The other ridges

have been interpreted as barriers and bars from approximately this same time
(Bergantino, 1971).

Jordan and Stewart (1959:980-1) describe Howell Hook as

".

an arcuate

ridge 65 miles (105 km) long impounding a 'lagoon' with a pronounced 'lagoon
channel'.

The ridge crest and the bottom of the 'lagoon' are generally

131

Figure

3-30.

Major geomorphic features of the south Florida area.

132

Figure

3-31. Bathymetric chart of the Howell

Hook area, south Florida.
Depth
contours in fathoms.
Supplemental
notations in meters.
(After Jordan
and Stewart, 1959.

133

smooth, but there are some isolated rises.

...

'lagoon channel' is 30 feet (9 m) deep

The deepest section of the

a channel is also indicated

by an embayment in the 100-fathom (183 m) contour

."

East of Howell Hook is an elongate 130 sq. km area of reef patches

(Jordan and Stewart, 1959).

There are also isolated patches near the

The large reef area lies in water depths of -137 to -156 m.

"lagoon" feature.

Local relief of the reefs is

A similar patch-reef zone occurs

to 5 m.

160 km to the north at -110 to -128 m.

As in the Central Florida area, the outer continental shelf is separated

from the middle modified karst shelf by a zone of very well-defined escarpments and terraces.

One of the most continuous and best-defined escarpments

on the shelf occurs at about -85 m in this area.

for some 240 km.

and -70 m.

It can be traced continuously

A second, but less continuous, escarpment occurs between -75

Lying between the two escarpments is a flat, terrace-like feature.

At about -80 m are several nose-like protuberances at the northern end

Toward the south at the same depth is a pronounced ridge-like

of the area.

feature.

Both the nose-like features and the ridge are interpreted as barrier-

beach complexes.

Two additional nose-shaped features occur at about -70 m.

The northernmost of these "climbs" contours to -60 m.

North of this contour-

climbing feature are two additional nose-shaped features associated with the
-60 m contour.

It will be recalled that Curray (1960) noted similar contour-

climbing features in the Texas area.

All of these nose-shaped features are

interpreted as barrier-spit complexes.

The configuration of these features

indicates a dominant longshore drift pattern from north to south.

The middle shelf, between about -55 and -30

Price (19

54)

m, is

has called modified karst topography.

karst are indicated in Figure

3-30.

characterized by what
Areas of most distinctive

These areas may correspond to the

of former bays where coquinas and other lime sediments particularly susceptible
to differential solution may have been deposited.

134

Between -40 and -18 m are a number of features which suggest barrier-spit
complexes

and shore trends.

Sinkholes are known to exist in the middle and inner shelf, though they
are not thought to be extensively developed.

Evidence suggests that sinkholes

and caverns become more frequent closer to the shore (Brooks, 1974:302).

There is a submarine spring located off Fort Meyers Beach.

This

feature, known as the Mud Hole, is reported to be saline (chlorinates of

19-20 o/oo), hot (97 F)

and apparently rich in trace metals.

Pyle, Bryant,

and Antoine (1974:298) note that the Mud Hole and several "deep holes" reported

by fishermen all lie within a belt 16 to 65 km offshore.

A number of onshore springs in this area have produced Paleo-Indian artifacts and bones of extinct Pleistocene animals (Neil, 1964).

Among the most

important of these are Warm Mineral Springs and Little Salt Spring, where

systematic underwater excavations have been conducted (see Chapter

6)

In the Venice-Naples area prehistoric sites have been reported in shallow

offshore areas.

Local collectors have found chert debitage, scrapers and

projectile points in shoals near Naples, Florida.

Ruppe (personal communi-

cation) has conducted systematic excavations in a submerged site in

of water offshore from Venice, Florida.

to 3 m

The site is a shell midden and, among

other things, has produced pottery from the early ceramic Orange Period.

Numerous artifacts and fossil bones have been dredged up in Tampa Bay.
The artifacts range from Paleo-Indian projectile points to ceramics.

Bones

of extinct animals have also been reported from the dredge spoils.

The southwestern part of penisular Florida between the Florida Keys and

Cape Romano is one of the most complex coastal areas in the United States.

This is an area of tidal channels and mangrove swamps, but there are open
lagoons.
In the western portion, a series of quartz and islands and shoals pro-

tected by vermetid reefs (Shier, 1969) has developed seaward of the

135

The most unusual feature of the area is a series of large islands

mangroves.

composed of oyster shells that lie more or less in a straight line within the

mangroves from Goodland Island to Chokoloskee Island.

only habitable land.

These islands are the

Their relief is the product of Indian midden accumula-

tions upon oyster bars subsequent to 1,200 A.D.

The record of the Late Holecene transgression is particularly good in

the western margin of the Everglades.

According to Scholl (1964) the sedi-

mentary succession from the mangrove swamps of southwestern Florida attests

to a 3 meter rise in sea level during the last 4,000 years.

The sequence

of transgressive sediments (Figure 3-32) consists of calcitic mud and fresh-

water peat deposited on bedrock.

On top of this layer is a unit of basal

FRESH -WATER
-MARINE

SWAMP ENVIRONMENT

PARALIC

SWAMP
ENVIRONMENT

SWAMP ENVIRONMENT

MANGROVE FOREST

MARINE DEPOSITS

. <T>

(fi

<T>

_ /TV

.'T>

a\

,M

SMELLY CLOLUTCC0US
Smelly n*E-GPAINCO
OUARTf SANO

S'LT

TO

121

FEET

fIBPOUS MANGROVE PEAT

W^

Figure 3-32.

ANO
FRESH-WATER PEAT
CALCIT.C -UK.

METERS

Idealized cross-section of coastal mangrove swamps

in the Ten Thousand Island area of the south Florida
coast.
The transgressive sequence consists of freshwater deposits overlying bedrock which, in turn, are
overlain by paralic swamps and marine deposits. Mangrove peat begins to form over freshwater calcite mud
(After Sholl and Stuvier,
essentially at mean sea level.
1967).

fibrous peat, largely derived from mangrove and other rooted plants, and an

overlying marine unit of peaty and calcareous shell debris (Whitewater Bay) or
shelly, quartz-rich sand and silt (Ten Thousand Islands area).

136

Judging from

radiocarbon dates (Scholl, 1964a), the mangrove peat unit began to form 3,0003,400 years ago after cessation of calcitic mud formation.

Within a period of

formation of mangrove fibrous peat in areas

a few hundred to a thousand years,

which are now waterways and intra-forest bays gave way to the deposition of
shelly brackish-water and marine sediments of the upper member of the transgressive sequence.

The environmental shift from freshwater to brackish-water

and marine milieus came about in response to a more or less steady rise in
sea level and marine inundation of former mainland swamps.

Because a con-

siderable body of evidence points to the probable tectonic stability of

southern Florida in Recent time, the recorded submergence is regarded as a

measure of an eustatic change in sea level (Scholl and Stuiver, 1967).

Based

on the age and elevation of fibrous peat overlying bedrock and freshwater

calcitic sediment, the rise in sea level across southwestern Florida 4,400 3,500 B.P. was 30 cm /100 years.

About 3,500 B.P., when sea level stood

1.6 m below its contemporary position, the rate of rise diminished by a

factor of five; since 1,700 B.P., the rate of rise has averaged only about
3

cm /100 years.

Figure 3-33
Cape Sable.

illustrates the positions of relict

shorelines on

Note that Shoreline Z has been dated at 4,950 + 120 years B.P.

The shoreline has apparently prograded seaward since that time and the three
capes have developed on the island.

The ages of relict Shorelines W, X and

Y remain to be established.

The Florida peninsula represents just part of a much larger geological

feature known as the Floridian Plateau.

The southeastern part of this plateau

has long been an area of deposition which shows evidence of subsidence.

Between Miami and Key West, bottom contours indicate the existence of a

crescent-shaped plateau, the Pourtales Plateau, the surface of which slopes

gently from a depth of about -180 m to -550 m.

Off Key West, at the outer

edge of the plateau, large sinkholes have been discovered at a depth of _ 250 m.

137

1980

100

Northwest

Cape

1610

100
Middle

Cape

rfewafer

Say

sea

peat

50
meters
vertical exaggeration

Figure

3-33,

12x

Cape Sable beach and relict shoreline feaA. Relict shorelines with radiotures.
carbon dates in years B.P. associated with
features.
Shorelines W-Z are low-carbonate
Shorelines of individual capes
mud ridges.
are shell beach ridges. (After Smith, 1968).
B. Typical cross-section showing modern and
buried peat deposits. (After Spackman e_t
al., 1964).

138

These holes, averaging 1 km in diameter and 140 to 170 m in depth, are evi-

dence that the Pourtales Plateau was once subaerially exposed and that therefore the plateau is more likely due to subsidence and downfaulting than to

erosion by the Florida Current which runs along the plateau (Hoffmeister
et al

1964; Jordan

e_t

al.

1964).

The Florida Keys form an arcuate, discontinuous band which follows the

shape of the Pourtales Plateau (Figure

3-34).

They are composed of two

distinct lithologic units, the Key Largo Limestone and the Miami Oolite
(Hoffmeister and Multer, 1964).

At the northern end of Miami the two diverge,

The more inland feature, the Miami Oolite ridge, was a marine limestone bank

which was lithified during the last interglacial about 100,000 years B.P.
The ridge curves to the southwest for 50 km and then westward to Cape Sable,
the southwest point of the Florida mainland.

It then overlies the Lower

Keys from Big Pine Key to Key West.

The Key Largo Limestone is an elevated, fossilized Pleistocene coral

reef of about the same age as the Oolite, being eastward and in slightly

deeper water.

At Miami Beach, it is found 3 m below sea level.

The ridge

extends 25 km south from Miami as a shoal which separates Biscayne Bay from
the Atlantic Ocean.

It then surfaces at Soldier Key and is exposed in the

Florida Keys for 200 km to Big Pine Key.

From Big Pine Key to Key West,

the limestone is submerged below the Miami Oolite.

The Key Largo formation

is also found in shallow water along the eastern shore of the Florida main-

land extending southward from Miami for about 65 km.

At the southwestern extremity of the Keys are ring-shaped reefs called
the Marquesas and Dry Tortugas.

correctly called atolls.

These are coral reefs which have been in-

They are formed on shallow sediment banks and un-

like atolls, are not associated with subsidence, nor are they fringed by

deep water (Smith, 1971).

139

cd

+-M
<u|
t-i

Q)
4->

W
H
cu

B
M-l

o
32
5-i

0)
4->

I
a
o
u
H

B
<y
5-i

H
a>

xi
+j

id
CO

H
J-i

O
iH
QJ

H
I

00

a>
5-i

3
H

140

While quartz is by far the major constituent of the beaches of Florida,

it represents only a small percentage of the beach material on the Keys.

On

both the east and west coasts of the state, the percentage of quartz decreases
until just south of Miami Beach it comprises much less than 50% of the material.

On Key Biscayne, quartz ceases to be an important beach constituent.

Still farther south, as well as at Cape Sable on the mainland, the beach is

almost entirely calcium carbonate in the form of coral, shell fragment, and

remains of foraminifera.

As an example, Martens (1936) cites a figure of

97.52% calcium carbonate for Upper Matecumbe Key.

The Florida Keys and the modern reefs represent a special case of fringing

Fringing reefs are formed by corals close to land and

reef (Smith, 1971).

in shallow water.

Their growth is directed toward the water's surface and

outward toward the open ocean.

This is a direct result of favorable condi-

tions seaward and upward and unfavorable conditions of increased temperatures,

salinity changes, and sediment deposition landward.

figuration shown in Figure 3-35.

The result is the con-

A broad platform of coral rock is formed

extending horizontally in a seaward direction.

The platform

continues to

grow until its base reaches a depth of about -27 m (Smith, 1971), the depth

*>,.

LAND
SEA
LEVEL

REEF

io' $!$! it

..

Figure 3-35.

Development of a fringing reef.

Broken lines
indicate stages of growth.
Rubble at the edge
of the reef forms a base for farther extension
in deeper waters.
(From Smith, 1971)
141

beyond which corals can no longer flourish.

Farther seaward extension may

take place as a result of broken coral forming a base of dead rock upon which

more coral may grow.

In addition, large-scale movement of land or sea level

can create conditions for farther extension of the platform.

The reefs of Florida, both the Keys and the modern reefs, are bank reefs

which differ somewhat from the usual fringing variety in that they were formed
This area lies within a marginal belt

farther from the shore (Smith, 1971).

between lower and higher latitudes which, as a result, was greatly affected
by the glacial ages.

During the glacial ages, the low latitudes remained

relatively warm and were able to support coral growth.

however, corals died, leaving the shore

unprotected.

In the marginal areas,

Waves wore away the

coast, creating a platform just below sea level and adding debris to the

offshore region.
Both the Keys and the living reefs are located upon a platform which was
once covered by a shallow sea.

During the

past

million years, the platform

has been subjected to periodic exposure and flooding as a result of with-

drawal of water during glacial periods and its release during interglacial
times.

Pleistocene reefs which grew over extensive parts of this platform

formed a thick layer of rock.

During the last glacial age, when sea level

was low, a platform was cut into the land.

When the ice caps melted for the

It now forms the base for the modern

last time, the platform was flooded.

reefs growing off the Keys.

A north-south section from the Florida mainland reveals the environment

described below (Figure
ridge rising

to 5

3-36)..

The southern

m above sea level.

coast of Florida is a curving

This ridge encloses the Everglades,

a freshwater swamp the surface of which is about 1 m above sea level.

Be-

yond the southern border of the Florida coast are the Keys, ranging 8 to 50 km

from the shore and extending westward beyond the western shore of the mainland

142

(A)

ySOti*^

Figure

3-36.

Map of Florida, a - southern coast,

- Keys, a'
living reef; d-d' - bryozoan and oolitic facies; e - Everglades,
- shoal water, e" - ship channel;
e
GSS - Gulf Stream and Florida Current.
(A)

Cross-section of Florida along the

north-south line of (A). Lettering is
(A)
Dotted lines inthe same as in
dicate hypothetical former conditions.
(Modified from Grabau, 1960.)
(B)

143

(Ginsburg, 1964).

The bay is dotted with small, low mangrove islands.

The

roots of the mangrove trees trap large amounts of sediment carried by currents.

Sedimentation has progressed to such an extent that a large portion

of Florida Bay forms mud flats at low tide.

Seaward of the Keys, at a distance ranging from 5 to 25 km, lies a

line of living coral reefs consisting largely of the branching coral Porites
and the calcareous red alga Lithothamnion
out of the water in a few places.

channel

9 to

10 m deep.

These reefs are submerged, rising

Between them and the Keys is a narrow

The channel represents an area of sedimentation of

coral debris, shells, and other calcareous material.

Seaward of this living

reef, the bottom slopes rapidly into the abyssal depths of the Florida Straits.

The Pleistocene coral reef called the Key Largo Limestone shows a large

variation in thickness.

At Key West, it is about 55 m thick; at Grassy Key,

52 m; at middle Key Largo, 21 m; and at the northern tip of Key Large, 44 m.

The composition of the Key Largo Limestone is typical of coral reefs

(Hoffmeister and Multer, 1964).

It is made up of massive coral heads many

of which are surrounded by smaller coral colonies, shells and shell fragments.

Reef-building corals are found in the formation from top to bottom but are
most prolific in the upper two-thirds of the structure.

Although the Key Largo Limestone represents an elevated coral reef,

probably less than one-fourth of its total mass is derived from coral (Cooke
and Mossom, 1929).

careous algae.

A large contribution to its structure was made by cal-

The great bulk of the material was derived from sea water

by a variety of organisms.
Since the solidification of the rock, solution and redeposition has
taken place.

In some places, the rock contains heads of coral replaced by

calcite that shows the structure of the original head.

In other places, the

rock is a breccia composed of angular fragments in a lime cement.

144

The breccias

represent loose material that has fallen into solution cavities and has been
recemented.
The surface of the Key Largo Limestone (and the Miami Oolite

as well)

is protected by a laminated crust averaging 3 cm in thickness and in places,

up to 13 cm.

It is believed to have been formed largely in the intertidal

zone, but the same condition has been reproduced in the laboratory under
In general, the lime-

subaerial conditions (Hoffmeister and Multer, 1964).

stone is found to be harder within 50 to 60 cm of the surface than below

(Cooke and Mossom, 1929).

Changes in salinity and other ecologic conditions were at times re-

sponsible for the encroachment of corals westward upon the bryozoan community.

When conditions were reversed, the bryozoans dominated again and forced the
corals to retreat eastward.

This process is evident in several intrusions

of Key Largo Limestone into the bryozoan f acies

As the bryozoan

f acies

ibid )

increased in thickness, oolites formed in ever-

increasing amounts in the area which is now the Atlantic Coastal Ridge, a
ridge extending in a northeast-southwest direction.

While this mound was

being formed, bryozoans flourished to the west where they laid down thick
deposits.

During a subsequent glacial period, sea level was lowered and these
deposits were exposed to the atmosphere.

Rain water flowing through the

interstices of rock precipitated calcite around the grains and formed indurated rock seen there today.

During this period, the eastern side of the

oolitic mound was considerably eroded by wave action.

The Lower Keys from Big Pine Key to Key West

different from the Upper Keys.

show some features strikingly

The Upper Keys form essentially an arcuate thin

line and are oriented in a northeast-southwest direction.

The Lower Keys, on

the other hand, form roughly a triangle with Key West at the apex and Big Pine

145

Key at the base (Figure 3-37).

Their orientation is in a northwest direction

and they lie parallel to each other, increasing in length from west to east.
The Upper and Lower Keys also differ in composition.

The Upper Keys are

composed of Key Largo coral reef limestone while Big Pine Key and those westward are made up of oolite.

However, all of these Lower Keys with surface

deposits of Miami OOlite are believed to be underlain by Key Largo Limestone

(Hoffmeister

ejt

al

1967).

The relationship in stratigraphy of the two

formations is seen at a contact at the southeastern end of Big Pine Key

(

ibid )

Here the oolite overlaps the old coral reef to the south.

cover is a relatively thin layer.

Chica Key it is

The oolite

For example, at the southern end of Boca

m thick, and 2-1/2 km to the north it becomes 10 m thick

(Hoffmeister et al.

1967)

An explanation for the northwest orientation and the shallow channels of

the Lower Keys is offered by Hoffmeister

e_t

al.

1967

The Upper Keys, which

are made up of coral reef limestone, parallel the edge of a reef platform.
The Lower Keys, which are made up of different rocks, have a different orientation.

Oolites probably formed on a platform just north of the coral reefs

which are now overlain by oOlite.

An east-west mound of unstable oolite at

least 10 m thick was formed behind the reef and extending the entire length
of the Lower Keys.

As the layer became higher than the reefs, the oolite

tended to encroach over them and to eventually cover them.

Tidal currents

then cut channels in the oolite normal to the orientation of the mound.

During

the subsequent glacial period, the mound was exposed and the rock became in-

durated.
rents.

When sea level rose again, the oolite was exposed to waves and curThe erosional effects of the waves and currents were concentrated on

channels so as to create the geomorphologic product which exists today.

146

->

?
lii'

<l

s-

147

CHAPTER IV
SEA LEVEL IN THE LATE QUATERNARY PERIOD

Introduction
The rise and fall of sea level during the Quaternary has fascinated

geologists for decades.

This is an area of investigation which is very

poorly understood due to the numerous difficulties and problems that re-

main unsolved and also due to the comparatively small amount of study
given to the problem.

The thousands of wells that have been drilled on

the Outer Continental Shelf have contributed virtually no information on

surficial deposits, and investigations pursued for purely theoretical

science are rare.

Nevertheless, a generalized view of the latest cycle of glacial lowering and subsequent rising of sea level has begun to emerge.

In this study,

we have attempted to synthesize from all the available extant information a

history of sea level change which we can relate to observed relict features

now submerged on the shelf.

In recognition of the importance of glacial con-

trol on sea level changes, we have also attempted to relate the sea level

changes to well-documented episodes of glacial advance or retreat.

Fluctuations of Sea Level

A chronology table and a correlation map treating fluctuations of

sea level are given in Volume III, Plates 1 and

3.

Plate

is a

chronology of Late Quaternary Events and presents an interpretative

history of the relative fluctuations of the land-sea boundary (with

respect to present-day sea level as a datum)

A curve of the relative

change of land-sea level is correlated with the Intervals A-K discussed

later in this chapter, with glacial periods, with physiographic units,

and with corresponding events in faunal evolution, artifact traditions,

and cultural sequences.

Plate

is a map correlating the sea level

1A8

changes with some of the relict features discussed in Chapter

3.

It is

important to stress the relativity of the sea level to land fluctuation,

since it is compounded of changes in sea level and changes in land level
(both with respect to some arbitrary datum, e.g., present sea level as

used here) and must therefore be expressed as a algebraic sum.

As Bloom (1967) has stressed, change in sea level may in fact lead
to opposing changes in land level due to isostatic effects.

Therefore,

the record of fluctuation we see at the present time may indicate a

somewhat amplified curve with respect to a curve of true or absolute

change of sea level.

Besides being amplified, the curve of relative land-sea level must

be in various ways distorted (again, with respect to a true or absolute
of sea level fluctuation)

curve

Such distortion may result partly from a lag in

isostatic response to sea level change.

Other distortions are tectonic sub-

sidence or uplift and subsidence from compaction.

Tectonic effects may be

of regional scale or of local scale (as seen around salt domes).

Changes

in the configuration of the geoid as a result of a sudden shift in

position of the earth's axis or in variations in the rate of rotation have

also been offered as explanations for sea level change (Newman, 1968).

The latest part of the curve is taken from a published version by Scholl
et

_al.

(1969), which in turn is based on numerous dates of various coastal de-

posits of southern Florida.

The earlier, partly overlapping segment (about

4000 - 9000 radiocarbon years ago) is constructed in part from published dates
of Rehkemper (1969) from Galveston Bay and Coleman and Smith (1964) from

the Chenier Plain of Louisiana.

to the curve of Scholl

e_t

al.

The displacement of this curve with respect

(1969) may partially represent a distortion

such as described above, perhaps a regional tectonic distortion, or geoid

changes.

The dip in the curve at about 8,000 years is based on geomorphic

evidence discussed later.

149

The curve from about 9000 years to about 10,000 years is based on two
dates from the northwest Florida area, one from Jervey (1974) and one from

Schnable and Goodell (1968).

Again, the shift in geographic area is asso-

ciated with a discontinuity which may well represent a distortion, possibly

from regional tectonic effects.

Beyond this time, only one possibly useful radiocarbon date is known.
This was reported by Curray (1960).

The sample is an excellent one, but it

was collected from near the Stetson Bank which, according to Edwards (1971),
is underlain by a salt dome structure;

therefore, the possibility of some

local tectonic effect exists.

The remainder of the curve is largely inferred from the forms identified on the shelf, Chapter 3, and from knowledge developed elsewhere which

indicates the history of glacial conditions (Geological History, below).

Paleoclimatology and the Deep-Sea Period

Some general considerations on Late Quaternary paleoclimatology are

important in a study of this kind, but in this area we are faced with the
fact that little attention has been paid to paleoclimate in past works.

Con-

sequently, there is little available evidence from which inferences can be

made.

Perhaps the most authoritative work of concern here is that of Watts

(1975).

Watts' pollen studies of cores from a lake basin in south-central

Florida showed a dry climate characterized by dune vegetation appeared, which,

in turn, was replaced about 4700 years ago by vegetation similar to that

of the present.

Graf (1966) has made some interesting paleoclimatic speculations re-

garding the northwestern Gulf coast area.

150

He considers the soil zones in

the Beaumont formation, which are associated with the Ingleside shoreline

features with their characteristic caliche zones, to be evidence of a former,

relatively more arid time.

He further believes

the characteristic micro-

relief features (the pimple mounds) of the Beaumont and older surfaces in
the area to be aeolian forms developed during a time of aridity post-dating
the Beaumont surface (and the Ingleside shoreline features).

Lundelius (1972), in his faunal analysis of the pit at Ingleside, considers the paleoenvironment to have been warmer and more humid than that of
the present.

His treatment of the age of this unit is noncommital, but he sug-

gests the possibility that it is younger than Late Wisconsin.

This fauna,

in our opinion, may well represent an early Holocene deposit in the order
of 8500 - 12,000 years in age.

The faunal material associated with fresh-

water dune ponds may have originated at this time when warming, possibly
induced by a surface layer of meltwater which covered much of the Gulf of

Mexico in a time of rapid glacial waning, brought to an end a cycle of

relative aridity that existed through much of the period of glaciation.

The dune pond formed under these wetter conditions.
At about this same time, there may have been partial stabilization of
the south Texas sand sheet and deposition of pond marls in the stabilized

dune areas as discussed earlier.

In Louisiana, Otvos (1975b) has provided some paleoclimatic interpre-

tations in his recent review of loess stratigraphy and distribution.

He states

that "the fact of loess accumulation in itself indicates a drier climate."

He

also notes that caliche is well-developed far to the south and east of the
loess localities in the Prairie formation, indicating even more widespread
dryness.

He states, "calcareous concretions do not accumulate in soils under

the present humid-subtropical climate of south Louisiana."

However, it should

be noted that calcareous concretions do develop in marsh and swamp deposits,

151

raising some questions as to an arid interpretation of their origin.*

While

probably Late Pleistocene or Early Holocene, the specific date of these features has not been established.
Part of the aridity of the Gulf coast may have been due to a reduced

frequency of tropical storms because of lower ocean temperature (Moran, 1975;

Adams, 1975a).

Hammond (1976) has presented in preliminary fashion some conclusions of

the Climap project, a major inter-institutional study of sea-floor sediments,
the results of which will shortly be published (Mclntyre et al.

Gater, 1976).
1)

1976; and

Some conclusions are that at about 18,000 years B.P.:

the Gulf Stream may have followed a southerly course toward Spain rather

than England; 2) equatorial oceans were as much as 6C lower in surface

temperature, and the world ocean average was about 2.3C lower;

was at least 85 m lower than present;

4)

sea level

3)

July surface temperatures on land

were as much as 15C lower than present in areas close to the ice sheets and,
on the average, about 5C lower over the continents; and 5) climate was gen-

erally drier than now, especially in the northern hemisphere.

Emiliani et al.

(1975) have recently presented results of an analysis of

the two deep-sea cores marginal to the DeSoto Canyon that are of the greatest

importance in interpretation of the Late Quaternary history of the Gulf of

Mexico.

The cores were investigated in several ways, including foraminiferal

studies, radiocarbon dating, and determination of oxygen isotope ratios.

The foraminiferal studies showed several interesting trends:

water forms

Spaeroidinella dehiscens

1)

Pulleniatina obliquiloculata

warmGloboro -

talia menardii and G. tumida ) were very scarce or absent throughout the cores
but increased in the uppermost core layers; 2) "temperate-warm" species

Otvos (1975b) cites aeolian dune trends associated with the Prairie
formation in southeastern Louisiana as further indications of drier conditions
152

Globigerinoides sacculifera-triloba and Globequadrina eggeri ) were present

in varying abundances throughout the cores; and 3) the "temperate-cold" spe-

cies Globoratalia inflata is rare in the uppermost zones but common in most

of the lower zones of the cores.

Radiocarbon dating of the cores was done on bulk carbonate samples and
for this reason probably overestimated their ages since clastic carbonate

contribution that is reworked from older deposits is inevitable.

et al.

Emiliani

(1975) estimated that dates in their core on the northwest flank of

DeSoto Canyon were about 5000 years older than dates on equivalent horizons
in the core from the southwest flank.

They considered this to be due to

greater amounts of reworked carbonate in the former.

They judged the more

basinward core to give accurate radiocarbon ages even though bulk samples
were used through a presumed correlation with the y/z boundary of Ericson
et al.

(1964).

The difficulty of interpreting the many oxygen isotope and planktonic

foraminiferal studies is that there may be regional variations that lead to

confused correlation.

Few such studies present adequately considered abso-

lute age determinations, and too often bulk samples of sediment have been
the basis of dating.

Samples may also be integrated concentrates of calcium

carbonate skeletal materials accumulated over so broad a time interval that

the resulting age is suspect.

It is quite apparent that a generalized

review of this entire methodology, clarifying both its weak and strong
points, is badly needed, but beyond the scope of the present study.

The foraminiferal studies in particular seem to present many methodological difficulties.

The underlying assumption that assemblage variations

can be interpreted to yield temperature variations through time is only

weakly substantiated by studies of modern assemblages.

The variation of fac-

tors other than temperature and the potential effect of these on the nature

153

of the assemblages deserve more emphasis.

Sea level changes of the magni-

tude described herein could well have impact on the broad-scale structure

and function of marine ecosystems, causing considerable shifts in the rela-

tive balance of populations of many marine species.

The ways that this might

be expressed in the planktonic foraminiferal populations (which collectively

are merely one component of the larger marine ecosystem) are by no means
simple.

Lacking knowledge of such ecosystem response, it is clear that

relating Variations of planktonic populations solely to temperature change

is certainly an oversimplification.

Vergnaud Grazzini (1975) has published one of the more interesting

examples of the oxygen isotope method of study of deep-sea cores.

His cores

from the Mediterranean Sea show particularly pronounced fluctuations, probably

due to the restricted nature of this sea and its location with respect to
the glacier.

Briefly, his interpretation of the cores is that they show

peaks indicative of cold water and a large glacial mass about 17,000 years
ago,

and warm water and a smaller glacial mass at about 55,000 years ago.
By way of explanation of the significance of oxygen isotope variations,

Grazzini

states that they "are caused both by changes in the ocean

surface temperature and by glacially-controlled changes in the isotope com-

position of the ocean water." He further points out that "most workers believe now that the last is the dominant factor and have even suggested that
the existence of any residual temperature effect remains to be demonstrated."

Grazzini postulates that the magnitude of isotope fluctuations

in.

the Mediterranean is exaggerated (with respect to the ocean gener-

ally) to a degree by "a 'dry effect' resulting from lower precipitation or

higher evaporation, or both" which may have increased isotope and salinity

concentrations in the water

He observes that in cold winter weather this

occurs presently in the Mediterranean, and "one may observe a sinking to the

bottom of the denser surface waters that have undergone extreme evaporation."

154

This is interesting because in the Gulf of Mexico, sometimes called

the "American Mediterranean," Emiliani

e_t

aj^.

(1975) have noted that isotope

variations are greater than in Caribbean cores and that this "may represent
an excess of evaporation in the Gulf of Mexico (which was closer to the

Laurentide Ice Sheet)."

Geological History
We have chosen the inner barrier features, which are comparable to and

perhaps correlative with the Ingleside shoreline forms, as the oldest forms
of concern in this study.

It is believed that their antiquity is perhaps of

a similar magnitude to that of known evidences of the presence of people in

North America.
Our correlations of the forms identified on the shelf with time inter-

vals in which they may have been created is presented in Volume III, Plate

3.

The correlations are admittedly provisional due to the rarity of any abso-

lute basis for the chronology.

For the most part, the correlations are based

on the sea level curve, but the forms have obviously influenced the construc-

tion of the curve and the two are not independent.

In a few cases, absolute dating adds to the information used.

Jervey

(1974) has provided a date which is possibly related to a barrier form which

trends into the nearshore zone just east of Destin, Florida.

age is 9070

320 years B.P.

The indicated

The date was on Chione cancellata shells from

a facies interpreted by Jervey as estuarine.

Since this species has a char-

acteristic shell sculpture of cross-hatched ridges (forming a cancellate

pattern from which its name is derived), it is possible to recognize evidence
of

abrasion readily.

Jervey states that the shells used for dating were

"well preserved," indicating that abrasion was slight and that the shells

were not transported from elsewhere.

Without such valuable observations, this

155

dated sample would be of little worth, but since Jervey is particularly concerned with transport, dispersal, and abrasion of such biogenic materials in
his work, we regard this sample to be of primary importance.

Another potentially useful but less credible date is that of Curray

(1960)

which comes from between a submerged ridge form and the Sebree Bank

off of the Rio Grande delta.

on Plate 3.

This is the ridge assigned to the H^ interval

In this case, the sample was dredged oyster shell.

As we have

discussed previously, there is much possibility of contamination in such samples from inclusion of transported older shell.

+ 270 years B.P.

The date obtained was 9530

The sample was dredged from 30 m of water, and while this

sample was not used in constructing the sea level curve, it does not plot
off the curve too badly,

especially if some allowance is made for depth of

water in which the oysters initially lived.

The only other radiocarbon date associated with a submerged shoreline
is that of Winchester

(1971) on a caliche nodule which was believed to have

formed in a soil zone developed in the Beaumont formation through subaerial

weathering at a time of lower sea level.

The date, then, indicates age of

the nodule, not of the beach, and implies that the beach is younger than the

nodule.

The age obtained from analysis of the nodule was 15,857 + 269 years

B.P.

Interval A - There is by no means any general agreement about the early

history of the Wisconsin glaciation and even less concerning the interglacial preceding it.

Barry

ejt

al.

(1975)

have recently reviewed the

thought expressed by some researchers that the initiation of the last glaci-

ation was possibly quite rapid.

The various authors cited have reported

sudden cooling trends at 115,000 - 120,000, 90,000 and 70,000 years B.P.

Attempts by Barry

ejt

al_.

(1975)

to model physical processes that might

produce such rapid changes have so far been unsuccessful, although a

156

degree of cooling of the right order can be achieved after 10,000 years
if some assumptions

(which the authors admit are dubious assumptions)

are made concerning circulation regimes and energy requirements.

As the level of the sea declined from its maximum high stand during the

Sangamon interglacial, there may have been one or more prolonged stillstands
or secondary maxima.

This is an area that remains very unclear due to prob-

lems in absolute age determination, particularly radiocarbon dating, that

have produced much misinformation in the literature.

Nevertheless, there

exists some probability that proposed interstadials, such as the St. Pierre

interstadial (ca. 60,000 years ago), may yet prove to have validity as sig-

nificant post-Sangamon interstadials which may have been related to sea levels
equal to or higher than the present.
to

Given this possibility, its relevance

the Ingleside shoreline features is obvious.

The Sangamon features of the Gulf coast area are best understood on
the peninsula of Florida.

ample

Perhaps the most interesting and well-known ex-

of these features is seen in the landscape of south Florida.

paleogeographic and stratigraphic interpretations of Hoffmeister

The

et^ al.

(1967) and Brooks (1968) indicate a shallow sea bank over much of present

south Florida, with oolite-characterized shoals, an extensive bryozoal

limestone-forming environment, and algal and coralline reefs. Th/U

dates

indicate such a landscape existed approximately 125,000 years ago.

Strong

supporting evidence of similar shoreline features at this time exists in

the Bahamas (Neumann and Moore, 1975) and in Barbados, where evidence also

exists of even higher levels, earlier in the Sangamon.

The inner beach ridges of the Cape Kennedy littoral accumulative form
are apparently of comparable age to the south Florida features just de-

scribed (Osmond

et al

1970).

157

These features cause us to wonder if there may not be Sangamon-aged

features around the entire Gulf rim.

Indeed, many have correlated inner

relict shoreline forms elsewhere on the Gulf with the forms in Florida
and

others have suggested Sangamon ages for such forms for other reasons

(e.g.,Otvos, 1975a).
If Sangamon forms do exist in Florida and if we assume their absence

elsewhere, then certain problems of interpretation arise.

One is that their

absence elsewhere most probably would indicate a tectonic differential, par-

ticularly a downward subsidence of the northern Gulf rim since Sangamon times,

with respect to peninsular Florida.

The other is that if the inner beach

ridge forms of the northern Gulf area (such as the Ingleside features) are
truly younger than Sangamon, then why did the later high stand of sea level

which produced them leave no important record of forms or deposits in the area
of the old Sangamon terrace of south Florida?

Some workers believe that the

Silver Bluff shoreline forms of that area possibly represent a weakly

developed, later, brief period of shoreline occupancy slightly above the

present level and are true correlates of the features of the northern Gulf.
However, other workers believe that the Silver Bluff shoreline is consider-

ably older.

Brooks (1973), for example, has stated that:

There is topographic evidence of a 6 to 8 foot

stand of sea level in most coastal areas of peninsular Florida and the Gulf Coast, but for the most
part, fossilif erous beach and lagoonal deposits are
lacking.
This stand of sea level is not Holocene
as many have thought (MacNeil, 1949).
In the area
of Marineland south of St. Augustine, Florida, excellent lagoonal deposits are associated with a coquina barrier. This highly fossilif erous deposit
is too old to date by the carbon-14 method.
It is
herein suggested that the deposit is about 90,000
years old and correlates with a late Sangamon event
(p.

II E-7).

To say that Sangamon features which seem to exist in south Florida

are absent along the northern Gulf rim through a net downward subsidence of

158

that area with respect to Florida is in conflict with much published opinion

concerning the tectonic character of the northern Gulf.

Fisk (1939), Bernard

(1950), Graf (1966), and many others have held that the western part of the

northern rim of the Gulf may actually be subject to tectonic uplift, which
is generally believed to be the response of more landward locales to subsi-

dence

occurring on the shelf.

Since a variety of forms suggesting older Sangamon and Pre-Sangamon

shorelines exists in Florida (Alt and Brooks, 1965) and along the Atlantic
coastal plain, their absence in the northwestern Gulf has been far too gen-

erally ignored.

Upwarping inland from a hinge line should have elevated

these shorelines, but if this has occurred, they have been totally eroded
away, which seems improbable.

An alternative is that the shorelines are

covered by younger deposits as a result of subsidence or downfaulting.

Consider for instance a hypothetical case.

Suppose a sea level stand

of 20 m above present sea level occurred about 125,000 years ago.

Relict

shoreline forms of this stand could have subsided to the modern sea level
by 62,500 years ago if they were subject merely to a rate of subsidence of
0.32 m/century, which is less than many other estimates of subsidence rates

along the Gulf coast.

If we then consider a hypothesis that at about that time

(62,500 years

ago) the Ingleside shoreline came into existence at a time when sea level

again stood at about 15 m above its present level, then it is probable that
the older (125,000 years) shoreline was already either submerged, buried,
or possibly washed away by marine erosion.

These are, of course, only hypotheses, and there is no particular evi-

dence for them.

However, they are important in order to show that there

could have been such a sequence of events.

The careful reader will note,

however, another dilemma in this hypothetical argument.

159

If the same rate

of subsidence we postulate for the first 62,500 years was continued through

the second 62,500 years, then the Ingleside shoreline should also have sub-

sided to below present sea level.

This is, of course, not the case, although

as Graf (1966) shows, this shoreline is virtually at present sea level in the

segments near the Brazos-Colorado delta and perhaps is below present sea
level in the Freeport Rocks area, as we have described in the section on
forms.

This dilemma, however

may result from the simplistic assumption that

subsidence proceeds at some continuous, linear rate, when actually it is

more likely to be episodic and quite variable in time and space.

Thus, the

high-standing Houston ridge segment of western Louisiana has experienced

little subsidence, and the segments near the Brazos-Colorado delta have subsided approximately at the rates assumed above.
but it does serve to point out an

Such speculation may seem idle

area of truly basic importance where there is great ignorance in consideration

of Gulf coast geomorphic problems.

This is the subject of active deforma-

tional processes (or neotectonics, as some term it).

It is clear that much

can be gained from more intensive study in this field.

Interval

B_

- The remainder of the falling stage

accompanying the early

Wisconsin glacial increase is assigned to Interval

of rapidly falling sea level.

B.

This was an interval

Coastal streams and larger rivers continued

to entrench during this interval.

Older fluvial and marine deposits were

exposed to weathering and erosion over greater and greater areas as the shoreline receded.

This regressive unit was composed of reworked earlier materials

and materials in transport over the shelf and along shore.

Much of these

may have been quickly removed as they became stranded.

The exposed terrain may have responded rather quickly to isostatic

compensation leading to relative uplift.

and may not have occurred instantaneously.

160

This adjustment was probably slight

Other things also happened to

affect the relative level of the exposed terrain with respect to the sea.
The exposure of fluvial and marine beds during an interval of low sea level

must have had some rather important effects, especially if that exposure persisted over a prolonged interval of time.
1)

Among some of the effects are:

volumetric change effects resulting from dewatering, consolidation, oxi-

dation of organic matter, leaching of soluble constituents, etc., all of

which are changes which may lead to varying degrees of local deformation
(mainly subsidence); 2) erosive effects which were highly variable - where
these were most intense, there may have been further isostatic compensation;
and 3) other effects such as mass movement, colluviation, cementation and

calichif ication
As sea level was falling, streams extended directly to the shoreline
and contributed the bulk of their sediment loads directly to the shelf.

Inland from the shore, entrenchment led to remobilization of sediment de-

posited as valley fills in previous periods of valley aggradation.

For

these reasons, sediment reaching the shelf possibly increased in amount

during at least part of such a falling stage, even if streams were possibly

diminished in volume due to net addition to glacial snow and ice storage.

Much of this remobilized sediment was of coarser-grade size which was redeposited, for the most part, in the nearshore zone after a period of en-

trapment by coastal current and drift systems.

After deposition in the

nearshore zone, continued fall of sea level led to re-exposure of the beds.
In this way, a thin, probably discontinuous, regressive unit of nearshore

deposits came into existence,

wherever this persisted, its importance,

as we will see, must have been great at a later time when the sea again rose

across this terrain, and these materials were again incorporated into the

nearshore zone and into transgressive deposits.

161

Interval
to 90

C^

A prolonged period of relative constancy of sea level at 60

m below its present level is believed to have existed throughout the

interval of 28 - 46 thousand years before the present.

During this period,

rivers and streams continued in their profile adjustments to the great base
level change of the previous falling sea level interval.

Sediment delivery to the coast was probably still rapid, at least in

the earlier part of this period, and coastal progradation may have continued
as sea level stabilized, especially in the deltaic areas.

Some isostatic

compensation may have continued into this time, leading to slight coastal
rise.

Away from deltas, slower coastal progradation or even long-term net

erosion by nearshore processes may have prevailed.

Both would lead, over a

long period, to a coastal terrace at or about the sea level of the time.

A prolonged interval of relatively stable sea level may mean that no

net glacial snow and ice buildup was occurring and that world climate patterns

apparently existed in a more or less stable state of this kind for some time.
We are aware of no particular explanation for such an event.

In the view of

Adam (1975b), an ice sheet once initiated will grow until the heat stored
in the surface waters of the ocean is used up and the energy gradient to

transfer water from the oceans to glaciers thereby diminishes.

Perhaps a

steady-state situation existed for some time thereafter in which each yearly
increment of solar energy led to sufficient transfer of moisture to the ice
sheets to more or less compensate for melting, but not to produce net growth.
In Adam's

(1975b) view, after the heat stored above the thermocline in

the world ocean has declined in amount

correspondingly)
heat occurs.

(and surface zone temperatures decline

less energy transfer as latent heat and more as sensible

This would perhaps signify somewhat drier climates in many con-

tinental areas since further warming of the air as it flowed from the sea
over the land would cause more water uptake.

162

Adam thought that a glacial episode would involve rapid growth of

glaciers while "heat of glaciation" derived from ocean storage was great,
then a sudden reversal and decline as this store declined and triggered a de-

cline in latent heat transfer.

The suddenness of the decline he felt was

also a result of a tendency of the glaciers to over-develop or over-extend.

Because of the over-extension, rapid melting ensued.

In an extremely

in-

teresting argument, Adam shows that rapid melting creates a surface water
layer which leads to high heat loss with no potential for storage to drive
This also contributes to glacial disintegra-

winter latent heat transfers.

tion by further cutting off nourishment.

However, it could be argued that other factors may operate that could

perpetuate a steady-state glacial condition for a long period.

While less

water is available from the cooler ocean, perhaps some compensation occurs
from evaporation over land areas.

An increased aridity over most ungla-

ciated land areas might then be seen, leading to still further reduction of
albedo over the unglaciated areas, which might also act to maintain the
glaciers.

Emiliani

et_ al.

(1975) have recently elaborated a hypothesis of con-

ditions during Stage 3 of the Emiliani interpretation of deep-sea core

stratigraphy.

They state that all micropaleontological evidence presently

available "... indicates that Stage

was 'cool' with temperatures closer

to those of a glacial age than those of an interglacial one"

1975, p. 1087).

ibid )

et al.

They also state, without offering evidence or cited work,

that "sea level during Stage

(

(Emiliani

apparently stood not much below the present"

This comment perhaps is influenced by published curves such as

Milliman and Emery (1968) presented, but these curves have few data points
in the interval of Stage

(and even these few are highly suspect according

to Thorn, 1971).

163

With their unsupported statement that sea level during Stage

much below the present; combined with their evidence that the Gulf
paratively cool at the surface, Emiliani
esting interpretation.

e_t

al_.

3
w?as

was not
com-

(1975) go on to an inter-

They suggest these tendencies "can be explained

concurrently if a large but thin ice cap had persisted over northern North
America during Stage 3, with an unusually rapid rate of accumulation and
ablation."

They further state that "... continued rapid ablation under

equilibrium conditions would supply the Gulf of Mexico with a continued abundant influx of ice meltwater while low temperatures would be maintained."
As Adam (1975b) has emphasized, the lower temperature of the Gulf and

an abundant influx of ice meltwater are somewhat mutually exclusive situations, or at least would not likely be stable over the long term.

that fall of temperature of the sea surface leads to

Adam says

reduced glacial

nourishment, then to surface layer formation by increased runoff, which

further reduces glacial nourishment.

Lindsay

et al

(1975) have reported a large (15 km-wide) form which

they regard as a submerged barrier reef at 100 - 110 m depth on the Campeche
Shelf.

This form is illustrated well by its profile which is reproduced

here as Figure 4-1.

This form possibly began to develop in Interval C and

continued its growth in Interval D.

In the southern part of the west Florida shelf, there is a narrow ter-

race at about 80 - 90 m which broadens at both its northern and southern

ends, where there is some suggestion of barrier forms.

There is little or

no published information on the nature of the bottom in this area, and the

bathymetry is also poorly known.

The terrace is broadest in an area about

169 km due west of Charlotte Harbor and in an area about 100 km west by

northwest from the Dry Tortugas.

164

a1

.tii-i a i .i

.1

.-Z.Um-i

7:.;..
j

.?
u.

.u vl.j

i..

>

i,

..

r
>

a.,

"

Q.
UJ

i.v

Jiucn.:><^'-

40;

4iiL

FOR^ REEF

60-

1*

. L.a-

.:st;

If.

:ji,.JL

!=t?f

'

_BARRiER_REgF

80MILES
J

L.

KM
Figure 4-1.

100 - 110 m terrace and drowned barrier reef on Campeche

1975).
Shelf (From Lindsay et al
.

We tentatively identify this feature as evidence of a prolonged low

stand of sea level and suggest that it correlates with the similar terrace
of the Campeche Shelf

Logan
stillstand

1
'

et al.

(1969) also suggested that there was a "prolonged

at about 91 m.

They felt this occurred prior to 17,710 + 50

years B.P. from a dated sample of "pelagic shells" from a formation related
to this level, but there is little reason to accept this date except pos-

sibly as giving a minimal age.

165

Interval

J)

- The time of the

Farmdalian interstade has often been sug-

gested as an interval in which a significant rise of sea level occurred.

Many workers feel that the level reached at that time may have nearly
equalled or exceeded the present level.
Thorn (1971) has

extensively reviewed the radiometric evidence of the

numerous Mid-Wisconsin high sea level stands that have been proposed and
finds that none of the dates are satisfactory.

Since this review, the situ-

ation has changed very little with the possible exception of a date reported

by Neumann and Moore (1975) from a cavern in the Bahamas.

This date meets

some of Thorn's criteria for satisfactory dates, but since it is not a marine
sample, the other criteria he proposes are not applicable.

The date was on a sample of a stalagmite, and both the radiocarbon and

Th/U methods were applied.

The radiocarbon age was 21,900 + 900 years B.P.

and the Th/U age was 22,000

350 years B.P.

The close agreement tends to

support a viewpoint that there has been no recrystallization of the sample.

This date indicates that at about 22,000 years ago, sea level was at
least 11 m below its present level.

There is no way of knowing from this

limited information how much lower it may have been, but this work suggests
a

methodology which may eventually substantially add to our knowledge of past

sea levels.

We suggest herein that the Farmdalian Interval did not lead to substantial
rise of sea level and is represented in the Gulf of Mexico by a prominent

barrier-strandplain form at a depth of 72 - 80 m.

This form is evident only

in the area of slow deposition off of southern Florida.

Intervals E and

F_

- These two intervals of time represent the resurgence and

maximum growth of the Late Wisconsin glacier.

at a relatively rapid rate.

On the shelf, sea level fell

The forms recording this are the dendritic chan-

nel forms seen in the areas of more detailed bathymetric coverage in the

166

south and central Texas sectors and on the shelf off of west Florida, some
slope breaks in the south Texas and east Texas sectors, and the Mississippi
trough.

The banks of the east Texas and west Louisiana sectors also show

platforms, notches, and submerged reefs which appeared at about the time of

maximum low sea stand.

Interval G - The rapid rise of sea level in the latter part of Interval F, re-

flecting rapid glacial decline, continued in the early part of Interval G.

Sea level rose rapidly by about 70 m from its maximum low stand, which led to

generation of shoreline forms at approximately the same level as forms which

had been created earlier in Interval D.

Algal and coralline banks which had

been exposed to weathering during the lower sea level became newly active.
Many previous forms were obliterated by coastal erosion and new deposition.
The size of some of the G-Interval forms may reflect their having been laid

over a base of prior forms and the contribution of materials reworked from
these prior forms.

Interval H - The onset of Interval H was marked by what perhaps was the most
rapid rise of sea level and the most rapid period of deglaciation of the entire

period under investigation.

This occurred roughly 12,000 years ago and

makes a convenient boundary for dividing the Pleistocene from the Holocene,
as has been done in this report.

Emiliani

et al.

18
(1975) find the lowest

concentration in their core

near the DeSoto Canyon to occur in an interval bracketed by dates of about

12,000 and 11,000 years.

maximum return of 16

This, then, indicates in their view the time of

concentrated water from the glaciers.

They estimate

from the bracketing dates that this occurred about 11,600 years ago and point
out that this closely coincides with Plato's estimate of the time of the

legendary Atlantean deluge.

The coincidence is indeed striking, but perhaps

167

more than it should be since the dates are probably in error (through presence of detrital carbonate as we have discussed elsewhere) and also since
it is not clear whether these ages are cited in radiocarbon years or actual

historical years.

The effect of detrital carbonate would most likely be

that the apparent ages would be older than the true age.

The effect of not

correcting radiocarbon years to actual historic years is that the ages are
too old, thus it is possible that the estimate of Emiliani

et_

al

(1975)

closely coincides with Plato's flood stories only by chance.

We make no reference here to the problematic Valders glacial episode
since Mickelson and Evenson (1975) have questioned its existence and main-

tained that the tills from which this interval was named are actually older
than the Two Creeks forest bed.

They point out that the till above the for-

est bed at Two Creeks may be evidence of a glacial episode after the warm

interval, and there is some possibility that some decline of sea level ac-

companied this renewed glaciation.

Geomorphic and archeological data from the coastal zone suggest

relative stillstand during subintervals H2 and H3, terminating about 8500

Extensive swamp development in the west Louisiana area suggests

years ago.

that conditions may have been somewhat wetter than those which prevail at

present.

The extinction of the Pleistocene megafauna had not yet occurred,

and mastodon, mammoth, horse, and other forms were abundant.

There are several lines of evidence which suggest that subinterval H4

may have been a time of catastrophic events.

Geomorphic evidence suggests

a reversal of the trend of gradual rise or even relative stillstand of sea

level that had occurred for several thousand years prior to this period.

Gagliano and

Thorn

(1967) tentatively place the Deweyville alluvial terrace

in this time interval.

It was also during this period that the extinction

of the Pleistocene megafauna probably occurred.

168

It is also noteworthy that

during this period of climatic and environmental instability, Archaic culture

emerged in the northern Gulf area.

Intervals

I_,

.J,

and K - These intervals represent a period of about 7000

years of more or less continuous rise of sea level from about -15 m at the

beginning of the period to the present level at the present time.

This is

the best documented interval of time with respect to useful radiocarbon

dates.

Nevertheless, there is still a wide range of controversy concerning

the nature of this last phase of history.

Particularly, there is much argu-

ment concerning whether the sea formerly may have been higher than at present
in the last few thousand years, as Fairbridge (1961, 1974) has repeatedly

maintained.

Fortunately, this argument has little importance in the present study

since it appears amply evident that regardless of when the sea reached its

present level or whether it ever stood higher than present, there still re-

main many possibilities of cultural resources over most regions of the shelf
from earlier times of lower sea level.

169

CHAPTER V

ARCHEOLOGICAL METHOD AND THEORY

Culture Areas
The archeological literature related to the coastal zone of the

northern Gulf is considerable.

Important sources of unpublished data

exist, in addition, primarily within the anthropology departments and

museums of universities and in the files of amateur archeologists in the

coastal states bordering the Gulf.

Although there are numerous important

summaries of the archeology both at the continental and regional levels, no

single work satisfies the requirements of the present study.
As a point of departure, the broad cultural areas and subareas defined
by Willey (1966) can be used (Figure 5-1).

For the purposes of the present

study, the northern Gulf will be divided into western, central, and eastern

areas.

As shown in Figure 5-1, the Western Gulf area corresponds to the

Northeast Mexico - Texas culture area of Willey.

The Central Gulf is

equivalent to Willey 's Lower Mississippi Valley Subarea, and the Eastern
Gulf area embraces both the Southeast Subarea and the Glades Subarea.

Although

beyond the scope of the present report, it should

it is

be noted that the subareas have also been segmented.

Willey (1949) de-

fined three segments of the Florida Gulf Coast within the Southeastern Sub-

The work

area as the Manatee Region, Central Coast, and Northwest Coast.

of Trickey (1958) would suggest that Mobile Bay deserves special treatment.
In Louisiana, Gagliano
er cultural units.

(1967) and Phillips (1970) have both defined small-

The Texas coast has likewise been segmented

Krieger, and Jelks, 1954).

Suhm,

Summary treatments exist for a number of these

archeological units as well as for a few of the subareas.

Phillips'

For example,

(1970) work on the Lower Mississippi Valley contains a compre-

hensive summary of coastal Louisiana archeology.

170

0)

u
CD
4->

M-l

<
c
o
H
60
0)

O
n

CD
4->

O
fi

0)

-C
4-1

M-l

O
W
cd
CO

K)

cd

w
a
CD
cd
0)
j-i

4-1

CJ>

m
cu
S-i

toO

171

Although the archeological literature of the coastal zone is generally voluminous, those intervals beyond the Late Archaic (older than
about 2,000 B.C.) are poorly known (see Volume III, Plate 1),

The explana-

tion probably lies in the fact that much of the record is contained in

drowned sites.

When the region is viewed through time, a waxing and waning of distinctive cultural manifestations at various places becomes apparent.

Al-

though the complete picture is a complex mosaic of moving peoples and culture traits, certain patterns and clusters of distinctive traits provide the

basis for systematic study.

From the standpoint of this study, important

traits to be considered include patterns of settlement and environmental

use, site morphology and content, and index artifacts and characteristic

assemblages.
Thus for each cultural manifestation we should strive to develop set-

tlement pattern - environmental use models.

The models should indicate the

relationship between site location and environmental setting (with emphasis

on geomorphic features)

Such models increase predictability of high

probability areas for site occurrence in reference to landforms.

The

models become useful when relict features are identified on the continental
shelf.

They provide an important basis for the prediction of high proba-

bility areas for occurrence of drowned sites.

The size and shape of the site (geometry), probable content (shells,

animal bones, black earth, burned rock, etc.), and the kinds and abundance of
artifacts that are likely to be found, serve to identify archeological sites
and to distinguish them from natural deposits or features.

The search for

drowned terrestrial sites involves testing for these distinguishing features of

archeological sites.

These distinguishing features can be thought of as

"cultural signatures" (in analogy to electronic signatures) in the

172

identification of underwater sites.

These signatures might be detected

by remote sensing and sampling techniques from the water surface or by

direct underwater examination and testing.

Thus, two major objectives of the study that are to be accomplished

through the review of coastal area archeology are:

1)

the development of

settlement pattern - environmental-use models, and 2) the development of

lists of distinguishing characteristics or "cultural signatures" which serve
to distinguish archeological sites from natural deposits and features, and
to distinguish them from each other.

Early Man
One of the most interesting aspects of this study is the

evaluation of the nature and extent of the earliest occupations of the

region.

This indeed touches on one of the least known and possibly the

most controversial topic of New World archeology.

Because so-called

"Early Man" sites are widely scattered, it is important to place the

northern Gulf region in proper perspective.

There is abundant evidence, largely in the form of fluted projectile
points, that the entire region had been occupied, or at least visited, by

11,000 to 12,000 years before present (B.P.).

Large numbers of fluted

projectile points have been reported from the eastern as well as the

western Gulf area (see Williams and Stoltman, 1965; Neil, 1964; Gagliano
and Gregory, 1965).

There is a growing body of evidence that man has been a resident in the
area for considerably longer.

In Texas, the Lewisville, Malakoff, and

Friesenhahn Cave sites have all been assigned to a "pre-projectile point"

stage by Krieger (1964).

The Avery Island site in coastal Louisiana, the

"Natchez Pelvis Find" in Mississippi, the so-called "pebble tool industries"

173

of Alabama, and certain artifacts from sinkholes in Florida also have "pre-

projectile point" implications.

Radiocarbon dates and geological associa-

tions at some of these locales suggest a minimum of 20,000 to 25,000 years

B.P.

for man's entry into the Gulf coast region.

While some researchers have presented evidence to demonstrate that

man has been in the New World for 50,000 to 100,000 years or more, other
scholars are highly critical of the validity of the evidence.

Although

this controversy cannot be resolved by the proposed study, the possibility

of these early occupations should at least be considered, and the kinds
of sites and artifacts that might be associated with them should be iden-

tified.

Richard

S.

MacNeish in two recent articles (1972, 1976) has ef-

fectively summarized the evidence for early man in the New World.

Citing

evidence from South, Middle, and North America, he defines four prehistoric
stages and traditions.

As shown in Figure 5-2, the three earliest tradi-

tions are attributed to Old World sources.

tradition, marked by the appearance of

The fourth and most recent

well-made projectile points for

big-game hunting, seems to be indigenous to the New World.

The following

is a summary of the four stages described by llacNeish.

Stage

I:

Core Tool Tradition

This is the most tentative stage dating

to more than 20,000 years ago in South America and possibly to more than

40,000 years ago in North America.

The difference in ages is attributed to

slow southerly dispersal of the tradition.

It is proposed that migrating bands

crossed the Bering Strait land bridge some 70,000 + 30,000 years ago and mi-

grated slowly southward.

gatherers.

These early people were unspecialized hunters and

They utilized a wide range of plants and animals including big game

when the opportunity presented itself.

The artifact assemblage includes crude,

large, bifacial and slab choppers, cleavers, hammers, scraping planes, and

crude concave- and convex-sided, unifacial scrapers or spokeshave-like objects.

174

25,000

30,000

50,000
60,000

75,000

Figure 5-2,

Chronology of
between major
for the three
is thought to

early man sites and traditions. Note

continental areas and the implied Old
earliest traditions.
The specialized
be indigenous to the New World (After

1976).
175

the time lag

VJorld origin

point tradition
MacNeish, 1972;

Pointed flakes may have served as simple projectile points.

Burned bone in

hearths and simple pebble tools may also be included.

Flake/Bone Tool Traditi on.

Stage II:

Dated sites in South America

range from 12,000 to as much as 16,000 years old.

In Central America, the

dates are from 15,000 to 25,000 years, and in North America from 25,000

People of this stage could either have entered the New World

to 40,000 years.

via the land bridge from Asia some 40,000 + 10,000 years ago or developed in
northern North America.

New concepts appear in the technology which em-

phasize stone flake tools, produced by percussion and pressure, as well as

the manufacure and use of bone tools.

Although meager, the faunal and

artifact evidence indicates a relatively unspecialized hunting subsistence.

The artifact assemblage includes flake projectile points or knives,

bone projectile points, pebble choppers, plano-convex denticulate scrapers,

burins, and rib-bone fleshers.

Stage III

Blade

Burin

and Leaf-Point Tradition

This stage is

represented by complexes in the 11,000 to 15,000 year range in South America

and Middle America.

In North America, they do not appear to be greatly

older, falling into the 13,000 to 25,000 year range.

Although data are

sketchy, these people are believed to have been specialized hunters of

There is a

big game or herd animals in a wide variety of environments.

marked advance in tool technology over the previous stage.

Fine, leaf-

shaped, bifacial projectile points, blades, and skillfully made flint

burins appear in the assemblage.

The burins suggest a comparable advance

in the bone tool technology.

Stage IV

Specialized Point Tradition

This stage covers the period

from 13,000 to 8,500 years ago throughout the New World.

It is believed

to be characterized by highly specialized hunting techniques.

These people

not only had a number of different hunting techniques, but the wide variety of
176

chipped artifacts and foodstuffs found at the sites suggests that they had

begun to collect a number of kinds of plants and to hunt and trap smaller
animals.

In some areas,

there is evidence of systematic collection of nuts,

seeds, fruits, and other plant materials.

Evidence for occupations during

this interval is widespread throughout the New World.

Many different kinds

of environmental settings were utilized, with a corresponding diversity of

subsistencepatterns.

It should be noted that present concepts of Stages

IV as well as III come from inland or interior site exploration.

Coastal

sites from these intervals in the Northern Gulf area remain virtually unexplored.

Characteristic artifact assemblages show much greater diversity than

in earlier stages.

Projectile point types include, but are not limited

to, Lind Coulee, Lerma, Hardaway, Plainview, Folsom, Clovis, Fell's Cave,

Lauricocha, and stemmed points.

Unifacial end scrapers, denticulates,

spokeshaves, knives, burins, and many other specialized tools occur.

Basic tools include pins, projectile points, and shaft straighteners.

Landform Associations
The principles of stratigraphy are well known to the archeologist

less-formalized and less-used set of concepts is the relationship between

site distribution and geomorphic features.

Coastal zone sites in favorable locales that are geologically stable

or where there is appreciable relief may be continuously occupied,

odically re-occupied, for long periods of time.

or peri-

For example, certain cave

sites along the South African coast were occupied for many thousands of years.
Sites associated with ephemeral features of low-gradient coasts are more

often characterized by occupations of shorter duration.

obtained throughout much of the Northern Gulf area.

Such conditions

Further, the periods

of occupation are likely to be associated with favorable ecological condi-

tions in the vicinity of the site.

177

In dynamic coastal situations, where changes in the positions of shore-

lines and streams are accompanied by shifts in settlement pattern and land
use, the occupation pattern is dictated by active and relict landforms,

such as stream banks, beaches, margins of estuaries, and important ecotones.

Studies of the distribution of sites by age or culture period in

coastal areas have also demonstrated the relationships between occupation

patterns and the evolution of the coastal landscape.

Simple but effective

demonstrations of this are found in places where progradation has occurred

over a considerable period of time and a series of relict shorelines has

developed.

The basic relationships of this frequently occurring situation are

shown diagrammatically in Figure 5-3.

Note in the figure that the mouth

of the stream has been a favored habitation place through time, and sites

are found not only at the present mouth of the stream, but also at former

stream mouth locations.

In this kind of situation, some of the site oc-

cupations may have continued even after the position of the stream mouth
shifted as a result of shoreline progradation.

However, in such instances,

it is the initial occupation that most closely approximates the age of the

feature upon which the site is located.

For this reason, sequential maps

showing initial occupations of sites are very useful in the interpretation of the geological history.

Another example of this principle

can be found in lobate delta areas, where areas of delta building have

shifted periodically as a result of upstream diversion.

Figure 5-4

illustrates diagrammatically the relationships which may exist in such a

case.

Sites are located primarily along the natural levee ridges of trunk

channels and distributaries.

While the period of initial occupation of the

sites associated with an abandoned river course or delta lobe may approximate
the time when it was active, habitation may have continued after the system

178

RELICT

SHORELINES

PRESENT SHORELINE

PRESENT SHORELINE

X'

D
WEATHERED SURFACE

Figure 5-3.

Distribution of initial occupation sites in a prograding

beach sequence.
A - oldest; E - youngest.

179

ACTIVE CHANNEL-NATURAL LEVEES

ABANDONED
RIVER CHANNELS

ABANDONED DELTA LOBES

ACTIVE DELTA

Figure 5-4.

GULF SHORE

Distribution of initial occupation sites in a lobate delta.

A - oldest; C - youngest.

ceased to function as an active delta lobe.

In some instances the habita-

tion sequence may be out of phase with the delta sequence.

An explanation

for these relationships is found in the cyclic nature of delta building

and the environmental succession which unfolds as the cycle progresses.

Upstream diversions of the river result in an influx of fresh water

and sediment along the coast in the vicinity of active distributaries.

180

Sediment deposition results in delta building and shoreline progradation.

Natural upstream closure of a major distributary cuts off the freshwater

and sediment influx, marine processes gradually become dominant, and the

delta lobe may retreat or deteriorate.

through a cycle.

Thus, the delta lobe evolves

The stages of an idealized delta cycle and associated

environmental changes are shown in Figure 5-5.

As indicated in the figure,

initial human occupation is often associated with the early stages of subHowever, it should be noted that biological productivity

aerial development.

is highest during the early stages of deterioration.

It is during these

stages that utilization of delta lobes by hunting-gathering peoples is

usually the greatest.

Another kind of situation is found in places where smaller, coastal
plain streams enter the Gulf.

with such streams.

Many favorable habitation sites are associated

As shown in Figure 5-6 (occupation "C" sites), sites can

be anticipated along the banks of the stream itself, along the margins of
the floodplain or valley wall,

in the deltaic area of the stream system,

and along the margin of the estuary near the mouth of the stream.

The morphology of such systems may be complicated by fluctuations in

the base level of the stream, resulting in alluvial and coastal terraces.

Relict stream courses are associated with the alluvial terraces, while relict

beaches and other shoreline features are found on the coastal terrace.

Figure 5-6 shows a characteristic pattern of initial occupation that might

occur in a typical system of this type.

Figure 5-7 illustrates still another pattern of initial occupation

sites that might be found in association with sequentially developed, re-

curved spit complexes and beach accretion on a major barrier- spit complex.
Note that the spit builds in a downdrift direction with a minor seaward

growth component.

The distribution of initial occupation sites reflects

181

HUMAN HABITATION AND BIOLOGICAL PRODUCTIVITY

AS A FUNCTION OF THE DELTA CYCLE
HIGH

SUBAQUEOUS GROWTH

natural environments

RAPID SUBAERIAl

GROWTH

|TIME SPA

DETERIORATION

t>

open bay
subaqueous levees
mudflats
fresh

marsh

brackish marsh
saline

marsh

swamp
lakes

oyster reefs

marginal beaches
barrier islands

human

habitation

initial

occupation

continued occupation

Figure 5-5.

Environmental succession of an idealized delta cycle.

The time
required for completion of a cycle varies from decades to thousands of years, depending on the size and complexity of the lobe.
Associated with the cyclic character of delta building is a
sequence of environmental changes.
These changes in turn influence the character of human utilization and exploitation.
(Modified from Gagliano and van Beek, 1975.)

182

\ /< RELICT _V ALLEY

* y'^WALL;

ALLUVIAL TERRACE

COASTAL TERRACE

\
ESTUARY
RELICT BEACHES

Figure

ACTIVE SHORELINE

ACTIVE BEACH COMPLEX

5-6l.

Distribution of initial occupation sites

associated with a coastal plain stream system.
A - oldest; C - youngest.

183

GULF SHORE
ACTIVE RECURVED
SPIT

COMPLEX

RELICT SPIT

COMPLEXES
DRIFT

Figure 5-7.

Initial occupation sites associated with sequentially developed,

recurved spit complexes.
This pattern could be produced either
by downdrift migration of the tidal inlet or re-occupation of the
same shoreline position following a fall and rise of sea level.
A - oldest; C - youngest.

ppwwww

RELICT SHORELINES

REWORKED ARTIFACTS AND

ECOFACTS, OCCUPATIONS A-D

PRESENT SHORELINE
DRIFT

Figure 5-8.

Initial occupation sites and reworked material on truncated

A - oldest: D - youngest.
shorelines.

184

this growth pattern.

This type of spit pattern may develop in two ways.

During periods of relative stillstand, inlet migration in

direction may result in this geometry.

downdrift

In such instances, the time in-

tervals involved may be a few hundred to a maximum of a few thousand

The pattern can also result from re-occupation of the same shore-

years.

line after an interval of sea level fall and rise.

In this instance, the

time interval between occupations may be thousands or tens of thousands

of years.

Figure 5-8 depicts a situation where a set of prograded relict shorelines (A-C)

has

been truncated at an angle by a later shoreline

(D)

Artifacts and ecofacts from sites associated with the initial relict shorelines may be winnowed out of eroded sites and redistributed along the later
shoreline.

In such an instance, a considerable hiatus may occur between

occupations C and D.

When shore sites are eroded by waves or wind action along a shoreline
or in a dune field > the coarse lag deposits left behind usually contain many

valuable data.

Artifacts of durable materials (bone, stone, pottery, etc.)

may be worn and rounded by abrasion and

other coarse inclusions.
the lag

reposited

along with shell and

Although the geometry of the site is destroyed,

deposits contain identifiable artifacts and ecofacts and are of

archeological value.

They indicate the presence of a site.

They contain

artifacts which may be datable and indicative of site activities.

Ecofacts

give clues to environmental conditions at the time of site occupation.

Eroded and reworked sites may be found in the distal end of delta
lobes (Figure

5-9).

Artifacts and ecofacts from eroded sites that were

once on natural levee ridges of distributaries may be redistributed along

185

the fronts of, and incorporated within the sands of, the transgressive

barrier islands that develop during the deterioration stage of the delta
cycle.

ABANDONED DISTRIBUTARIES

GULF SHORE

TRANSGRESSIVE DELTA
BARRIER ISLANDS

REWORKED ARTIFACTS AND

ECOFACTS, OCCUPATION A
Figure 5-9.

Initial occupation sites and reworked material in a

A - oldest; B - youngest.
transgressive delta.

Ecofacts
The remains of plants and animals occur commonly in archeological

sites and may be particularly abundant and well preserved in coastal

sites.

They include shells of pelecypods, gastropods, and other marine

organisms, bones, seeds, wood, peat,

f oraminif era,

pollen, spores, etc.

They may constitute the bulk of the archeological deposit,

may

be

hundreds

of

meters

in

length

186

and

ten

or

which

more meters

thick.

They may be confined to individual strata, such as bone beds or

organically rich deposits.

They may also occur as inclusions within a site

matrix of sand, silt, clay, or other sedimentary deposits.

If the ecofacts represent remains of fauna or flora that constituted

a part of the diet or were otherwise utilized by the site inhabitants,

they may contribute significantly to an understanding of past economic

activities.

Calculations of population size have been made for coastal

shell middens, and seasonality of occupation has been determined from

studies of annual growth rings of shellfish, deer antler, and seed remains.

Ecofacts are also of primary importance in providing samples for

radiometric dating.
In addition to their use in dating and cultural interpretation, eco-

facts are also of exceptional value in establishing the environmental con-

ditions in the vicinity of the site during its occupation.

Archeological

sites often contain concentrated samples of ecofacts gleaned from neighboring environemtns by inhabitants of the site.

Shellfish remains from coastal

sites provide an excellent illustration of the use of faunal remains and en-

vironmental interpretation.

Shellfish are sensitive indicators of salinity,

water temperature, turbidity, and other process factors that define environments and sub environments.

(For an example illustrating characteristic shellfish

assemblages from the western and central Gulf, see Parker, 1960.)

Consider

three molluscs commonly found in archeological sites of the northern Gulf

and their environmental implications:

*Molluscs *
Unio sp

*Salinity *
Fresh

Rangia cuneata

Brackish

Crassostrea virginica

Saline

Figure 5-10 depicts a hypothetical coastal estuary which has become fresher
through time as a result of blockage of its mouth by barrier spits and
187

'
S

o
u s
o

#\

cd
CD

<+-l

CO

<

d)
ha

o
ha

CO

tf>

(I)

e
O

J*

15

CD
ha

(0

0)
ha

5-4

M-l

00

z
g
r<
Q.

ex

3 3
CO
O
CU
o
o
5-i

0)

4-1

4.)

rH -H

LU

CD

OQ

<

5-i

TJ

UJ

Lli

'E

2
c
ha

ha
0)

(0

(0

>

(0

ha

o
3

'

CU

>,

M-l

5-1

>H

Cd

4-1

0)
CO

cd
i-i

CD

CO

rH

CD

cd

4-J

4-1

3
3 O
-H H

-H T3

X4J
S

CJ

(7)

</>

cc
< LU
Iz
< >

cc

<**

cd
5-i

4J

o
5-i

0)

iH

x:

ps

4-1

-H
rH

&
!>.

ja

(2
CU

e
cd
cu

4-1

3
cu
3 U

T3

Cu

co

CO

cd

co

5-i

CO
>>

x>

a,
&. T3
cd

CU

>

3 o

TJ -H r-l
rH
H

S
<4-l

O
4J

cd

"4-1

CO
>

5-1

CU

>

CJ>

CU

<
s^

CU

e
o
O

rH

CO

CU

cd

>, 00

CO

LU

4-1

I I

w
Q
N

CtJ

UJ

4J ^-\

cfl

Z
Z
LU UJ
Q Q
Q Q

<

CO

CO

CO
CU

TJ
rH

4-1

co

CO

O
O
o

5*

cd

rH

4-1

rH
rH
CU

CO

cd

cu
CO

00 rH
x:

d
m
r-i

CU
5-i

3
00
Pn

188

4-1

CO
ai

4-1

CU

CO

cd

3 3

.-1

cd

3 > 3
o cu y
-H rH a
4J
o

3 O
CO

4-1

CU
CO

M
3
3
O

>>

CU

>

5-i

cd

UJ

cu

rH

Middens located around

outgrowth of streams emptying into the embayment.

the estuary reflect this environmental change through changes in shell con-

Shell middens located on prograding beaches may be composed of dif-

tent.

ferent shell types, while a site on a bluff with a long sequence of occupation shows a vertical change in shell types.

Because of the dynamic processes associated with shore zone environments, archeological materials frequently become incorporated into
tary environments.

sedimen-

Common examples are sites on the banks or on point bars

of meandering streams, on actively prograding beaches, on coastal dunes,

and around ponds, bogs, or sinkholes.

In subsiding areas or in certain situations during intervals of rising

sea level, entire sites may be slowly covered with sediment.

Gradual sedi-

ment burial of entire mound groups, preserving the entire geometry of the
sites, is a rather common occurrence in the Mississippi delta area (Figure
5-11).

Archeological Sequence and Sea Level Fluctuation

The Ria Cycle

Clearly, a major fluctuation of sea level will bring about great

changes in coastal environments, landf orms

and settlement patterns,

which will be reflected in the archeological record.

An often-repeated

sequence of events in the Late Quaternary history of the northern Gulf

area is related to a fall and subsequent rise of sea level.

Such a se-

quence can be called a "ria cycle," since it produces drowned river valleys, or rias.

The ria cycle has been chosen to illustrate archeological relation-

ships related to sea level fluctuations because ria cycles have occurred on

several scales during the past 55,000 years and left a distinctive record

189

SHELLS

NATURAL LEVEE
Figure

5-11.

SILT

Shell mound on a subsided natural levee ridge enveloped by marsh

mud (After Russell, 1967).

within the study area.

Figures 5-12 and 5-13 illustrate a hypothetical ria

cycle and its effects on coastal landforras and the archeological record.
The ria cycle is especially important in the interpretation of conti-

nental shelf archeology.

As indicated in Chapter

a number of ria cycles

have unfolded within the time interval under consideration in this study.
As illustrated in Volume III, Plate 3, one major, and three minor ria cycles

have occurred during the past 25,000 years.

Each of these cycles presumably

had a major influence on the distribution of landform and associated

habitation sites within the study area.

190

a
o
H

U
CO

rH
0)

>

O
O
iH

CD

rH

CD

CO

-H

CD

CO

c/}

4J

cO

-H
qj
a,
rH >>

CO

M
z
co

^
>-

m
1

a)

M
&C

tc
H -H
5-1

3
T3

fin

C
H

4->

c c
3
B o
a- j:
O CO

CD

H
a;

cu

>

rH

CD

CJ

T3

>,

6
O

CD

CJ
V-i

<4-l

JZ
4J

XI

X.

cO

-u

rH -H

3
14-1

T3

OJ
CJ

-U

CD
cO

a -h
CD
O
3 O
cr
a)
co

rH

CO
(-1

OJ

XI

en
io

b
3

co

CO

3
o o
CO

>,

H -H TD
4J

4J

QJ

CD

co

4J

-C

4-1

4-1

cj

a rH3
s ^
>,

CN]

rH
LP)

0)

M
3
tod

Pn

191

CO

T3

3
H

V'

CO

CO

CU

62

CO

0J

UJ

X.

bC

o c

o>

T3

CU

U
cti

>
z

rl

gg

T3

[0

4J

CO

3
2

O
rH

en

UJ

0>

CO
CU

O
LW
O C
rl
3

CJ

-o

to

It)

cu

H
CJ

^
CJ

3
5

r*.

4-1

a O

3
5

HI

CO

4-1

3
5

CO

H
-l

U
^^

v ~

CU

rfl

CU

Vj

4-1

^H

4J

rt

>-,

CO
CO

CJ

r-

OJ

CO

3
5

111

3
a.

cd

x. rH

r-i

iH

!-J

m
1

111

rt

H
iH

4-1

o
lO

CO
UJ

LU

DC

tr

3
a

CO

cu

CC

CC

CC

4-1

CO

o
N

LTV

rl

rl

CO

DC

CC

3
LU

M
h

Fn

C
rl

rl

4J

OJ

CJ

cu
CO

111

CU
rl

C
5
-C
CO

iH
14-I

_l

<
Q

CO

111

3
s

CU

El

4-)

CU

rl

rl

>
CU

CO
"4-1

iH T3

c
CM

3
2

3
2

CO

CD

CO

r-i

CO

13

CO

a)

4-1

(3

+
<

3
UJ

CC

CC

3
2

3
LU
2

>

o
<

r-

CO

<

co

IL

a
CU
6
CJ

CO

CO

CJ

0)

rH
I

LO
cu

u
3
60
r-i

192

4-1

O u

co
UJ

CO

CO

rl

UJ

LU

UJ
CO UJ

CD

4-1

tH

13 C
rH H

-I-

CO

CO
r-i

rl

>
c
cu

CHAPTER VI
SELECTED TYPICAL ARCHEOLOGICAL SITES OF THE NORTHERN GULF

Site Inventory and Dating

An attempt was made to compile an inventory of known sites in the

northern Gulf area that may have been occupied from the time of formation
of the Ingleside shoreline (A2 interval circa 55,000 years B.P.) until sea

level was stabilized at approximately its present stand (the end of Interval
J,

3,500 years B.P.).

Site data was obtained from various sources, such as the Texas

Archeological Research Laboratory, at The University of Texas at Austin,

the Department of Geography and Anthropology at Louisiana State University
in Baton Rouge, and the Department of Sociology and Anthropology at the Uni-

versity of Southwestern Louisiana in Lafayette.

The Louisiana Archeological

Survey and Antiquities Commission, the Department of Archives, History, and

Records Management of the Florida Department of State, and the Temple Mound
Museum, Fort Walton Beach, Florida, added significant site information.

In

addition to the institutional sources, site information came from selected

amateur archeologists throughout Texas, Louisiana, Mississippi, Alabama, and
Florida.

A review of the literature on northern Gulf archeology also yielded a

long list of sites.

problems, however.

The use of such a literature search has its inherent

We are forced to base our site locations and occupation

dates on information which, at times, is fragmentary, poorly reported, poorly

analyzed, and sometimes entirely erroneous.

is always risky,

Relying on secondary sources

especially when data is poorly described and illustrated,

making a critical review of the interpretations most difficult.

In some

cases, published reports are so vague that it is impossible to locate sites,

even by county.

193

In dealing with unpublished site reports, another problem was frequently-

encountered.

The file cards often contain very cryptic descriptions, such

as "Archaic (?) flint cnips found."

Such locations were considered on the

assumption that most were, indeed, Archaic Period, initial occupation sites.
The site locations resulting from the inventory are presented on the map
in Volume III, Plate 4, entitled "Archeological Sites, Initial Occupation."

As indicated by the title of the plate, an attempt was also made to establish
the time of initial site occupation.

The locations of the sites are indicated

on a map of the area by a symbol which denotes the initial occupation period.

Tabular listings of initial and subsequent site occupations are given in

Volume III, Plate 5, "Projectile Points and Site Chronologies."

These tables

are divided by area, Western Gulf, Central Gulf, and Eastern Gulf.

For each

area there is a listing of the time span of the projectile points found there,
and also a listing of the time period of the site discovered there.

Although it was deemed important to present the occupational sequence

to the greatest extent possible, there are many pitfalls in establishing abso-

lute

and even relative dates.

The shortcomings of radiometric dating are

well-known and will not be discussed at length in this section other than to
note that the problems increase as the age of the material increases.
The value of utilizing artifact assemblages and "index" artifacts for

establishing relative chronology is a well-established tool.

The technique

becomes shaky, however, when correlations are made far from the place where
the chronological position of the artifacts has been established through con-

trolled excavation and absolute dating techniques.

methodology
invention
and

Other limitations in the

are time lags related to diffusion, the possibility of independent

of the same form in widely separated areas or at different times,

the many other difficulties related to using artifacts as indicators of

culture.

194

In spite of all these difficulties,

the validity of the methodology as

applied to a regional problem can be demonstrated.

For example, radiocarbon

assays of organic samples associated with fluted projectile point forms at

sites widely dispersed across the continent indicate that they were in vogue

during a relatively narrow time interval, and thus are valuable index forms.
For the period from 11,500 to 3,000 years B.P,, chipped stone projectile

points are the most universal and distinctive artifact type and are the basis
for estimating the chronological position of sites.

The term projectile point

is something of a misnomer, as many were clearly utilized as knives or other

special-purpose hafted tools.

Nevertheless, each artifact included in this

broad category is usually distinguished by a particular set of attributes

which is the basis for classification.

In some cases, such as the Clovis

points in the Plains area, the age of the points has been established.

points with similar attributes are found in Florida,

When

it is generally assumed

that they are the same age as those from the Plains, but at this time they

have not been securely dated.

Depending upon the area of origin of the fluted

point tradition, of which the Clovis point is an early form, and the routes of

dispersal from the area of origin, the fluted points of Florida could be older
or younger than those of the Plains area.

"Culture lag" plays an important

role in such instances.

For the period from 11,500 to 3,000 years B.P., chipped-stone projectile

points are the most universal and distinctive artifact type for estimating the

chronological position of sites.

In some cases, such as the Clovis and Folsom

points in the Plains area, the age of the points is well established.

However,

in many instances, either the absolute age of the point has been poorly estab-

lished or it may vary greatly across the study area.

chronologies listed in Volumes III, Plate

5,

The projectile point

reflect these problems.

For instance, the San Patrice point in Texas (see Volume III, Plate 5)
is seen as

beginning approximately 5,000 B.P. and lasting well into the first
195

millenium A.D.

In Louisiana, the time estimates are considerably ditterent.

Data from that state suggest

that the San Patrice point has a time range of

10,000 B.P. to 8,000 B.P., a difference in initial dates of 5,000 years!

"Culture lag" can play an important part in such instances.

Clovis points,

for example, which have been dated quite accurately in west Texas, may have

taken longer to disperse eastward to Florida, where such precise dates are
lacking.

Thus, one cannot reliably fix the initial, and even terminal, dates

on the point as being the same for the different areas.

That certain points are thought to have longer time spans in one place
than another is illustrated by the Lerma point, which in Texas has an estimated

range of 10,000 to 2,000 B.P., and in Louisiana of 9,500 B.P. to 6,500 B.P., a

difference of 5,000 years.

Since the charts shown in Plate

are compilations

of data listed in various publications dealing with regional chronologies, and

are subject to margins of error and assumptions in each publication,

the

charts should be viewed, in most cases, only as a compromise among several

publications.

When a projectile point's time span is the same for two or

even all three of the Gulf regions, then a more trustworthy

chronology can be based upon it.

The reader should be warned, though, that

in a few cases, the publications dealing with point chronologies copy dates

from each other, so that it can appear as if the point has the same exact age
and time span in two separate areas.

Point typology is another factor in regional projectile point comparisons.

Similar, if not identical, points are given different appellations in

different regions throughout the Gulf, although technologically they have the
same characteristics.

Big Sandy points from Alabama and Bolen points from

Florida are an example of this problem.

Certain point names, like Gary, have been used as catch-all terms to
include several varieties of points which have no really common attribute
except being crudely chipped and having a contracting stem.

196

Garys can be

shouldered or non-shouldered, range from 10 cm to

cm in length, and can

have been either knives, projectile points, or lance tips.

If the data base for projectile point tradition sites appears to be

thin,

then that for earlier sites can be described as very meager indeed.

We are limited to a mere half-dozen sites in the entire region that are re-

ported to have preprojectile point characteristics and descriptions and illustrations of cultural materials from these few sites are virtually nonexistent.

The chronological position of these earlier sites is based on a

few radiocarbon dates and stone tool assemblages.

One final word should be mentioned about Plates

and

5.

All sites

were listed by number rather than by name, since it was a less cumbersome
method.

However, in many cases, the responsible state agencies have not

assigned numbers to all sites, even when the name and exact location of the
site is known.

In such instances,

letters have been temporarily substituted

in the hope that eventually a number will be assigned.

Tate Cove Site is a good example.

16 EV "A" for the

Its location, artifacts, and age are

known, but a number has not yet been assigned.

In retrospect, the task of compiling an inventory of all known sites
in the Gulf coastal area dating older than 3,500 B.P. is not an impossible
one.

As one might expect, relatively few sites are known from the early

time intervals.

Selected Sites

From the published and unpublished data reviewed during the course of
this study, summaries of selected sites are presented.

For each of the three

regions into which this report is divided, a significant site, or sites, is

described in relation to its age, ecofacts, artifacts, associated landforms,
and other pertinent data.

Each of the three regions is subdivided into Pre-

projectile Point, Paleo-Indian, Archaic, and Poverty Point Periods.

197

Except

for Texas, where no Poverty Point site has been reported, all three regions

contain descriptions of sites of each period.

Although this section presents only a summary of existent terrestrial

site data, it does point out the great diversity of site geometry, landform

association, and cultural content, and is believed to be representative of the

prehistoric cultural resources of the continental shelf.

Western Gulf Pre-projectile Point

In the western sector Gulf area, as in most places in North America,

the evidence for human occupation beyond approximately 12,000 years B.P. is

fragmentary and scattered.

Accumulating evidence gathered over the years,

however, cannot be ignored, and Pre-projectile Point sites certainly warrant

strong attention.

Two sites in particular have played important roles in

the establishment of Pre-projectile Point

traits in the northern Gulf area.

or "chopper-scraper

"

complex

These are Lewisville and Friesenhahn Cave.

Lewisville, except for a problematical Clovis Point discovered in

Hearth No.

1,

has for many years been one of the pillars on which the whole

Pre-projectile Point theory rests.

The site was located in a large borrow

pit excavated by draglines of the U.S. Army Corps of Engineers in 1949-51

(Crook and Harris, 1958:233).

It was situated on the west bank of Elm Fork

of the Trinity River, just north of Lewisville, Texas.

Subsequent inundation

by water backing up behind the dam for which the borrow pit was originally

excavated has since sealed off the site.

Before this modern inundation, a series of 21 hearths was discovered

in the lower levels of the Upper Shuler geological formation (Wormington,

1957:58; Crook and Harris, 1958:234).

This formation of yellow, sandy clay

is believed to be of Lake Wisconsin age, and the range of fauna present at

the site would seem to bear this out.

Two radiocarbon dates, derived from burned vegetable fiber and solid

wood charcoal recovered form two of the hearths, measured in excess of 38,000

198

years.

Aside from the controversial Clovis Point, three other manmade arti-

facts were located in close proximity to the hearths.

pebble chopper, a scraper, and a hammerstone.

These were a crude

The possibility that these

three other artifacts were introduced onto the site in the same manner as the

Clovis Point (possibly a "plant," or by mixture of later material caused by

the dragline)

and that the hearths are really natural burned areas or wood-

rat nests and are not manmade^has caused skeptical reactions among some

archeologists.

According to the excavators, however, no such possibilities

occurred and the site was not affected by an over-zealous artifact "planter."
If one accepts the age and authenticity of the artifacts

(excluding

the Clovis point), then a picture of an ancient Pleistocene campsite emerges.

From analysis of both the flora and fauna, Crook and Harris (1958) detailed
a climate that was warmer and wetter than present.

The evidence of pond-

deposited clays also argues for this climate difference.

Crook and Harris

also hypothesized a repeated, seasonal occupation of the site by small groups

or bands of migratory peoples.

These peoples were "hunters of opportunity,"

not relying on one specific animal as the basis of their economy, but rather on

whatever they could catch.

The remains of larger fauna (mammoth, horse,

bison) found at Lewisville are believed to be animals killed and butchered

away from the site, with only selected portions of the carcass brought back
to camp.

There were no complete skeletons of these large herbivores found

at Lewisville.

In any event,

the site may well serve to illustrate an open

campsite, similar to possible sites of the same vintage located out on the

continental shelf.
The other Pre-projectile Point site that has received considerable

attention is Friesenhahn Cave, located on the Friesenhahn Ranch, 21 miles

north of San Antonio, Texas.

Evans (1961), in his summary article on the

cave, states that Sellards (1919) was the first to mention the wealth of

Pleistocene fossils being found by local collectors.

199

It was not until 1949

however, that systematic excavation of the cave revealed its true importance.
Over thirty

genera of mammals, birds, reptiles, and amphibians were dis-

covered (Evans, 1961:7).

However, the most striking find was that of over

forty "flints that show a crude but nonetheless definite pattern of chipping, resulting in steep, almost vertical edges with small, protruding
'beaks.'

These artifacts are plano-convex and cannot be accidental..."

(Krieger, 1964:45-6).

One such artifact, discovered beneath the articulated

skeleton of a large saber-tooth cat

Dinobastis serus ) is illustrated in

Figure 6-1.

Figure 6-1.
Steep, endretouched artifact found
beneath saber-tooth cat at
Friesenhahn Cave (After
Sellards, 1952).

^:

.:.<

..*.

Aside from these artifacts, a series of small bones was recovered

which showed cut and polished ends.

It is possible that these cuts were made

by a carnivorous animal, and the polishing resulted from the bones passing

through the creature's intestinal tract.

Whatever their true nature, the

bones, if taken under consideration with the artifacts, would seem to mark

Friesenhahn as a significant site.

The cave itself (Figure 6-2) was formed as a solution cavern in the

limestone bed underlying the southeastern part of Edwards Plateau (Evans,

1961:7).

The original entrance, through which the fauna traveled, opened

towards the northwest, but was eventually filled with debris so that all

200

EGEND
MAIN TRENCHES

SHALLOW
EXCAVATIONS

UNEXCAVATED AREA

PRESENT VERTICAL
OPENING FROM CAVE
ROOF TO SURFACE
CAVE FLOOR

SINK HOLE

IN

FISSURE

CAVE FILL

STALAGMITES (ABOVE SURFACE

OF FILL)

DEBRIS -FILLED PASSAGE FROM

IN

PLEISTOCENE OPENING TO SURFACE

F

MARGINS OF FILLED CHANNEL

DEPOSITS

SKELETON OF SABER -TOOTH CAT (DinobastfS serus)

SKELETON OF PECCARY (Mylohyus nasutus)

Figure 6-2.

IN

FLOOR

Floor plan of Friesenhahn Cave, showing trenches and articulated

skeletons
(After Evans, 1961).

201

outside traces of it vanished.

Present entrance to the cavern is obtained

by a vertical shaft approximately nine meters deep, through which the exca-

vators had to descend.

Four main trenches (Figure 6-2) were dug through the floor of the
cave, exposing four distinct zones of fill (Figure 6-3).

Zone 1, the

deepest, was composed of limestone blocks, gravels, and red clay.

Only a

few small mammal bones and some turtle shell fragments were recovered.
Zone

consisted of water-deposited clays in which were found carbon-

aceous vegetable remains, limestone grit, and coarse rock debris.

The zone

also contained the articulated remains of an adult and an infant saber-

toothed cat, and a skeleton of a peccary

Mylophus masutrus)

Under the

adult cat was found the scraper in Figure 6-1.

Zone 3 consisted of "banded, concordally-fractured,

gritty clay with

inter -bedded thin layers of small limestone and flint gravels" (Evans, 1961:
15).

The clays were deposited in ponds, and carbonaceous material was preThis zone also possessed the greatest quantity of

sent in the entire zone.

Included were Dinobastis and another saber-toothed cat

fossil vertebrates.

(Smilodan sp.), a large number of juvenile elephant bones

young American mastodon

Mammut americanum )

turtle carapace fragments, bison bones

(

Equus sp.), camels

dirus )

Elephas sp

a relatively large number of

Bison sp.), and the bones of horse

Camelops sp.), deer

bear, and coyote.

Odocoileus sp.), dire wolf (Canis

The diversity of fauna suggests that extinctions

had not begun to be evident.

Zone 4 is fill of a once-flowing stream channel on the cave's floor.
The channel fill is a mixture of older deposits reworked from below, frag-

mentary fossil finds, and rocks and clay.

Also occurring in this channel was

a large freshwater clam shell that could only have come from a stream several

kilometers away.

Either it was washed in from that distance or carried into

202

a
o
H
U
co-

rn
qj
l-i

o
H

43

a
CO
V-i

00
H
4-1
crj

M
4-1

CO
toO

C
H

5
O
43
CO
#\

0)

it

>
co

(3

43
cO

43
(3
QJ
CO

/-.

iH
H vO
u a\
Pn iH
0)

UJ
CM -

-J

<
o

iH

CO

CO

43

13
cO

>

#.

O >
(3
0)

W
V-i

OJ
4-1

M-l

Mh
<

H
H
o->

co

>H

43

M-l

H
4-1

M-l

o o
CO
CO
tot)

(3

4-i

-H
(3

iH

cO

(3

CO

O M
QJ
4-1
>
O QJ

OJ

CO

co
1

CO
CO

o
U

CO
vjO
QJ

M
3
toO

H
Pn

203

qj

43
4-1

M-l

the cave by man.

Although this is slim evidence for the occurrence of man,

it becomes more significant if one considers the scrapers and the possibility

of the polished bones as man's work.

According to Evans, the faunal analysis leads one to conclude that the
cats were catching the juvenile elephants and mastodons and dragging them

back to the cave.

The adults of the other species were probably brought to

the cavern in a similar manner.

This raises the possibility that if man were present at the cave during
the Late Pleistocene, as evidenced by the choppers and scrapers, then he may

have likewise played a part in accumulating the bones of the fauna.

Western Gulf Paleo-Indian

In a 1940 paper, E.H. Sellards described artifacts and vertebrate

fossils from a series of sites in the Mission River drainage of Bee and
Goliad Counties, Texas.

The sites (41 BE 2, 3, 4, 5, and 41 GD 5, 6) were

associated with the Berclair terrace, interpreted by Sellards as a riverine

equivalent of the coastal Beaumont (Figure 6-4)

Material was recovered along Blanco and Medicine Creeks, and controlled
excavations at the Buckner locality (41 BE 2) yielded both artifacts and
fossilized faunal remains.

Artifact finds at the other sites, although never

excavated to any major degree as at the Buckner Site, serve to illustrate that
this whole drainage area is a prime location for early man sites.

At the Buckner Site, the Berclair terrace is the surface of a sequence

of alluvial fill deposited in a stream valley cut into the Goliad formation

(Pliocene)

(Figure 6-5).

Since deposition of the Berclair sediments, a second

interval of entrenchment has occurred, and a sequence of fill is in progress.

In the lower reaches of this younger valley, which is entrenched into the

Beaumont formation, the stream is now aggrading its channels.

204

A.

150 HORIZONTAL

Figure 6-4.

Relationship of Berclair Terrace to older Tertiary

Goliad Formation, Pleistocene Lissie Formation,
Beaumont Formation and the Late Holocene floodplain
deposits.
The upper section of B is through the
confined valley, not the upper reaches of the stream.
The lower section is in the vicinity of the Lissie
Formation.
The "fill" and "low terrace" deposits
are considered to be Late Holocene (After Sellards,
1940).

Deposits of the Berclair formation range from light-gray to dark-gray

sand, gravel, silt, and alluvial clay derived from Tertiary rocks of the re-

gion, particularly the Goliad formation.

A typical section consists of cross-

bedded gravels at the base, poorly stratified sands in the middle, and silts
and clays at the top (Figure 6-5).

The percentage of gravel and coarse sand

decreases in the lower reaches of the valley, and secondary calcium carbonate
in the form of caliche, in varying concentrations, is everywhere present.

205

ARTIFACTS
Stake C
f

Stake

Stake E

LU

Z
LU

SILT

O
o

-1.5

</)
ULi

SAND

3.0

E
c

GRAVEL

4.5

HI

z
w
o

60

\^m
i>

ri

ii

aas

PLIOCENE FOSSILS
:

:
i

^jfaaajIlIllAliilililMfcMithliilllhhMSliiM

VERTICAL DISTRIBUTION
1

inch = 9

Vertical scale

inch =

Stake C

Artifacts
# Artifact or cluster of

m
3 m

Horizontal scale

o Fossil or cluster of Fossils

Stake D

7
1

D:
c>
2r

a
a

,p

aa

a
a

<

HORIZONTAL DISTRIBUTION
Scale

inch =

Artifact or cluster of Artifacts

a Fossil or cluster

Figure 6-5.

^c7

oa

og

aa a

"aWn

*
T"

t-

1Z

O D

*.

Stake E

of Fossils

Horizontal and vertical distribution of artifacts at

the Buckner Site (41 BE 2), showing stratified zones
of gravel, sand, and silt
(Alter Sellards, 1940).

206

The excavations at 41 BE
the edge of the terrace.

were carried out in massive degrees along

The silty

Two distinct horizons came to light.

upper horizon (0.6 - 2.3 m)

hearths (Figure 6-5).

yielded about 200 flint chips, burnt rocks, and

Among the artifacts were a Clear Fork gouge, a leaf-

shaped knife, crude, leaf -shaped blades, a hand axe, and a Morhiss, or Peder-

nales point.

Also in this upper horizon were areas where mud -dauber nests

had been hardened by fire.

No fossils occurred in this horizon.

Below the upper horizon was a sand stratum void of artif actual content.
This extended from 2.3 to 4 m below the surface.

The lower horizon of gravel was found to underlie this sterile, sandy

level and extended down from

4 -

5.8 m

in depth.

In this horizon, Sellards

discovered points which appear to be (from published photographs) a Folsom,

a Yuma,

two unf luted Clovis bases, and a possible Big Sandy type.

Associated

with these were Clear Fork gouges, oval knives and blades, circular scrapers,
scrapers with denticulate spokeshaves, an axe or adz, fire-hardened mud-dauber
nests, and assorted Pleistocene vertebrate fossils.

It is

this lower hori-

zon which shows clear Paleo-Indian traits, and its separation

by sterile sand

from the upper Archaic horizon which helps support the belief in the antiquity
of the site.

This series of sites is significant for several reasons.

First, it

provides a model for sites associated with alluvial terraces along small,
coastal plain river systems.

Similar relict features appear to be quite com-

mon on the continental shelf.

The second reason for their importance relates to the relative dating
of the late Quaternary features in the study area.

For some reason, this im-

portant work by Sellards has been lost in the literature, and its implications
have escaped the attention of recent investigators of the region's Quaternary
geology.

Putting Sellard's well-described findings in the terminology used

in this study, we arrive at the following interpretations:

207

Formation

Culture
Period
-

Late Holocene

I-K

Archaic

Middle Holocene

Low terrace and fill

Upper Berclair
Lower Berclair

Period

Interval

Early Holocene

Paleo-Indian

Lissie

Pleistocene

Goliad

Pliocene

(?)

(?)

Since Sellard correlates the Berclair terrace with the Beaumont terrace, these data present a serious challenge to the generally accepted age

estimates of the Beaumont terrace and its equivalent across the Gulf coast.
Most geologists would estimate the Beaumont to be 25,000 to 120,000 years old,
Two additional locales should be mentioned in our consideration of

possible relationships between Paleo-Indian occupations and the Beaumont

terrace in the western Gulf area.

Both are beach deposits where the present

shore is eroding into deltaic facies of the Beaumont Formation.

At both

places artifacts and bones of extinct Pleistocene vertebrates have been

collected along the beach.

The first of these locales, the Sargent Beach Site (41 MG 4), is lo-

cated between the mouth of the Brazos River and Matagorda Bay.

In 1957

James Searcy reported that the site consisted of a low, black clay bank

approximately 45 cm high, fronting on the Gulf

(Files of the Texas

Archeological Research Laboratory, University of Texas at Austin).

Searcy

reported finding bison bones, fossil shells, "elephant" bone fragments and
tusks, slivers of mastodon teeth and several flint artifacts.

In 1970 W.

B.

Neyland reported that the locus of the black clay bank had completely eroded
away, but that mastodon teeth were still found along the beach.

208

This locale

is significant because it is situated upon relict natural levees associated

with a late Pleistocene or Early Holocene delta of the Colorado River.

A similar situation is found at McFaddin Beach (41 JF "A") located

between the mouth of the Sabine River and the High Island salt dome (Figure
Here the modern beach is being nourished by material derived from the

6-6).

Beaumont formation, which is eroding offshore.

Relict distributary channels

and natural levees of an old Trinity River delta exposed on the surface of
the Beaumont terrace just north of a narrow belt of Late Holocene marsh

presumably extend to the offshore area beyond the beach.

It is from the surface of,

or within these, Beaumont deposits that

artifacts and bones are being eroded.

Projectile points from the beach in-

clude Eden, San Patrice and Clovis-like fluted forms.

(

Mammut sp.), giant tortoise

horse

Geochelone sp.), camel

Bones of mammoth
(

Camlops sp.), and

Equus sp.) have also been found.

A neighboring Rangia shell midden, the Willow Lake Site (41 JF 28), also
associated with a relict distributary levee of the Trinity River, has produced
a single Scottsbluff Point along with pottery sherds and other

artifacts.

Western Gulf Archaic

One major problem which confronts archeologists studying the Texas

coastal region is the fact that the Archaic of this area is believed to be
an extremely long-lived tradition.

The Aransas Culture of the south and

central Texas coast, for example, has an estimated time-span ranging from
5,000 years B.C. to 1,000 A.D.

It is very difficult to place a site of the

Aransas Culture in an accurate time frame without absolute dating controls,

such as C-14 dates.

The Johnson Site (41 AS 1; Campbell, 1947), the type-

site of the Aransas Culture, is a case in point.

The site was excavated and

reported on before the advent of radiocarbon dating.

Thus, its position in

the 6,000 year history of the Aransas Culture is difficult to determine from
the published data.

209


cu

>s

U-l

4J

cu

H
4-1

CJ
CU

CO

S-i

<4-4

VJ

cu

>-l

o.

05

4-1

<D

y-i

X4J

CO

-a

Mc

G
O

CO

o
cu
CO
S-i

3
O

CJ

CU

T3

iH

4-1

r-l

CO

a.

CO

O
G

o
0)

TJ
a)
CJ

G
O

0)

x
H

T3
V-

(X

cfl

U-4

CO

CO
CU

a)
cu

4J

>

CO
CO

CO

0)

i-H

4J

4J

CO

CO

)-l

X X
X

cu

4-1

>

4J

CU

CO

cu

CJ

a
0)

13

<u

3co

r^
c0o\

4-i

CH

>,

cOiH]

4-1

4->

CO

X
H

H
0)

k,-*

CO

4-1

CU

CJ
CO
a)

rH

P-.

4-1

CO

CU

x>

4-1

CJ
CO

G
i-l

* 4J

X
a)

cu

4-1

H
CO

xo

O
M
U-l

X
T3
H

CO

H
4J
CJ

CU

4J

U-l

CO
1

CO

pa

Xo

T3

CU

CU

x4J a

CO
(U
CJ

CO

pt,

&
CU

s
CU
CO
CO

4-> PQ
H
-o T3 CO *-'
X) C H

CO

CO

a
o
W

CO

4-1

CO

i-i

CO

fn
CJ

0)

v/
0)

M
3
00

210

G CO CU
H
>
O H -H
a- O 06

For that reason, we have chosen an Archaic site from the northern Texas
coast and have avoided the problem of exactly how old an Aransas Culture Site

really is.
The introduction of pottery into the Galveston Bay area has been securely
dated through the efforts of a number of workers at about 100 A.D.
1967; 1973; Aten and Bollich, 1969; Aten, 1971; and others).

(Ambler,

The early ceramic

periods of this area are generally thought of as "ceramic Archaic," that is,
the continuation of the coastal Archaic tradition with the addition of

ceramics to the artifact assemblage.

Thus an Archaic site (i.e., pre-

pottery) of this northern coast area would be at least 1800 years old, and

probably older.

It would then have been occupied during a period of slightly

lower sea level, and sites of similar nature may be expected to be located
on the continental shelf.

The Jamison Site (41 LB 2) in Liberty County, Texas (Aten, 1967), is

just such a site.

Although the upper levels are rich in pottery, the lower

stratigraphic levels of the site contain artifacts, which are probably

Late Archaic.
The site itself is located on an outcrop of the Deweyville terrace at
the western margin of the Trinity River f loodplain (Figure 6-7)

This loca-

tion would have afforded the occupants of the site both the uplands of the

terrace and the lowlands of the floodplain in which to carry on food procure-

ment activities (Aten, 1967: 1).

Excavations carried out by members of the Houston Archeological

Society from 1959 to 1961 revealed the site to be an earth midden.

analysis units were therefore established.

Four

Aten concluded that these lower

two units represent a preceramic occupation of the site.

Included in these

units were dart points of the Ellis, Gary, Neches River, Palmillas, and

Williams types.
at 4,000 B.C.

Williams points have a time range beginning approximately

and lasting well into the time of potter.

211

Palmillas, Gary,

u
1500

Legend
topogrophic scorp
(23')

overoge elevation of
physiogrophic surface

m"
DAYTON

&.

Figure 6-7.

Location and physiography of the Jamison Site, 41 LB

(After Aten, 1967).

and Ellis points have dates assumed to range from about 2,000 or 1,000 B.C.

into pottery times.

It can then be inferred,

if we disregard the earliest

dates for Williams points, that these two units would have been occupied

sometime between 2,000 B.C. and 100 A.D.

Similar Late Archaic tool and point assemblages have been found at
sites in Texas and Louisiana, although no detailed comparisons have been

attempted as yet.

These sites could conceivably be contemporaneous with the

212

lowest levels at the Jamison Site.

It can also be assumed that some sites

having such Late Archaic manifestations at one time existed at locales which
are now beneath the surface of the Gulf of Mexico.

Although wave action

would have altered them in some instances, it is possible that they would
exhibit features similar to the Jamison site.
The exact relationship between the site and the Deweyville scar has

unlikely that the Deweyville scar

not been established, but it is not

was active during the period of initial occupation.

is located at an important ecotone

the escarpment between the Beaumont

surface and the Trinity River floodplain.

are

likely locales for Archaic sites.

nental

In any case, the site

Such valley wall escarpments

Similar drowned locales on the conti-

shelf represent high- probability areas for site occurrence.

Central Gulf Pre-projectile Point

We will be concerned here with two important discoveries believed to
be of Pre-projectile Point age:

the Natchez Pelvis find in Mississippi and

the Salt Mine Valley Site on Avery Island, Louisiana.

In 1846, a Natchez physician by the name of M.W. Dickeson published a

short account of some fossilized bones he had collected during the period

from 1837 to 1844 (Dickeson, 1846).

He described the remains as coming from

a gulley overlain by 9.1 meters of sediment.

In addition, these remains were

said to have been located some 60 cm to 90 cm below the skeletons of three

Megalonyx in a stratum of blue clay.

human pelvis.

Among these fossils was a black-stained

The pelvis has since been lost, but its stratigraphic position

and relation to extinct Pleistocene fauna have made it one of the most famous
of all early man finds in the New World.

In 1889, Joseph Leidy presented a listing of the fauna found by

Dickeson to be associated with the pelvis.

213

These were Megalonyx jef f ersoni

Megalonyx dissimilis

Ereptodon priscus

Mylodon harlani

Mastodon americanus

Equus major, and Bison latifrons (Leidy, 1889:9; cited in Ouimby, 1956:77).
In 1895, Dr. M. T, Wilson reported the bones of a Mylodon

and the

Natchez Pelvis to be of the same age after analyzing them by the fluorine
dating method (Wilson, 1895:725).

The Natchez Pelvis data then lay dormant

and practically unknown until 1951 when T.D. Stewart rediscovered the sig-

nificance of the find (Stewart, 1951).

Finally, in 1956, George

Quimby, Jr., pursued the Natchez Pelvis

I.

problem all the way to the actual creek bottom from which the bone had
emerged over 100 years before.

His concise summary of the situation

surrounding the find and the exact creek location of the pelvis are
adequately reported, and there is no need to review all of his findings here.

Suffice it to say that any attempt to locate more portions

of the skeleton of which the pelvis was a part would prove futile due to the

intense eroding of Mammoth Bayou, the creek in which the pelvis was originally
discovered.

However, the loess bluffs surrounding Natchez are prime loca-

tions for more finds of this nature.

These loess bluffs are deposits of wind-blown silts derived from exposed channels and bars of the Mississippi Valley, located immediately to the
west (Figure 6-8, 6-9, and 6-10).

The fossiliferous blue clay at the bottom

of the loess deposits has been interpreted as an eroded surface of the Mont-

gomery terrace (Quimby, 1956:78).

The geology of the Lower Mississippi Valley loess deposits has been

studied by a number of recent workers.

Among the most important recent

work is that of Snowden and Priddy (1968) and Otvos (1975)

These authors

have described the distribution and stratigraphy and cite radiocarbon dates
from two locales.

Dates from a roadcut along U.S. Highway 61 near Vicksburg,

Mississippi, are shown in Table 6-1.

The stratified loess at this locale

was deposited between 24,000 and 18,000 years ago during the Farmdalian and

214

'

CO

H
CO
CO

>
oo

O
H
O
CU

cu

4=

u
-U

X)

CT3

Z H
m
o

CO

H
4J

H>

a>

>.

c a.
O N

CU

>
CJ

c
H

e
o
H

4-J

cd

*"-

r-.

vo
o>
.-1

4J

a
a)
CO

u
H
00

00
H

CU

4J

U o
CU

CO

o
r-<

c H
3
cfl

en

00

CU

6
o

5
o

M-t

T3
CU

CO

-a

M
N
H

t-i

CU
r-

H H

H
U-(

(X -H
ni
(X T3
0)
T3 H O

ao
1

t0
CD
!-i

3
BD

P^H

(taai)

N0I1VA3H3

215

CO

'SB
tt%j

v
-

.-T^!
,

^soSB

Pfflj

BB

MwB

hmiJM

Figure 6-9.

"*

Fossil locale on Tunica Bayou, West Feliciana

Parish, Louisiana.
The situation here is believed to be somewhat similar to the Natchez
Pelvis locale.
Fossil bones of Mastodon deposited on an erosion surface on Miocene marine
clays are exposed by stream erosion.

*.* -i\. *fttffc ;4U *

Figure 6-10.

Loess deposits in roadcut along

U.S. Highway 61 near Vicksburg,
Mississippi.

216

STRATIGRAPHIC
UNIT

RADIOCARBON AGE
(YRS.

B.P.)

17,850 + 380*

18,200 + 500*
18,640 + 380*

WOODFORDIAN LOESS
(INTERVAL "F")

19,200 + 420*

19,250 + 350*
20,500 + 600*

21,270 + 440*

WOODFORDIAN LOESS
(INTERVAL "E")

21,800 + 500*
22,600 + 700*
22,600 + 800*

FARMDALE LOESS
(INTERVAL "D")

23,550 + 750*
23,550 + 1000*

25,600 + 1000

FARMDALE SOIL?

*FOSSIL PULMONATE GASTROPOD SHELLS

A
FOSSIL WOOD

Table 6-1.

Radiocarbon ages of Vicksburg, Mississippi,

loess deposits (After Snowden and Priddy,
1968).

217

Woodfordian substages of the Wisconsin.

Only the upper Woodfordian loess

unit, believed to be correlative with the Peorian loess of Illinois, is

present at Natchez.

South of Natchez, Otvos (1975) has dated this upper

unit at Tunica Bayou in West Feliciana Parish, Louisiana.

He reports radio-

carbon dates on snail shells of 20,690 + 250 years B.P. and 21,750 + 310 years
B.P.

The Natchez Pelvis, then, would appear to have a minimum age of about
18,000 years, when loess deposition is believed to have terminated.

The

age could be more than 22,000 years old, when deposition of the Woodfordian
loess began.

The Natchez find is important because it provides evidence that

man was in the central Gulf coast area during Interval E or F, before or during
the maximum Wisconsin low stand of the sea.

Thus, we do have some basis, how-

ever tentative, for the hypothesis that archeological sites may be associated

with features on the continental shelf formed during the Late Pleistocene Early Holocene rise.

This coincidence can be seen clearly on the chronology

plate, Volume III, Plate

1.

In the southern portion of Louisiana, five major salt domes protrude

above the surrounding low- lying terrain (Figure 6-11).

On one of these,

Avery Island, some of the earliest evidence of man in the central Gulf area
has been recorded.

Gagliano (1967) published results of the work he had been conducting on

Avery Island since 1962.

He also reviewed all previous explorations at the

island carried out by Owen (1863), Leidy (1866; 1889), Joor (1895), Veatch
(1899)

and others in which the presence of Pleistocene fauna and artifacts

were revealed from what was termed the "bond bed" (1967:32).
The Salt Mine Valley Site (16 IB 23), the actual location on the

island where evidence of early man Pleistocene vertebrate fossils became

apparent, seems to have been formed by a repeated process of stream valley

cutting and filling (Figure 6-12)

218

CO

c
CO

H
en

3
O
CO

u
4->

c
o
l

4-

O
CO

en
}-(

d
p
CO
CD

u
H
43

CO

00

O
H
CO

>>
43
P,

^
O

oo
C3

CO

LO

H -H

e
CO

J-l

00
CO

CU

CJ

4-1

M-l

>H

CU

00
H

219

c
>
M
CO

<

60

d
H

R
co

4J XJ

d c
0)

td

CO
CD

en
CM

-n

PQ
H

U
PM

CO
i-H

03

O
o
o
r>

vO

>
Vj
0)
4-1

Hd
CD

2
CD

d
H

S
n)

CO

r~~
CD

c^
iH

4-1
s

O
O d

co

^ H

u
o
4-1

co

i-l

60
co

o
(1)

i-l

4-1

CO

M-l

<

H v-'
60

O
60
o d

i-H

CD

60
CD

-H

H
H
-H
M-i

4-1

>s
Ph

>

CD

CO
CD

CO

Hi

P3 T3

O
O
o

14-1

d
co

co
PC

>

CO

^3

,c

TJ

60

co

00
CO
1

o
o
o

iH
CO

>
S-i

<u

CD

-H

4-1

i-l

4-1

CO

CD

cj

a
H

l-i

CO
<D

3
O

4-J

Pm

co

CD

h
3

60
H
Pn

220

i-i

Gagliano Q.970; 16) hypothesized that in the first stage of this process, 12,000 - 11,000 B,P.

upthrust rock salt.

(Intervals R^ and H 2 )

"Stream erosion exposed

The surface of the rock salt was very irregular, and

there were deep solution pits.

Channel sands and gravels and organic debris

collected in solution holes and over shallow pits."

During the second phase (11,000 - 8,000 B.P., Intervals H3 and H^)
of these geological changes, the valley began to fill with sediments carried

by Iron Mine Run, the stream draining much of the island surface and traversing Salt Mine Valley.

This stream and others like it shifted back and

forth across the valley, depositing an interbedded sequence of channel sands

and gravels and overbank silts and clays (Gagliano, 1970: 16).

Phase

(8,000 - 6,000 B.P., Intervals H 4 and I) of the valley was

marked by erosion and/or solution of the lower salt deposits, which resulted
in a new, smaller valley within the older valley walls.

The final phase (6,000 B.P. - Present, Intervals

by another period of stream sediment deposition.

and K) was marked

Erosion of the banks of

the older valley walls has also added to the fill.

In the initial excavations at Salt Mine Valley by Gagliano, begun in

1962, the key bed, from which Joor, Veatch, and others had recovered fossil

bones, was exposed at a depth of approximately six meters and continued down
to about seven meters.

In 1968 the work in Salt Mine Valley was continued with a program

of core drilling and excavating.

During this study, the bone bed was

dated and tied into the same bed exposed in Pit V, which up until then
had not been accurately dated.

The dates and a description of the fauna

and artifacts recovered from the fossiliferous layer are reported below.

221

During 1969, the International Salt Company began excavating a new mine
shaft in the valley (see Figures 6-13 and 6-14).

A pit twelve meters in dia-

meter was excavated down to the basal rock salt.

This was accomplished by the

use of a clam-bucket working with a steel casing.

The material was removed

by levels and transported to another part of the island where it was systematically

Material was removed from a total of four

arranged for screening and analysis.

levels, reaching a depth of 9.6 meters below the surface (-2.8 meters MGL)
(Gagliano, 1970:7-8).

In 1970, part of the fill from the lowest levels

(-.46 meters to -2.80 meters MGL) was screened.

Again, it was clear that the bone bed was the prominent feature in which

numerous fossil Pleistocene vertebrate bones and human artifacts were found.
The artifact assemblage consisted of chipping debitage, anvil stones, hammerstones, bipolar core tools, steep-edge chipped tools, vegetable cordage,

cane basketry, and a single bone projectile point (Gagliano, 1970:11).

levels from -2.5 to -2.8 meters MGL, shells of Rangia cuneata

In the

brackish-

water clam, were found in remnants of a possible hearth, thus pointing to the
likelihood of a marsh environment relatively nearby at the time of occupation.

Man's early presence at Salt Mine Valley is marked, as just stated, by a

rather diversified assemblage of artifacts within the fossil bone bed, which
itself was deposited in the lowest levels of the site, atop the basal salt.
This bone bed has been dated (Gagliano, 1970) at 10,900 B.P.
300 years and 11,950 B.P.

(8,950 B.C.)

(10,000 B.C.) + 300 years.

In this stone bed, and thus assumed to be in association with the

artifacts, Joor (1895) reported the remains of several extinct species.

These included: dire wolf

Floridanus )

Equus complicatus )

harlani )

Me gal onyx sp

Canis dirus ), saber-toothed cat (Smilodon

giant, big-horned bison

eastern horse
frlylodon

Bison gigantobison latifrons ),

plains horse

mammoth

Mammut sp

capybara

Hydro choerus )

siscutata ) (Galiano, 1967:40).

mastodon
,

Equus scottii )
(

Mastodon sp

and giant tortoise

ground sloth

ground sloth

Geochelone cros-

A radiocarbon date on some fossil ground sloth

222

<u

>
H
4-1

cO
i

CD
j-i

X)

c
0

*r

C
Q

Tl
4J

Cd

U
X

0)

CO
QJ

iH

o
,C
0)

n
o
M-l
.

/^-N

o
r^

33

d
o

CT\

iH

4J
cO

o,

O
c
cO

H
6C
(S

M
s
<*>

H
to
CO

J-i

CO

cd

* U
^"N M-l

cO <C
CN v-^
PQ

vO
iH

en
CD

u
d

4-1

>>

4-4

CO
01

CD

i(

CO

>

0>

CD

O
CO
M-l

Vj

3
.03

M-l

4-1

rH

0)

cO

bO

CO

cd

i
I

bO
H

223

Historic

European artifacts

Ploquemine period pottery

e Prepared cores ond blodelets
f

Bipolar artifacts
Gravers, denti cu otes,
o Extinct found remain

Ridge ex cav otions

Pit

cordacj

Boreholes

&

7750 BC 550

8^50 BC
10,050 BC

2
*

3.00

300

?^W:W^W^ 7^ ^ r
b

+
+

Figure 6-14.

shaft

jj

New mine

Chipped grovel

'

'

Ale;

+\? / +

\-'-r

4-

+_

Idealized cross*section through Salt Mine Valley (16 IB 23).

MGL reference datum equals mean gulf level. For locations
of control sections, see Figure 6-13 (After Gagliano, 1970).

vertebrae, originally found by Joor in 1890, yielded the age of 8,390 B.P.
(6,440 B.C.)

+ 140 years (Gagliano, 1967:34).

However, the vertebrae used for

this date lacked precise stratigraphic location within the bone bed.

Two distinctive lithic assemblages have now been identified from the

fossil bone bed.

The first, recovered from Pit V in the 1962 excavation, has

been describee as a bipolar

assemblage (Gagliano, 1967).

reported a similar industry from the Debert Site

MacDonald (1968)

in central Nova Scotia, where

the material was associated with fluted projectile points.

An excellent se-

quence of radiocarbon dates obtained from hearth charcoal at Debert yielded

an average date of 10,600 + 47 years B.P.

MacDonald groups the products of

the bipolar technique into a class of artifacts called pieces esquilles

224

These artifacts, first recognized in Upper Paleolithic assemblages of the

Old World, are believed to have been used as wedging and slotting tools

associated with the groove and splinter technique of working bone, antler,
ivory, and hard wood.

MacDonald (1968:89) makes a basic distinction between pieces esquilles

and bipolar pebble cores, the former being incidental products of the use
of small pebble fragments in splitting and grooving bone and related ma-

terials.

In contrast, bipolar cores may produce similar forms, but were

probably designed to yield usable flakes.

Following this reasoning, Mac-

Donald describes five types of pieces esquilles from Debert

1)

rectangular

(multiple columnar fractures resem-

(small wedge forms); 2) "pseudo-cores"

bling remnants of polyhedral or cylindrical flint cores from which lamellar

flakes were struck); 3) "pseudo-burins"
of true burins,

(lack negative bulk characteristic

thus believed to be incidental to bone-splitting function of

the artifact class); 4) on-end scrapers (resulting, again, from splitting use);

and 5) on-other artifacts.

It should be noted that at Debert, both hammer-

stones and anvil stones are associated with pieces esquilles

While the possible splitting and grooving functions of the bipolar

artifacts from Salt Mine Valley were recognized, their classification rested
on the assumption that the basic forms were produced by direct hammerstone

blows directed to pebbles placed on anvil stones.

This conclusion was based

not only on the occurrence of anvil stones and hammerstones, but also by ex-

periments which produced comparable forms.

Characteristics of the bipolar

types described from Salt Mine Valley are illustrated in Figures 6-15
and
6-16.

The types include longitudinally split pebbles, side-split pebbles,

split, pebble-core fragments, cores with flat striking platforms, triangu-

loid cores, fusiform cores, irregular blocky cores, core scrapers,

cortex
flakes, multiple-scar flakes, and
beaked forms.

More recently, burin forms

have been recognized on bipolar cores

from Salt Mine Valley.

225

TRIANGULAR
CROSS - SECTION

PRISMATIC

CROSS

SECTION

FUSIFORM CORES

LONGITUDINALLY
SPLIT PEBBLE

WEDGE -SHAPED

CORE AXIS

PLATFORM

EDGE

ANGLE

90

REBOUND
FLAKE SCAR
PITTING

CORE WITH

TRIANGULOID CORE

SIMPLE PLATFORM
BEAK

CORTEX

SIDE SPLIT PEBBLE

5

CM.

Figure 6-15.

Characteristics of fr'pmlar cores from Pit V, Salt

Hine Valley (16 IB 23) (After Gacliano, 1967).

226

m
5
1

Figure 6-16.

Bipolar artifacts

cm

from Pit V, Salt Mine Valley Site

A-C, Longitudinally Spilt Cores; D, I,
J, Core Fragments; E-G, Multiple~Scar Flakes; H, Fusiform Core; K, L, Side Split Pebbles; K, L, M, Beaked
Forms (After Gagliano, 1967).
(16 IB 23).

227

This typological discussion has been introduced here because the

general class of artifacts is believed to be broadly distributed in the

northern Gulf area and highly indicative of Paleo-Indian and earlier Lithicstage occupations.

A second, highly significant lithic assemblage came to light from

Salt Mine Valley as a result of the 1969 excavations in the new mine shaft

The artifacts were manufactured by steep-edge chipping on relatively flat

stream pebbles (Figure 6-17and 6-18). The resulting scalloped edges were utilized
as gravers and burins.

Some have been worked down or rounded from use.

These artifacts are reminiscent of those described by Krieger (1964:46) from

Friesenhahn Cave (see Figure 6-1)

As previously noted, the bone bed is remarkable not only for the stone

tools which it has produced, but particularly for the quantities of bone and

perishable artifacts.

The excavations in the new mine shaft yielded

number of perishable artifacts.

a.

A small fragment of basketry or matting

made from split cane was found in lumps of sand.

(Figure 6-19), the plaiting is unmistakable.

Although the piece is small

Cordage occurred both as short,

individual strands (Figure 6-20) and as interwoven bundles.

The bundles sug-

gest sandals or fragments of larger, loosely plaited fabric.

A socketed, bone projectile point fashioned from the long bone of a

small animal or bird was found partially encased in a grass bundle in the

lowest level of the excavation (Figure 6-21).

Distinctive grooves or cutting

marks and polishing are evident

Several small pieces of wood that appeared to be cut or intentionally

burned on one end were recovered.

A small stick with a distinctive flat cone

on one end is illustrated in Figure 6-22.

It should be noted that while the bone bed in the

new mine shaft has

been stratigraphically correlated, where dates have been obtained with Pit

228

Steep, edge-chipped artifacts from the

New Mine Shaft, Salt Mine Valley.

Figure 6-17.

Figure 6-18.

1^^

|^ j*

^>

Steep, edge-chipped artifacts from the New

Mine Shaft, Salt Mine Valley.

229

Figure 6-19.

Fragment of split-cane basketry from New Mine

Shaft, -2.5 to -2.8 meters MGL, Salt Mine
Valley Site, (16 IB 23).

W"""^

^i^sefEk^

iljljjjljllip;

WT****

t"

jjt-*

Figure 6-20.

*m

}a

Pieces of three-strand cordage from New Mine

Shaft, -1.89 to -2.23 meters MGL, Salt Mine
Valley Site.

30

Socketed bone projectile point from New Mine

Shaft, -2.5 to -2.8 meters MGL, Salt Mine
Valley Site.

Figure 6-21

Figure 6-22

Cut wood from tfew Mine Shaft, -2.5 to -2.8 meters

MGL, Salt Mine Valley Site.

231

V and Boreholes

B M and "I'', organic material from the new shaft has not been

satisfactorily dated,

There remains the possibility that the two assemblages

from the valley are somewhat different in age.

An alternative explanation is

that they represent different activity areas that were more-or-less contempor-

aneous.

The pre-projectile point component at Salt Mine Valley can be summarized

as follows:

During phase

1)

of the valley's geological history, while the valley

was forming and the rock salt was exposed, wandering hunting-gathering peoples
left artifacts in stream and solution pit deposits.

artifacts are bones of a diversified assemblage

Associated with these

of extinct Pleistocene verte-

whether these animals were killed at the site or killed elsewhere and

brates,

brought to the site has not been established, but the latter explanation seems
most likely.
2)

Following this, while Iron Mine Run was slowly filling the valley

with sediment, artifacts and bones were again incorporated within the deposits,
this time along the margins of small streams.
3)

Radiocarbon assays have tentatively fixed the pre-projectile com-

ponent at approximately 12,000 years B.P.

4)

Salt domes with surface expression may have provided attractive habi-

tats for early inhabitants of the region.

It follows that submerged salt domes

on the continental shelf should be regarded as high-probability areas for site

occurrence.

Central Gulf Paleo-Indian

As with the preceeding section on the Pre-projectile Point Period,

Avery Island will again serve as our principal Paleo-Indian site.

From an area to the northeast of Salt Mine Valley, around the fringes
of De Vance's Pond, Gagliano (1967:96) reports the surface findings of San

232

Patrice, Lerma, and "Lanceolate-type" projectile points.

This would seem

to indicate the pond area as a favored stopping spot during the Late Paleo-

Indian Period.

Additional evidence comes from Pit VI and the surface area to its northeast (Analysis Unit 2) where prepared cores and bladelets have been found
(Gagliano,

1967:51-52)

(Figure 6-23).

These artifacts have been classified as

cores with prepared striking platforms, cores with natural striking platforms,

blocky cores, cortex flakes, flakes with multiple scars, flake and scrapers,
flake side scrapers, and bifacial artifacts.

During the 1969 excavations, several pits were excavated into the ridge
deposits of Salt Mine Valley (see Figures 6-13 and 6-14).
lieved to have been formed during Phase

These ridges are be-

of the proposed valley sequence (ca.

11,000 - 8,000 B.P.) and are composed of stream-deposited sands and gravels.
In these ridge deposits were located various artifacts conforming to those

described previously for prepared cores and bladelets.

Although no true projectile points were located in stratigraphic context, the similarity of the prepared core and bladelet artifacts

to Paleo-

Indian tool assemblages from elsewhere warrants the inclusion of this com-

ponent in that time frame.

The projectile points from around De Vance's Pond

help support the belief that Paleo-Indians did inhabit Avery Island, and most
likely Salt Mine Valley, conceivably while following a seasonally based,

migratory, hunting and gathering pattern.

Aside from Avery Island, surface collections from two

Late Paleo-

Indian - Early Archaic sites serve to illustrate projectile point types and
other characteristics of the period in the central Gulf area.

The Bayou

Grand Louis Site (16 EV 4) is associated with a relict, crevasse distributary channel on the Prairie Terrace surface of south-central Louisiana.

The

old crevasse channel, which once led into an extensive backswamp area, can
be identified only by relict channel scars and poorly defined natural levee

233

Figure 6-2 3.

Cores, bladelets, and bifacial tools from Salt Mine Valley

A, B, E, G, End Scrapers; C, D, I, Flake Side
(16 IB 23).
Scrapers; J, K, Cores; F. H, L, Multiple Scar Flakes; M,
Scraper or Knife; N, Drill or Awl (After Gagliano, 1967).
234

ridges.

The site is distinguished by artifacts scattered over several acres

in cultivated fields along the crest of an old natural levee.

The artifact assemblage (Figure 6-24) is distinguished by projectile

point forms including Cumberland, Angostora, Merserve, San Patrice, Palmer,
Ellis, Gary, and Williams,

Hammerstones, pitted anvil stones, unifacial

scrapers, and bladelet tools also occur.

bles are also found.

Debitage and some unworked peb-

There is no apparent local source of stone from which

the artifacts were manufactured.

The Prairie Formation with which the site is

associated consists of alluvial deposits of silts and clays.

While the arti-

facts are thinly scattered, they are distinctive because they represent the

only material coarser than silt that is present.

It is a relatively common

situation in coastal and alluvial sites to find that stone artifacts represent
the coarsest sediment or rock particles present, representing coarse inclusions

well beyond the size range of the natural sediments present.

The Palmer Site (16 EBR 13) exhibits some similarities.

It,

too,

consists only of scattered surface finds on an old alluvial terrace, although the paleogeography is somewhat different.

The Palmer Site is situ-

ated on the margin of a poorly defined scarp marking the margin of the

Amite River valley in southeastern Louisiana.

When the site was occupied,

the reaches of the river in its vicinity are believed to have been estuaries.

The projectile point assemblage from the Palmer Site (Figure 6-25)
is distinguished by San Patrice points, Kirk Serrated points,

and broad, simple-stemmed forms.

Plainview,

Unifacial scrapers and use-polished, gouge-

like tools (probably digging stick or "hoe" blades) are also present.

While Bayou Grand Louis and Palmer are typical of Late Paleo-Indian Early Archaic sites, and similar sites are likely to occur on the continental
shelf, their discovery will be extremely difficult.

The artifacts are

limited in number and scattered over a relatively large area.

235

The surface

Figure 6-24.

Projectile points from Bayou Grand Louis

A, Cumberland; B, Angostura;
(16 EV 4).
C, Meserve; D, Palmer; E, San Patrice;
F-L, Unclassified.

Figure 6-25.

Artifacts from the Palmer Site (16 EBR 13).

A-C, Unclassified broad-stemmed projectile
points; D-E, Kirk Serrated; F, Plainview;
G, San Patrice; H-I, Unifacial end scrapers;
J, "Digging" celt or adze; K, Circular biface with graver burr; L, Granite hammerstone; M, Ground hematite (plummet?);
N-Q, S, Unclassified; R, Kent.

236

expression of the relict features is subtle, and there does not appear to
be a distinctive geometry or internal structure.

Such characteristics are

too subtle for positive interpretation through present remote sensing tech-

niques.

They are likely to be found only through chance artifact recovery

in a bottom dredge or some other fortuitous discovery.

Central Gulf Archaic

One of the least known periods in the prehistory of the northern Gulf
area is the so-called Middle Archaic.

The paucity of specific site data

makes this interval difficult to discuss except in very speculative terms.

Centain assemblages of projectile points and stone artifact types have

been tentatively assigned to the Middle Archaic, but this is largely through
inference and comparison with sequences established in other areas.

For

example, mixed projectile point assemblages that include Kirk Serrated, Eva,

Palmer, Morrow Mountain, Tortugas, Williams, and Marshall might be assigned

to an Early-to-Middle Archaic interval.

In Louisiana, at least part of the Amite River Phase may be assigned

to the Middle Archaic.

The phase has been defined entirely on surface col-

lections from what are in all probability multi-component Archaic sites

along the middle Amite River.

Amite River Phase sites most typically are found on riverine terrace
surfaces in close proximity to scarps marking the alluvial valley margin
of the river (Figures 6-26 and 6-27).

They are located along those reaches

of the stream where chert gravel is a common component of the bed load.

Sites are identified by a profusion of debitage and chipped tools.

At the Williams Gravel Pit (Figures 6-26, 6-27, and 6-28), a chipped
adz and a large sandstone mortar were recovered from sand and gravel deposits,

indicating active use of the bars by Archaic peoples.

237

Figure 6-26.

Site distribution and morphological relationship in the

middle Amite River area. An "*" signifies a site not shown
(After Gagliano, 1963).
of Vol. Ill
on Plate 4

238

c
o

100

ra

>

W;M

Undifferentiated Recent deposits

Prairie Terrace deposits

Pre-Prairie Pleistocene deposits

Figure 6-27.

Generalized cross-section of the Amite River Valley,

showing site-terrace relationships. Location shown
(After Gagliano, 1963).
in Figure 6-26

Sandy

clay with

pea gravel.
_o

Cross-bedded sand and

gravel. Inclusions of

manganese coated
gravel throughout.

Manganese coated sand

Q.
a)

and gravel.

6-

Manganese coated adz

Figure 6-28.

Stratigraphy of high floodplain exposed at Williams

Gravel Pit (16 EBR "A"), location shown in Figure
6-26
(After Gagliano, 1963).

239

The geologic context of the Williams Gravel Pit finds and the morpho-

logic relationships of sites in the area are significant in the present

study.

The overall setting is similar to the Mission River drainage in

Bee and Goliad Counties, Texas, as described by Sellards (1940) and dis-

cussed previously.

Within what might be called the present floodplain of the river,

there are actually two levels which indicate surfaces of stream deposition.

The upper level, or high

floodplain

is somewhat dissected, has a moder-

ately well-developed soil profile, and is covered with water only during
flood.

Several feet below this high floodplain is a level of active and

abandoned point bars.

At first inspection, it might appear that the upper

surface was formed during floodstage, and the sandy bars represent
stage

However, deposition by the river seldom exceeds the

features.

height of the highest river bar, and only clays are deposited above the

bar crests during flood.

In the segment of the Amite River under con-

sideration, the stream is still adjusting its gradient by downcutting.

Active deposition is presently restricted to sandbars and channel deposits in and below the low

level.

In their studies of river systems, Wolman and Leopold (1957:105).

have found that a channel has an associated

floodplain whether stable and

flowing on bedrock, gradually eroding a valley, or gradually depositing a

fill.

Furthermore, when the aggraded valley fill of a floodplain is

incised after its deposition, the original level becomes a terrace.

They define an alluvial terrace as "....an abandoned floodplain whose

surface no longer bears the normal relationship to the stream bed" (1957:105)
The true terrace, by

annual flood.

its

The upper

definition, must not be overtopped by the

floodplain of the middle Amite River area is

occasionally flooded, but does not "....bear the normal relationship to

240

the stream bed."

The upper surface, then, might be interpreted as a

poorly-defined terrace.

It is the deposits of this upper surface that

have yielded Archaic artifacts in the Williams Gravel Pit.

The slope relationships between the upper level and the Prairie

terrace are such that the two surfaces merge immediately north of
the area depicted

in Figure 6-26.

Like the Prairie, the high

floodplain

deposits are coarse and presently supply much of the sand and gravel needs
of Baton Rouge and adjacent communities.

Because of the very slight dif-

ferences in elevation between "high" and "low" surfaces, the deposits are

undifferentiated in Figures 6-26 and 6-27.

Gravel deposits, which are shown schematically in Figure 6-27,
are formed by concentration of the coarsest particles eroded from the older
deposits.

Gravel is abundant in channels and on point bars immediately

downstream of tributaries and in places where streams cut and rework older
concentrations.

The gravel consists predominantly of tan, brown, red, and

black chert of varying degrees of purity, with occasional quartzite pebbles

and small slabs of ferruginous sandstone.

Extensive site areas are found on the Prairie terrace along the scarp

separating Prairie and Recent

floodplain surfaces, i.e., the Baywood Site.

A few middens, such as the Doyle site, have been found on abandoned point
bar ridges within the active

floodplain.

All of these sites are associated

with gravel accumulations, the largest complexes being located on the

Prairie surface near the junctions of tributary streams and the river valley,

i.e., the Bluff Creek Site.

The occupants simply selected spots where

streams were actively eroding and

depositing

gravels, resulting in a

readily available supply of raw material for tools and weapons.

Creek, Doyle, and Baywood Sites

typify the

Amite River Phase.

241

The Bluff

settlement patterns of the

Recent work in the area has disclosed a series of sites along the

escarpment separating the Prairie and pre-Prairie Pleistocene uplands.

While the sites thus far found are characterized by chipped stone tools,
their age and relationship
race sequence remain

to the geologic history of the river and ter-

to be established.

Putting the geological and archeological sequences of the middle Amite

River into the terminology of the present study, it can be summarized as
follows:

FORMATION

CULTURE PERIOD

INTERVAL

PERIOD

J-K

Late Holocene

Low floodplain
and fill
High floodplain
(terrace) and
fill

Middle Archaic

Prairie Terrace

Possible associated
sites

Middle Holocene

Early to Middle
Pleistocene

Pre-Prairie
Pleistocene

In one sense, the middle Amite River sites can be considered as quarries

whose locations were dictated by the availability of stream gravel.

However,

the diversity of the artifact assemblage suggests a much broader range of

activities and has led to the conclusion that the principal sites were at
least seasonally occupied and that during these periods, the occupants were

242

engaged in a full range of economic and social activities largely unrelated

to the manufacture of stone tools

(Figure 6-29 and Figure 6-30).

Small, chipped, gouge-like tools often exhibiting polished bits are

now believed to have functioned as hoes or digging-stick tips and may have

been used in "root gathering" or "early agriculture

Distinctive pro-

jectile point forms are indicative of specialized hunting activity.

Bifacial knives and blades are related to these activities.

Another sug-

gestion of at least semi-sedentary activities is the occurrence of small,

low, circular earth mounds.

not been established.

The specific function of the mounds has

They are usually solitary, but 4ccur sometimes in

pairs.

In addition to Amite River sites, there are other suggestions of a

Middle Archaic age for the introduction of mound building in the central
Gulf coast area.

The most important of these is the Monte Sano Bayou Site

in southeastern Louisiana (16 EBR 17).

Here Haag, Ford, and Gagliano

(unpublished manuscript), in a salvage archeological project in 1967, es-

tablished that two large conical earth mounds located on an alluvial terrace at the valley escarpment of the Mississippi River were mounds con-

structed over cremation platforms

(Figure 6-31 and Figure 6-32).

A char-

coal sample from the cremation platform within one of the mounds yielded
a date of 6,220 B.P.

occur

140 years (4,240 B.C.).

at the site, including a stylized

Poverty Point traits

grasshopper bead of red jasper,

Poverty Point- type microf lints, and the similarity in construction techniques of the mounds with those at the Poverty Point Site. This casts some doubt
on the validity of this surprisingly early date for mound building.

How-

ever, another relatively early date from the Banana Bayou Mound (16 IB 24),
a small conical cremation mound on the Avery Island salt dome,

credibility to the date.

gives some

Charcoal from the cremation platform within the

Banana Bayou Mound was dated at 4,440 B.P. + 260 years (2,490 B.C.).

243

r^

^3 B

III ^
he

W-M

V
ll

>,

[
V
'

^J^L J
W

V
i*^l

H^HJ

Figure 6-29

Amite River Phase Projectile Points.

A-C, Almagre; D-F, unclassified; G,
Morhiss; H, Webb (?); I-K, 0-P, Kent;
L-N, Assymetrical; Q-R, Shurala; S-T,
Wells.

Illllil
Figure 6-30

Amite River Phase Artifacts. A-D,

Ovate and triangular blades; E-G, chipped celts; H, unclassified; I, "turtleback"; J-K, double-ended gouges; L-M,
gouges; N, "leaf-shaped" point or blade;
0, notched drill; P-Q, straight drills;
R, double-ended graver; S, oval knife;
T-U, "petal-shaped" flakes; V-W, lamellar flakes; X-Y, pointed hammerstones
244

Figure 6-31.

Map of Monte Sano Mounds (16 EBR 17).

The mounds
are situated on the Prairie Terrace surface near the
escarpment marking the eastern valley wall of the
active floodplain Q f the Mississippi River.
245

c
o
e

rl

a,
cu

43
4-1

o
S3

CD
4-1

H
(/)

O
c
C/2

cu
4-1

c
o
CD
4-1

<
(3

42

co
a)
co

3
O

iH

4-i

nj

S
M

C
CU

4-1

43

a u
cu

rj
CO
CO

4-1

O
H

X)

o
CN

CI
I

CU

3
oo
H

246

PI

Thus, our highly speculative model for the Middle Archaic of the

central Gulf coast area includes the trait of mound building.

Until dis-

proved, conical earth mounds, occurring singularly or in small groups,

should be considered as a possible element of the Middle Archaic.

We have indulged in some speculation regarding the Middle Archaic,

as inspection of the sea level curve (Vol.

3,

Plate

1)

will indicate, be-

cause drowned Middle Archaic sites can be anticipated on the continental

shelf.

In fact,

the postulated sea level reversal during interval H, may

have continued during the Middle Archaic, providing an explanation for the

paucity of coastal area sites related to the period.

The settlement and land-use pattern along the middle Amite River provides another model for predicting high probability areas associated with
relict alluvial features on the continental shelf.
As previously mentioned, there are currently few published

reports dealing with central Gulf coastal Archaic sites.

One of the ex-

ceptions, however, is the Copell Site (16 VM 102), located in an orange

grove on Pecan Island, Louisiana (Figure 6-33).

This site was first brought

to the attention of archeologists through two short reports compiled by

Henry B. Collins (Collins, 1927; 1941).

Later, Ford and Quimby (1945)

gave considerable review to the site and its artifactual assemblage in

their report on the Tchefuncte Culture.

In this report, Ford and Quimby

placed the Copell Site in the same cultural context as Tchefuncte.

They did recognize, though, that the lack of pottery at Copell probably
meant the site was earlier than the typical Tchefuncte Culture sites.

In

1947, Martin, Quimby, and Collier placed the Copell Site in the Archaic

Period.

However, in later studies (Mclntire, 1958; Haag, 1961), Copell was

still considered as Tchefuncte, although some doubt was expressed.

247

It

is

9220'

**** BcocK
P7

'

-.

"" -.
\

ridge

(VVVVVJ
JvXnS&J

CKenier

Shoreline trend indicoled

on airphoto mosaic*

Vp

Moish

Prairie

Terrace

Lake
Gulf-bottom contour

in

feet

6
'

'l

MIIES

'

Figure 6-33.

Uplands

Km

The Copell Site (16 VM 102) on Pecan Island, Louisiana.

Pecan Island is one of a series of relict beach ridges,
or cheniers, along the Louisiana coast.

248

now generally assumed that the Copell Site is not a Tchefuncte Period site,
but rather a Late Archaic Period occupation.
1967; Gagliano et al

Two recent reports (Gagliano,

1975) have suggested that the name Copell be applied

to a Late Archaic phase for the area of southwest Louisiana.

The site itself is situated on Pecan Island, a relict Gulf beach ridge,

or chenier, rising out of the Louisiana marshes.

Such cheniers offered the

Indians stable, high ground throughout the centuries

and can be considered

as prime locations for the discovery of Archaic and Paleo-Indian sites in

the continually subsiding Louisiana marsh area.

Indeed, scattered Paleo-

Indian points have been reported from Pecan Island.

Since we are here concerned with the Late Archaic Copell Site, a trait
list may be in order as a summary of the artifacts collected by Collins and

analyzed by Ford and Quimby.

stemmed, ovate- triangular

These artifacts are

projectile points, knives, retouched blades, a boatstone, a bar weight,

eted

sock-

points manufactured from the bones of a variety of mammals and birds,

an atlatl hook, bone perforators, cut and drilled bones, and various un-

worked animal bones (Ford and Quimby, 1945:17).

associated with the Copell burials.

All of these objects were

Indeed the site is unique in that it

was a burial Site rather than a habitation area. The burials themselves were
sometimes laid upon layers of red and yellow pigment and were placed in the
ground in tightly flexed positions.

They probably also occurred in pits,

although Collins could not discern their outlines in the excavations (Ford
and Quimby,

1945:16).

There were 20 males and 13 females, with the mean

age at time of death being 55 years for the males and 46.2 years for the

females (Snow, 1945:Table XVII)

buried at the site.

The only description of the depth or areal extent of the site is that
the burials were unearthed from several feet below the ground surface.

249

One other interesting item associated with the burials was asphaltum,
a commonly occurring item along the Texas coast.

This may possibly il-

lustrate ties to the west with the Aransas Phase and other Archaic peoples
of the Texas coast.

Another Late Archaic Period site in the central Gulf area is the Cedarland Plantation Site (22 HC 30)

located near the mouth of the Pearl

River in Hancock County, Mississippi

(Figure 6-34).

The site, located in 1957, is a semi-circular, stratified, oyster

shell and earth midden (Figure 6-35)

Pearl River estuary (Gagliano, 1963;

situated on a low bluff overlooking the

Gagliano and Webb,

Immedi-

1970).

ately across a swampy depression to the south-southeast of the site is

another Indian habitation locale, the Poverty Point-age Claiborne Site

(22 HC 35),

which is discussed in the following section.

The Cedarland midden material, as illustrated in the cross-section in

Figure 6-35, is basically composed of two strata.

The lower stratum is made

up "predominantly of oyster shell (Crassostrea virginica ) with bones of small

animals, deer, bear, fish and waterfowl, and with charcoal and artifacts

intermixed (Figure 6-36).

Remnants of small, clay-lined hearths (Figure 6-37)

are scattered throughout the shell" (Gagliano and Webb, 1970:47).

The

upper stratum is a black, organic sand containing the remains of animals,

charcoal, artifacts, and clay lumps.
was also noticed in this upper

An increase in the numbers of hearths

stratum.

The artifacts recovered at Cedarland

(Figure 6-38) became the basis for

defining the Late Archaic Pearl River Phase (Gagliano, 1963).

Most commonly

occurring artifacts and raw materials were marine shells, terrace and stream
gravel, ferruginous sandstone, red jasper pebbles, limonite, orthoquartzite

from southern Alabama, pink and white meta-quartzite from Arkansas, sandstone

250

>
H

06

_l

ll

<

H
tfl

CH
r*

o
H
4-'

Rj

i
QJ

p-i

cc
CJ

H
60

O
O

H
,3

C
M

*>

o>
.H

6
~3

o
C

cd

CO

CO

i-i

o
H

30
0

4-1

O u

^3
H

e
o
u
4-

ui

10

T3

H <U
T3 H
U-l

o
4-1

H
00

H
-a

N
,

*
ro
kO
0)

u
3

M
H
U,

251

o
M

CU

T3
H3
H

>,

"

-O
-O
01

CU

00 rH

C
H

0
CU

0)

.>

TJ

>
o
c

CO

CO
<u

en
to

<o
i-l

oo

CO
cu

4.J

co

cC

m
co

m-i

,o

<d
^

cj>

to

0)

>h
CO
CJ

CN

O
rl id

CU

C
O C
JO

CO

u-<

>, rH
iH J= iH
CX CJ
CO

CO

5-i

CO

00
H

en

CO

to

cu

5-1

tc

4J

M
CN

'.-:

Jt

CD

rH

^-^

O
4J

CU

-H iH

o
M
CX

CO

4-1

4J

CO
QJ

TJ

4J

C
CO
r-1

CO

Oh
CO

-a

(0

C
o

cO

4->

CJ
0)

cO

cu

H
m
I

CU

u
00
H

252

cu
cj

cr
cr.

Figure 6-36.

St rati graphic view of Cedarland Site.

Lower
portion of site is an oyster shell midden,
upper portion is black earth midden.
Date:
12/27/65

Figure 6-37.

Cross-sectional exposure of clay-lined hearth

in black earth midden in Late Archaic Cedarland
Site (22 HC 30), Hancock Co., Mississippi.
Date:
12/27/65

253

Figure 6-38. Artifacts from the Cedarland Site. A -- Hale Point;

D - Pontchartrain Point; E B - Macon Point; C- Macon Point
Point
G - drill; H,I - bipolar
Gary Medium Point; F- Ellis
lints;
microf
Busicon shell scraper or gouge;
L
cores; J,K
- gorget; P sandstone saw;
M
nutting stone or anvil; N
R - sandstone,
weight;
winged
quartzite
atlatl
quartzite bead; Q
Graveyard
plummets.
from
(M is
cylindrical atlatl weight; S,T
Site, 16 ST A) (After Gagliano and '"ebb
1970).
;

254

hones and saws, hammers tones, deer ulna awls, shell scrapers

perversum ) and choppers (Mercenaria mercenarla campechiensls )

,

Busycon

Winged

atlatl weights, red jasper beads, and straight drills were also common.
The most common point types consisted of Gary typical, Pontchartrain,

Macon, Hale, and Palmillas, with Ken, Yarbrough, and Gary large and small

appearing in lesser numbers.

Another interesting characteristic was the occurrence of bipolar cores.

Bipolar cores are also reported from Archaic sites in Texas

(Honea, 1965)

suggesting that this technique, identified with Paleo-Indian levels of

Avery Island, may have persisted for long periods or have been revived for
special purposes.

The burin-like forms on bipolar cores of the Avery Island

assemblage have not been identified at Cedarland.

A single radiocarbon assay from the upper level of the site gave a date
of 3,190 B.P. - 130 years

(Gagliano and Webb, 1970:69).

sumed, however, that the site was occupied before that.

It can be as-

A relationship be-

tween Cedarland and other Pearl River Phase sites has been noted by Gagliano
and Webb, especially for the Cedar Point (16 ST "A") and Graveyard (16 ST 4)
sites.

The transition from the Archaic occupation at Cedarland to the Poverty

Point at the neighboring Claiborne

Site will be considered in the following

section.

Central Gulf Poverty Point

Recognition of the Poverty Point Culture, one of the most interesting

phenomena of North American archeology, can be traced to an article by Clarence H. Webb (1944) which describes certain traits from the type site in West

Carroll Parish, Louisiana.

The real importance of Poverty Point Culture un-

folded as a result of the excavations at the Jaketown Site in Mississippi

(Ford, Phillips, and Haag,

(Ford and Webb,

19 56)

1955),

followed by those at the Poverty Point Site

255

Although the Poverty Point culture had been identified at the Jaketown,
Poverty Point, and other sites in the Mississippi Alluvial Valley, as late
as 1963, the earliest established component in the Louisiana coastal area

was Tchefuncte.

Various researchers had postulated pre-Tchefuncte occupations,

but had no substantial proof upon which to base this supposition.

Ford

and Quimby (1945) reported Poverty Point Objects from the Tchefuncte Site
(16 ST 1),

the Little Woods Middens

(16 OR 1-5), and Big Oak Island

all basically Tchefuncte Period (500 B.C. - 100 B.C.) sites.

(16 OR 6),

They attributed

the presence of the objects to their being hold-overs incorporated in the

Tchefuncte cultural milieu.

Earlier, C. B. Moore, in

his early reconnaissance surveys of the lower

valley's archeological remains, reported Poverty Point Objects from the Bayou
Sorrel Site (16 IV 4), the Schwing Site (16 IV 13), and the Miller Site
(16 SM 6)

(Moore, 1913:12-16).

In a 1963

article in American Antiquity

Gagliano and Saucier reported

three new Poverty Point sites, two of which, Bayou Jasmine (16 SJB 2) and

Linsley (16 OR 40), had been revealed by dredging operations.

The third

site, Garcia (16 OR 34), had been recorded along a heavily wave-washed beach.
In that same year, the Pearl River Phase of the Late Archaic Period

ported (Gagliano, 1963).

was re-

Situated across a swampy area from the Cedarland

Site, the type site for the Pearl River Phase, was the Claiborne Site
(22 HC 35), a Poverty Point Period locale.

In this section, we will first

discuss the Claiborne Site and then Bayou Jasmine, Linsley, and Garcia.
The Claiborne Site (Figure 6-35) is located only about 45 mi from Cedarland,

The site itself is almost identical in shape and orientation to Cedarland,

containing a semi-circular midden deposit of black earth and shells.

In

contrast to Cedarland, however, the dominant shellfish forming the midden

was the clam Rangia cuneata.

This suggests a decline in the salinity of

the water in the Pearl River's estuary from Late Archaic to Poverty Point
times.

256

Poverty Point Objects were exceptionally abundant (Figure 6-39).

These clay balls are believed to have been used as cooking stones in

roasting pits or boiling stones in skin-lined and water-filled boiling

holes.

The most common type of Poverty Point Object was the melon-shaped

variety, followed in frequency by biconical types (Gagliano and Webb,

This statistic, plus the counts on the other types of objects

1970:52).

of Claiborne, shows a marked similarity to the percentages of various types

of objects from the Poverty Point Site (16 WC 5) itself and the Linsley Site
(16 OR 40)

to be discussed below.

Only minor differences in size and color

and amount of sand present, show up among objects from the sites (Gagliano
and Webb, 1970:52-57).

Virtually the entire Poverty Point assemblage is represented, leading

to the conclusion that Claiborne was a regional center participating fully

in the Poverty Point trade network and cultural organization (Gagliano and

Webb, 1970:57).

A small conical earth mound was also associated with the Claiborne
Site,

323 meters east of the midden, but was totally destroyed before any

data could be obtained from it.

Other artifacts and ecofacts described as being abundant at Claiborne

are animal bone, terrace and stream gravel, and Pont chartrain- type pro-

jectile points.

Gary large, Gary typical, Macon, Hale, Kent, Shumla,

Williams, Morhiss, and Carrollton are the other common points at Claiborne.
The typical Poverty Point-type microf lints, marine shells, red jasper, fer-

ruginous sandstone, limonite, steatite, orthoquartzite, crystal quartz,

magnetite and hematite, sandstone hones and saws, hammerstones

bone points and antler tine points, antler

f lakers,

socketed

antler scrapers or

wedges, bone pins (Figure 6-40), steatite vessels and sherds, fiber-tempered
pottery, untempered pottery, two- or three-hole gorgets and sandstone

257

Figure 6-39. Poverty Point Objects from the Claiborne

Site (22 HC 35), Hancock County, Mississippi. A - B, "Mulberry roughened" spheroid (perforated) C, Spheroidal perforated; D, Cylindrical, horizontal grooved; E, Cross-grooved;
F, Cylindrical, horizontal grooved (perforated); G, Cylindrical, spiral grooved; H, Cylindrical; I, Melon-shaped grooved;
J, Melon-shaped; K, Biscuit-shaped; L, Melon-shaped, spiral
grooved.
;

Figure 6-40.
Selected artifacts from the Claiborne Sice.
Perforated, shell gorget; BandD, Bone hairpins; C,
A
E, and F, Bone projectile points and fragments; G, Drilled
canine tooth; H, Bone awls or pins; I, Bone points and
,

f lakers.

258

gorgets, grooved and perforated plummets, tubular beads, and three female

figurines round out the majority of artifacts of the typical Poverty Point

assemblage at Claiborne (Gagliano and Webb, 1970: 66-69).

Of particular interest is the fact that Claiborne has produced more

fiber-tempered pottery than any other site in the central Gulf region.

Approximately 200 sherds of such kind were found.

All were classified as

either Wheeler Plain or Wheeler Punctated (Figure 6-41)

Steatite sherds and a cache of ten complete steatite vessels

(Figure 6-42)

were found in sterile soil below the midden.

In summation, Claiborne can be viewed as a site of the Poverty Point

Period, having been occupied, most likely, by the same population type that

had lived at Cedarland during Late Archaic times.

in plan and face the estuary in such a like manner

ceive of another hypothesis.

Both sites are so similar

that it is hard to con-

The population shifted from Cedarland to

Claiborne, for whatever reasons, with the advent of the incorporation of

the Poverty Point cultural milieu.

Further evidence of this shift in locales is seen in the radiocarbon

determinations.

A charcoal sample from the upper level of the Cedarland

midden yielded a date of 3,190 B.P., while a charcoal from the base of the
Claiborne midden was determined to be 3,100 years old (Gagliano and Webb,
1970).

Thus,

the shift occurred approximately 3,150 years ago.

Claiborne was most likely the major trading or ceremonial center of the
area during Poverty Point times, while lesser satellite villages or special

activity sites were situated in the marsh and swamp of the surrounding
coastal region.

The Garcia, Bayou Jasmine, and Linsley

Site were just

such locales.

Based on bore-holes and radiocarbon dates (Figure 6-43) and the artifacts recovered, Gagliano and Saucier (1963:326) established the Bayou

259

10
I

Figure 6-41.

'

'

cm

Base of a Wheeler Punctated, fiber-tempered

vessel from Claiborne.

.i.u. m.ijnm.i ii iijunni i .uj..n

<

<

iu.j li

.ijtnn.mi lliMn ni

Figure 6-42.

'

jun4 mm

im.u

Figure of the five steatite bowls found as a group,

A and B show the gouge marks
at the Claiborne Site.
left from shaping of the vessels, while C, D, and E
Photos courtesy of
have been ground to smoothness.
Mr. Norvell Roberts, Winchester, Ohio.
,

260

Jasmine Phase of the Poverty Point Period for the Linsley and Bayou Jasmine
Similarly, the Garcia Phase was established for the Garcia Site

Site.

(location shown in Figure 6-34).

Basically, the phases are distinguished by age.

The older, Bayou

Jasmine Phase is characterized by larger amounts of typical Poverty Point

Objects (Figure 6-44), Rangia cuneata shell middens, abundant faunal remains,
bone artifacts, and three radiocarbon dates averaging 3,690 B.P.

At the

Linsley Site, Poverty Point Objects were even encountered clustered in fire
pits, brought up whole in the dragline bucket spoil.

It may be noted that

these spoil loads were taken from depths of approximately 2-7 meters below

mean gulf level, and from what is believed to be the top of an old buried
Mississippi River natural levee (Gagliano, 1963:16).
The Bayou Jasmine Site spoil was though to have come from a similar

ridge at a depth of 1.8-2.3 meters.

Recent excavations by Louisiana State

University have revealed, however, that midden material and peat deposits
extend down to at least 5.5 meters below the surrounding marsh and swamp.
The Garcia Phase is assumed to be the younger of the two phases based

on the similarity between the microlithic assemblages

and the famous Poverty Point and Jaketown Sites.

(Figure 6-45) at Garcia

Radiocarbon dates are

lacking from Garcia.

The Garcia Site itself is described (Gagliano and Saucier, 1963:326)
as

"

victim of rapid shoreline retreat which consists of a heavily

wave-washed beach accumulation plus a possibly undisturbed shell deposit

to 4 feet of water up to 1,000 feet from the shore."

Hints of

lying in

earlier

Paleo-Indian and Archaic Period occupations at the site are re-

vealed by Dalton,

Meserve, Lanceolate, and Archaic- type points.

The stone

assemblage, including the microf lints, reveals a trading network with the

north and east, since much of the lithic material came from northern

261

6<Jt>

<>

00 CD 00

<B
140

120 130

+ +

UJ
_1

2490+

1740

o
<r

(A

1590+120

1890

a.
t* ei

Pi

<
(A

msmm
II

>-

w
S

lfc5

i'^*

^ Oxr
S s> 3

w\
>\

aj

<z5
Z<
UJ

UGC

0.

!^

u
CO

0J
.
-

ec
u.

co

a
>

UJ
a

H
_
:

co

*m

C
.-J

UJ

+1

o
0)
o
CM

/>

11

o
0.

a>

lAl

ro

_J

V)
CM

i^^^^^SE

</>
h.
-

oc

o
>o

co
CO
CO
co

QJ

4-1

c
H

O
Pm
>>
4-t
l-i

OJ

O >
H o

o>
ao

T3 Pm
CO
J-i

aj

JZ

X)

4-1

_)

CI
CO

<4-l

n
>,

*-'*

f
^

>> PM

_J

o
^

O
>-l

CO

V)

a>

rd
CJ

m
o

T3
<H

0)

>

4-1

C
o
rO
M

i-

?z.

CO

c
o

a.

^~.

0)

a)

J3

o>

(A

C
o

4-1

H
CO

i0

% o

Ac-

OJ
CO
CO

cO

X.

>-i

Ph

Ml
H

Z o

4-1

CO

IO

4-1

CO

CD

1
CO
cO
-)

dJ
T-i

3
o O

J3

CO

!^

0J

CO

>-(

CQ

O
XI
*

c
o
H

OJ

4-1

IW

4-1

cO

4J

1-3

CO

ro
~d-

vO
QJ

M
3

M
H
Pn

262

OJ

CJ

Bayou Jasmine Phase artifacts from the

Figure 6-44.
Linsley Site (16 OR 40) A - net impressions on baked
clay; B-I - Poverty Point Objects; J - deer ulna awl;
K - bone net spacer (?); L - bone pin; M - sandstone
hone; N,0 - unclassified projectile points; P-R socketed bone points.
.

T|J

~\

T 1

F'

Figure 6-45. Microliths and other artifacts from the

Garcia Site (16 OR 34). A - "Poverty Point-type" microliths; B - quartz fragment; C,D - quartz crystals;
E - expanded-base drill; F - celt fragment; G,H hammerstones; I - sandstone disk saw; J - schist slab;
K-M - grooved plummets; N - sandstone hone.

263

Mississippi, Arkansas, Missouri, southern Alabama, and the Appalachian and

Piedmont regions of the southeastern United States.
The area in the vicinity of the mouth of Pearl River

(Figure 6-34)

provides another important model of settlement and environmental use for

prehistoric peoples whose economy was based primarily on hunting-fishinggathering.

The area offered a wide variety of ecological conditions, in-

cluding river bottom swamplands, pine hills, coastal beaches, and brackish
and salt marshes, all of which converge in the vicinity.

Each of these

environments supported a characteristic fauna and flora? and each offered

a unique situation for hunting and gathering.

The river itself provided

a natural avenue of travel giving access to upland products and raw mater-

In view of these many attractions, it is not surprising that the

ials.

area reveals a long occupational history

pattern that is repeated at most

river mouths throughout the study area.

As illustrated by the number of sites shown on the map (Figure 6-34) the

ceramic history is well known.

Ceramic materials ranging in age from Plaquemine

to Tchefuncte are represented, and there are good correlations between human

occupations and geomorphic development.

For example, the Tchefuncte Period

is the earliest occupation found on the beach-ridge

of the river.

times.

complex at the mouth

This suggests development prior to, or during, Tchefuncte

The three sites shown near the upper margin of the map represent

the northernmost Rangia shell middens on

Pearl River and probably ap-

proximate the maximum extent of brackish-water encroachment during their

occupation.
In contrast to the many Rangia shell middens, the Cedarland Planta-

tion and Graveyard

(

Sites are composed almost entirely of oyster shells

Crassostrea virginica )

During the Pearl River Phase, when both were in-

habited, the mouth of the Pearl was probably more estuarine and local salinities

were

somewhat higher, permitting oyster growth.

264

Conditions changed,

and by Poverty Point times,

when the Claiborne Site was occupied, the

brackish water clam Rangia had replaced the oyster as the dominant shell
type in the midden, indicating that the estuary was being filled by a

growing delta in the vicinity of the site and that the saline zone of the estuary had shifted seaward.

It is also interesting to note that in the

Poverty Point midden, shell is generally less abundant, and the midden
consists largely of black earth with very abundant remains of mammals,
birds, and fish.

In the middens of the Tchefuncte and subsequent per-

iods, brackish water Rangia shells predominate, making up the bulk of the

middens.

Other physical changes in the area have resulted from recent fault

movement.

The Prairie Terrace, marking the western margin of the stream

valley, has been downfaulted, and the Graveyard midden, once on the valley
margin, is now partially below sea level.

The Garcia beach deposit, as mentioned previously, contains Poverty

Point elements and is believed to be related to an early Mississippi

River Delta development.

village at Claiborne.

The site was probably a satellite to the major

The earliest evidence of occupation thus far found

in the area comes from the innermost of the abandoned beach ridges

Ansley, Mississippi) and from the Garcia beach deposit.

ridge,

(at

From the beach

a single side-split quartz pebble with a graver spur produced by

steep-edge chipping has been found.

At the Garcia

Site

the evidence

consists of an unf luted Clovis Point and a number of Dalton points.

The McKeen Site (16 ST 52)

a camp

located on a knoll overlooking the

valley scarp, is unlike any other in the vicinity of the river mouth.

The as-

semblage is typically Early Archaic, composed exclusively of chipped artifacts, and includes a Kirk point.

This Early Archaic occupation is be-

lieved to have been contemporaneous with the development of a series of

large Deweyville stream scars impinging on the valley wall.

265

Immediately

north of the area shown on the map, these scars emerge from beneath a
wedge of alluvium to form part of the Deweyville Terrace complex.
By studying the settlement pattern and refuse of habitation sites
in the Pearl River mouth area, it is possible to reconstruct a significant

part of the Holocene geological history and ecological succession.

The

following sequence of events is suggested.

CULTURAL
PHASE & PERIOD

EVENTS

INTERVAL

PERIOD

Post
Tchefuncte

Development of coastal marshes

continued.

Late Holocene

Pontchartrain
Phase
Tchefuncte
Period

The lower estuary of the Pearl

was essentially filled and accretion beaches near the mouth
of the river were fully developed.

Late Holocene

Bayou Jasmine
Garcia Phase
Poverty Point
Period

The Pearl River delta advanced

within the embayed valley.
Mississippi River distributaries approached the area
from the west modifying environments from Gulf Sound to
marginal deltaic.

Late Holocene

Pearl River
Phase Late Archaic
Period

The shoreline was re-established along the innermost

of the accretion beach
ridges and the river valley
was embayed as a result of
rising sea level.
Sea level
reached its present stand.

Late Holocene

Early Archaic
Period
(McKeen Site)

Deweyville streams were active and ecological conditions

somewhat different.
The McKeen site was a number of
miles upstream from the existing coast; sea level was approximately 20 meters below
its present position.

Middle Holocene

H4-I

Paleo-Indian
Period

Late Paleo-Indian occupation

on shoreline or near-shoreline
features at Garcia site and
innermost of abandoned beach
ridges.

Early Holocene

H3

266

Eastern Gulf Pre-Projectile Point

One Pre-projectile Point Period site, Skelly, will be de-

scribed from the eastern Gulf.

The Skelly Site is one of a number of

important quarry sites located in the vicinity of Dothan, in south Alabama.

The Dothan area is one of subdued karst topography with numerous sinkholes

and residual outcrops of more resistant chert and cherty limestone.

William

H.

Mr.

Emanuel, formerly of Dothan, and other local collectors re-

ported the sites and described them as containing a profusion of lithic

material.

Emanuel (personal communication, 1975) was impressed by several

features of the sites.

area.

First, archeological sites are numerous in the general

Examination of large surface collections from neighboring sites revealed

characteristic artifacts from all documented culture periods thus far recognized
in the region, ranging from Paleo-Indian to late Prehistoric.
of the material is heavily patinated.

Second, much

Third, Emanuel found and reported hand

axes very similar to Acheulian-type hand axes which he had collected and

studied from the Grand -Pressigny area of France.

The Skelly Site is situated on rolling hills around a small dry lake
bed.

While the total extent of the site is not known, it is so large that

it was brought to Emanuel's attention by an Army pilot who reported its

aerial appearance as a curious "speckled area."

to be lithic artifacts, patinated white,

and pastures (Figure 6-46).

pits on the hills.

bris.

The speckles turned out

that litter the cultivated fields

Important features of the site are old quarry

The margins of these pits are paved with workshop de-

Where exposed by a roadcut, several levels of chipping floors can be

seen to dip into the old pits and are interbedded with red lateritic sands
(Figure 6-47).

After visiting the site with Emanuel, the senior author initiated a preliminary survey of the area.

267

The survey was conducted by Mr.

Figure 6-46.

Figure 6-47.

Debitage in cultivated field at Skelly Site

near Dothan, Alabama.

Stratified chipping floors sloping into old

quarry pits at Skelly Site near Dothan, Alabama,

268

Tom Ryan, then a student at Louisiana State University, during the summer
of 1967.

Ryan located about 20 sites and examined a number of large col-

lections accumulated by members of a local amateur archeological society.

Characteristic artifacts from all periods ranging from Paleo- Indian to late
Prehistoric were identified in the surface collections; that is, evidence
of all documented culture periods thus far recognized in the region could

be found.

A number of sites were found where lithic material occurred

with associated index artifacts and assemblages from various periods of the
established sequence of the area.

Although surface collections suggest a Pre-projectile Point age for much

of the occupation (Figures 6-48 and 6-49)

because the site remains to be

systematically worked, surface-collected material has important implications for

The following types appear in the assemblage:

this study.

Ovate and trianguloid hand axes - frequently rhomboid in cross-section

and with tool edges on the margins produced by secondary chipping

Heavy choppers - crude percussion chipping

Bifacial discs

Plano-convex scrapers - (scraper planes)

Large ovate- to- trianguloid biface blades
Leva llo is- like cores and blades
Flakes and blades with denticulate and graver spurs - (one or more)

Sandstone hammers and abraders

Heavy polyhedral chert hammerstones

Sandstone anvils

Bipolar cores - (infrequent)

Burins
The possibilities for occurrence of similar sites on the continental

shelf or the eastern Gulf are reasonably good.

The chances of detection

of this type of site by remote-sensing and bottom-sampling techniques are

269

Figure 6-48.

Skelly Site Artifacts.

A, D, Ovate bifaces;
B-C, E, choppers or cleavers.

Figure 6-49.

Skelly Site Artifacts. A-C, Ovate to rectanguloid bifaces; D-E, Bifaces with denticulates and graver burr on margins;
F-G, Trianguloid blades (possibly projectile
points or knives).

270

also reasonably good because of the abundance and density of worked stone
and the extensive surface area.

Eastern Gulf Paleo Indian

The discovery of human fossil remains and extinct Pleistocene fossils
in an old stream bed near Vero, Florida, (Sellards, 1916) represents a rare,

albeit controversial, situation relative to Paleo-Indiar occupation in the

eastern Gulf area.

Geographically, the Vero Site lies beyond the boundaries

of the Gulf coast study area, but its coastal setting and significance warrant

inclusion.
The construction of a drainage canal between Vero and Gifford, Florida,
in 1913 revealed the presence of vertebrate fossils in a cut through an old

stream bed (Figure 6-50).

This cut displayed three zones of deposition, which

eventually were given temporal estimates on the basis of geological and

Figure 6-50.

Map of Vero area showing canal from

which human fossil remains have been
1" = 4000'.
1 - Pine lands;
found.
2
Pleistocene Beach; 3 - Stream
Human remains came from
valley.
this valley, west of the railroad
and public highway crossing (From
Sellards, 1917a).

271

'

biological data, and will be described subsequently.

Tbe valley of the stream

at the fossil locale is relatively limited, varying in width from 107 to

152 meters.

The stream system, which has been greatly modified by modern channeliation, passes through a pan-shaped depression lying west of a sequence of

Pleistocene dune deposits believed to be related to the Silver Bluff shoreline (see Figures 6-50 and 6-51)

The depression may have originally formed

as a shallow lagoon behind the Silver Bluff beach-dune complex, or it

may have resulted from the buildup of the dunes parallel to an old
In either event, the depression

escarpment against the Pamlico Terrace.

evolved into a perched freshwater dune lake, and it is in the deposits of

this dune lake (Figure 6-51, Bed 2) that fossil human bones have been found

with bones of extinct Pleistocene vertebrates.

Of the three strata revealed by the cut (Figure 6-51)

the lowest and

oldest deposit has been related in time to the Sangamon interglacial stage
(Weigel, 1962) and is basically composed of a marine marl.

Fossils from

e Extant faunal remains

d Pottery
<

Hardpan

8 X:jv';}}::]&<

c-Flint spalls
b-Extinct Pleistocene vertebrate remains

^m. ...
/*" A

a Human bones
''

''

''

'''''''''..

'

...

'

'

! ....'
:. '

'

.'

"-

-'
.

-.

'

..

.,.

'

ve

'

---

=4

"

*^

v&R^Sf^Z^
r

"::&>:#: b.: :'S-'-r8200 BP 960

VriiV-V:'
L >3
0,000 BP Y. '.\\'&l
> .- ^:vVv->:^v^v;':^v.-v:-" a :;:
5

'

<if

2.

'

'

-' .-'.

~
fet#iitf**jgyiijuft^fi
_V&&i^^
d s vjjjflft?
3550 BP fM C-Xvl^:*iS-SS^r
+3m msl ^ii^W-^^^c^M^^2
i
A p p ro x

,:.

:-:

'

:.

"

;V

;;.

j'J:~\

""''-!.

Anastasia marine marl

>r

']

_/

Bed
" -

J.

-.

<?

-v^

>37,000 BP

<i^

-.Y-

Figure 6-51.

"Jv

?'>

^V-^v* ^^/.^O-.^:

East-west cross-section through fossil deposits at Vero.

Section is diagrammatic and scales approximate.
(Modified
from Weigel, 1962.)
272

this deposit are exclusively invertebrate species characteristic of a marine

habitat.

Analysis of the molluscan faunal remains shows no major morpho-

logical difference from present-day species, indicating the persistence of

invertebrate forms through time.

The second stratum is composed of cross-bedded sand at the bottom.

The

appearance of a freshwater marl at the top of this layer represents a shift

to freshwater conditions.

All fossils of extinct vertebrates from the site

have come from Bed 2, supporting the assumption that it is a Pleistocene-aged

deposit.

The vertebrate fossils were located in close proximity to human

remains as well as flint spalls.

Analysis of the fossils shows a predominance

of freshwater species, most of which were warm-climate animals.

Of

the 122 vertebrate species found in Bed 2, four were fish, seven

were amphibians, 27 were reptiles, 37 were birds, and 47 were mammals.

Forty percent of the mammalian forms are now extinct, as are three percent
of the reptiles and amphibians and eight percent of the birds.

The 28

identifiable land and freshwater invertebrates found do not differ from

modern species.
A checklist of fossil vertebrates from the locale prepared by Weigel
(1962) contains the following extinct forms.

Geochelone sellardsi
Ciconia maltha
Ectopistes migratcrius
Paramylodon harlani
Megalonyx jef fersonii
Dasypus bellus
H olmesina septentrionalis
Synaptomys aust ralis
Hydrochoerus sp
Canis ayersi
Vulpis palmaria
Panthera augusta
Smilodon sp.
Mammut sp
Mammuthus sp.
Mylohyus sp.
Equus sp
cf. Tanulpoloma mirifica
Bison sp. (probably Bison latifrons)

Extinct land tortoise

Extinct stork
Passenger pigeon
Harlan's ground sloth
Jef fersonian ground sloth
Extinct armadillo
Extinct armadillo
Extinct bog lemming
Capybara
Extinct wolf
Extinct fox
Extinct jaguar
Saber-tooth cat
Mastodon
Mammoth
Peccary
Extinct horse
Extinct camel
Extinct bison

273

The first human bones discovered in Bed

were found 100 m west of the

bridge in the south bank of the canal beneath 45 - 60 cm of marl rock.

second discovery of human bones was made in June, 1916 (Sellards, 1917), and

included an astragalus, a cuneiform, part of an ilium, two phalanges, a limb

bone section, and other bone fragments.

A deer scapula was found at the

same level as, and between, the astragalus and the cuneiform in Zone 2.

All of the bones from this zone were randomly distributed throughout the zone
and were usually found in an imperfect state of preservation.

A strong case

has been made (Sellards, 1917) for the contemporaneity of the human and

Pleistocene fossils, based on their aforementioned similar manners of occurrence (scattered and poorly preserved), on their proximity within the
strata, on their location in and beneath undisturbed geological formations,
and on their equal degrees of mineralization.

The absence of any nearby

fossil-bearing deposits reinforces the assumption that the deposits are primary.
Bed 2 also contained a thin, sharp-edged flint, a small worked spall, two

small spalls, and part of a bone implement, which may possibly have been
engraved.

The sharpness of the spall's edges helps to negate a theory of

water travel to explain their appearance in the deposit, and since the closest
flint outcrop is 160 km northwest of the site, the original source has yet
to be determined.

An abrupt erosional non-conformity marks the transition from Bed

to

the uppermost stratum (Bed 3), which is composed largely of muck and loose

sand, averaging 60 cm in thickness

marl.

and frequently overridden by freshwater

Human skeletal remains, bone implements, pottery, chipped stone pro-

jectile points, and ornaments have been collected from this stratum.

The

vertebrates from this third zone represent extant forms only.

Studies of

Zone 3 fossil plants reveal a former abundance of hydrophytic

plants, with

the resulting interpretation of Bed

as having been a pond-marsh habitat

during its period of deposition.

274

Further indications of the age of the deposits and their aquatic associations during deposition are gleaned from radiocarbon dating and geological data, especially pertaining to Bed

2.

Weigel (1962:11) summarizes

some of the relationships as follows:

The difficulties involved in establishing correlations of Florida Pleistocene fossil deposits with stages
The age
of the Pleistocene were reviewed by Bader (1957).
of the Vero bone beds is relevant to these problems and to
Two of
the chronology of Florida Pleistocene shorelines.
the shorelines, the Pamlico and Silver Bluff, are closely
associated with the Vero beds. The Pamlico shoreline lies
about 25 miles west of Vero and its terrace surrounds the
The Silver Bluff Shoreline lies a few feet
bone beds.
east of the railway at Vero and shoreline features indicate a sea level about 10 feet higher than present (MacNeil,
The contact between beds 2 and 3 is somewhat less
1950)
than 10 feet above present sea level; thus the deposit area
was subject to tidal action of the Silver Bluff sea.
.

At approximately +3 m MSL, the Silver Bluff was the only shoreline to

approximate, or exceed, the level of the existing shoreline since the

Pamlico (Sangamon Interglacial)

Since the fossil bed (Bed

2)

seems to be in-

timately related to deposition of the Silver Bluff dune sands, the age of
this shoreline is of considerable importance to our problem.

There has long

been considerable divergence of opinion regarding the age of this feature,

which is very well defined along the south Atlantic coast.

One school of

thought regards the Silver Bluff as post-Wisconsin in age and relates it

to the hypsothermal interval (post-glacial maximum) which began, according
to Deevey and Flint

(1957), about 9,000 years ago.

However, the concept of

a hypsothermal high stand has been largely disregarded by more recent

workers.

A second interpretation might relate the Silver Bluff shoreline to a

Farmdalian (Mid-Wisconsin) sea level high stand of 25,000 to 30,000 years

ago.

As we have discussed in Chapters

and 4 of this report, the evidence

for the Farmdalian high stand is suspect.

Still a third interpretation, and the one that would probably be fa-

vored by most contemporary Quaternary geologists, would relate the Silver

275

Bluff to the Sangamon Interglacial (see Chapter 4).

Following this

correlation, the Silver Bluff shoreline might be anywhere from 90,000 to

55,000 years old.

While the age of the Silver Bluff remains to be established, radiocarbon dates shed considerable light on the age of the fossil-bearing bed
(Bed 2) at the Vero site.

As shown in Figure 6-51, three radiocarbon dates

were obtained on samples from Bed

2.

A charcoal sample taken from the very

bottom of the bed yielded a date of >30,000 years.

The date of 8,200 + 900

years was obtained from carbonaceous material collected from the lower onethird of the bed near the limits of the basin.

Weigel (1962:10) does not

believe that this sample represents the oldest part of Bed

portions

older

treme limits.

2,

since the

thin out at the edge of the basin and are absent at its exThe 3,550

top part of Bed 2.

+ 120 years date

is from charcoal collected from the

Weigel (1962:10) states, "This bed clearly appears to

represent a period of continuous deposition from something over 30,000 years

ago until about 3,500 years ago and embraces that period during which many

Pleistocene forms became extinct in Florida."

In view of the

nonconformity at the base of the bed and the consider-

able hiatus which it implies, the degree to which the >30, 000-year-old date
is representative of Bed 2 and its contents is somewhat questionable.

Per-

sistence of a dune lake for a period of 30,000 years, during which great

fluctuations of sea level and variations in climate are known to have occurred,
also raises questions concerning Weigel'

interpretation.

The two younger

dates seem to be more consistent with the faunal and archeological record
as we presently understand them.
It is significant to note that the fauna and flora
indicate

moderate climate during deposition of Bed

2

and

length.

2.

warm,

The nonconformity between Beds

represents a hiatus of unknown duration, but apparently not of great

The faunal and floral record of Bed 2 indicates Late Holocene condi-

tions not significantly different from those which are obtained at present.
276

From the published descriptions of the site, the following summary

interpretation can be made,*

Bed

Events

Interval
and Stage

Period

3*

Deposition of sand and muck with

bones of extant fauna, pottery
and other artifacts.
Conditions
similar to those at present.
Bo gy> ponded conditions.

Late
Holocene

Erosional
Nonconformity

Erosion implies reduction of

standing water level through
reduced ground water level, improved drainage of pond or reduced base level of pond drainage system.
Final extinction
of Pleistocene fauna.

Middle
Holocene

H4-J

2*

Development of perched dune lake,

Deposition of sands with freshwater vertebrates. Fauna and
flora indicate warm, moderate
conditions.
Extinct Pleistocene vertebrate fauna abundant.
Human bones and chipped stone
indicate presence of man.

Early
Holocene

H3
(possibly
H1-H3)

4*

Silver Bluff shoreline active,

creation of tidal lagoon (?).
Deposition of coastal dune deposits.
Creation of pan-shaped
depression in which Beds 2 and 3
were deposited.

Late
Pleistocene

Nonconformity

Major hiatus of tens of thousands of years.

Pleistocene
Wisconsin
Glacial

5-7*

Deposition of Pamlico sands.

Shoreline approximately 25 miles
to the west.
Sands probably deposited under shallow marine
conditions as strand plain.

Pleistocene

1*

Deposition of shallow marine

marl, Anastasia Formation.

Pleistocene
Sangamon
Interglacial

Numbers after Weigel

1962,

277

or

Early
Holocene

A
or

Pre-A

One of the most exciting and scientifically rich archeological sites

to be located in recent years is the Warm Mineral Springs Site (8 SO 19) in

Sarasota County, Florida.

(8 SO "A")

Both it and neighboring Little Salt Springs Site

have afforded the opportunity to study well-preserved organic

material in stratigraphic levels, employing underwater archeological

techniques.

The information gained, and yet to be gained from these sites,

should contribute substantially to our understanding of Late Quaternary

geology, climate, archeology, flora, fauna, and hydrology of Florida.

Another site, the Fish Creek Site (8 HI 105), in Hillsborough County,

does not have the archeological or scientific potential present at Warm Mineral
Springs, but it does offer the reader an example of the ordinary, as opposed
to the extraordinary Warm Mineral Springs, Paleo-Indian site situated along

Florida's bay margins.

We will begin our discussion at Warm Mineral Springs

and then turn our attention to the Fish Creek Site.

A short

paper appearing in American Antiquity (Royal and Clark,

1960) first brought attention to the Warm Mineral Springs sinkhole.

publication, Royal

In that

and Clark discussed and illustrated a remarkably well-

preserved human skull and brain, along with various artifacts obtained from
the sinkhole.

They also mentioned the now well-known ledge, at a depth of

approximately 13 m below the water's surface, from which the skull came, and
the three distinct stratigraphic zones of the ledge (Royal and Clark, 1960:
285-6).

A single radiocarbon date of 10,000 + 200 years B.P. was obtained

from a burned log taken from Zone

3,

the lowest on the ledge.

The skull was

thought to be somewhat younger since it came from the base of Zone 2, the

intermediate zone.

Royal and Clark did feel, though, that an Early Archaic

age for the skull was a strong possibility (Royal

278

and Clark, 1960:285).

It was not until 1973, however,

that substantial, stratigraphic

evidence emerged during the excavation of a small test pit on the "13
meter" ledge by Carl Clausen, then of the Florida Bureau of Historic
Sites and
Properties (Clausen, 1972).

In a more recent report, Clausen, Brooks, and

Wesolowsky (1975) present a detailed account of the test pit and the im-

plications of the data recovered from the excavation.

Warm Mineral Springs originated when a subterranean cavern in the Tampa

Limestone collapsed (Figure 6-52).

This caused the formation of the present

sinkhole, approximately 73 m in diameter and averaging about 38 m in depth

(Clausen, Brooks, and Wesolowsky, 1975:193).

At a depth of about 13 m, a small ledge rings most of the sinkhole

(Figure 6-53).

This ledge, ranging from

to 6 m in width, was formed as

softer rock and clay materials both above and below it eroded it away.

Along

the northern side of the spring, the ledge forms a relatively wide platform

upon which sediments and organic deposits have accumulated.

Here it was pos-

sible to carefully examine the ledge's three sedimentary zones and excavate
the test pit.

These zones, 0.9 to

m in thickness, are described by Clausen,

Brooks, and Wesolowsky (1975:96-7) as:

Zone 1 - An algal slime, the stratigraphically superior

It is composed
layer, ranges from 0.20 to 0.50 m in thickness.
of a soft, aqueous, black algal ooze with shells of the same
small snails still prevalent in the spring, Heleobops docima
and Pyrogophorus platyrhachis
The bones of alligator, tarpon,
and turtle are occasionally found in this layer.
Zone 2 - Calcitic mud, the middle layer, ranges in
thickness from 0.15 to 0.50 m.
It is predominantly a gray, unconsolidated calcitic silt. There is some pine bark, oak
Two distinct
leaves and other plant debris in this layer.
Snail
layers of wall tufa represent periods of spalling.
shells are common, especially Physa cubense Heleobops docima
Vertebrate remains are uncommon and
and Helisoma trivolvis
are those of frogs and mice.
Radiocarbon dates on charcoal
from the upper and lower portions of this zone are 8,520 + 400
years:
6570 B.C. (W-1243) and 8,600 + 400 years:
6650 B.C.
(W-1241).
Zone 3 - A leaf bed, the bottom deposit, varying in
thickness from 0.10 to 0.80 m.
This is the most variable of
all the strata, consisting of alternate bands of terrestrial
plant debris, predominantly leaves, twigs and small logs, seeds
and charcoal intercalated with calcitic mud layers that contain
.

279

4->

7t*

c
O

en
0)

h
4J
cO

ID

rH

CD

CO

M
c
H

ir

*
cn

o
o
M

pa

j_)

cn

0)

u
*'

'Cm-.';

c
<u

cn

If
cn

c
a)
cn

co
<-\

M
0)

to

4J

rH

M-l

c_>

<

s^

IW

cn

oo

3 C
O -H
H TJ

w H

H -H
cn

ja

00

c
Vi
0)

H T3

5 o
O B
Xi
cn

TJ

3
cO

cn

oo

3
U

cn

H rH

cO

C/2

0)

rH rH
CO

U
CU

A
^i

c c
H -H

cn

^^

m
MN

>-i

o>

CO -H rH
12 T3

c
3
O O

S-l

cn

iw

a
2
CO

CNJ

m
vD
01

u
3
oo
u->

280

>%

S
o

cn

rH

si

-0-

-1

-MSL- =- -

SURFCIAL sand;

-1

-3

-2

-4

-3

-5

-4

-6

-5

-7

-6

-8

-7

-9

-8

-K)

-9

-11

CHONE SANDS
LIMESTONE SANDY.
HARD, GRAY

/
LIMESTONE SANDY PHOSPHORITE.
SOFT.

GRAY

LIMESTONE SANDY, PHOSPHORITE

FLECKS, HARD, OFF.

SAND AND CLAY, CALCAREOUS,

GRAY

PHOSPHORITE,

-12

-11

-13

-12

-14

-13

-15

-U

-16

-15

-17

-16

-18

-17

-19

-18

-20

-19

-21

SAND AND

_=

CLAY,

CALCAREOUS,

PHOSPHORITE, SOFT, GRAY

LIMESTONE SANDY
CLAY, HARD, GRAY
-

SOME

EAN SEA
LEVEL

METERS BELOW
SPRING SURFACE

Figure 6-53.

WHITE

Generalized cross-section through the wall of Warm Mineral Springs.

Note the ledge at the approximate depth of 13 m, and the three zones
of sediment on the ledge.
The two pins shown standing vertically
out of the three zones mark the location of Clausen's test pit
(After Clausen, Brooks, and Wesolowsky, 1975).

281

fresh-water and terrestrial snail shells as well as fragments

of wall tufa. Fallen stalactites occur as well as fronds of
the sinkhole fern Thelypteris normal is encrusted with a heavy
calcitic layer. A small stalagmite has been found in situ
within this stratum. Terrestrial snails are more abundant and
diverse in this zone. Helisoma trivolvis and Physa cubense
Identified plant
are the most common fresh-water species.
remains represent the following species: Pinus elliotii ,
Sabal palmetto , Quercus virginiana, z. laurifolia , Ampelopsis
arborea , Carya sp. , Phylolacca rigida , and Thelypteris nor malis.
The most common polynomorph is a pyospore of a species
of Chi amy do monas or a related fresh-water algae.
Vertebrate
remains so far identified consist of man, deer, opossum, raccoon, rabbit, squirrel, mouse and frog.
Radiocarbon dates on
charcoal collected by Brooks from this zone are 9,370 + 400
7420 B.C. (W-1245) , 9,500 + 400 years:
years:
7550 B.C.
(W-1212) and 9,870 + 370 years:
7920 B.C. (W-1153), from top
to bottom respecively.

In the area of the test pit, only Zone 3 remained completely intact,

however, since relict-hunters had removed most of the first two zones earlier.

Three radiocarbon samples were taken from this zone in order to pin down its
time of deposition more accurately.

The Zone

deposit was 70 cm thick at

this point, and the radiocarbon dates, in descending order in the zone, were

8,920 + 190 years, 9,350 + 190 years, and 9,220 + 180 years before present,

respectively.
3

The bottom radiocarbon date came from the 70 cm level of Zone

(Clausen, Brooks, and Wesolowsky, 1975: 198).

The test pit was then estab-

lished near the sinkhole's sloping wall and measured one meter by one-half
meter, a rather small pit, but one which yielded considerable data.

While preparing the area for systematic excavation, it was necessary to

clear away a portion of Zone
pit could be facilitated.

into Zone

3,

3, so

an approach giving easy access to the actual

During this clearing process, at a depth of 25 cm

a fragmentary left ilium of a human juvenile was discovered.

Based on the size of the ilium fragments and comparison with modern pelves,
it was determined that the ilium belonged to a child about six years of age

at time of death (Clausen, Brooks, and Wesolowsky, 1975:205).

Following this encounter, the test pit was excavated by removal

of 10 cm levels.

At a depth of 35 cm into the Zone 3 deposit, an

282

almost complete first sacral vertebra of

a human juvenile was un-

The degree of

covered (Clausen, Brooks, and Wesolowsky, 1975:201-202).

synotosis occurring in the vertebra indicated an age of six years when the

individual died (Clausen, Brooks, and Wesolowsky, 1975:204).

There was no

evidence to conclusively state that both the ilium and the vertebra came
from the same individual, but such a possibility is indeed high.

carbon sample taken from the same level of Zone

One radio-

as the vertebra yielded

+ 190 years before present (Clausen, Brooks, and Wesolowsky,

a date of 10,260

1975:203).
In addition to the two human bones encountered in a controlled excava-

tion, which by themselves are extremely important finds, Clausen and his

associates have interpreted the sediments of the three zones with regard to
sea level rise and fluctuations over the past 10,000 years.

As determined by Clausen, Brooks, and Wesolowsky (1975), then, we can

make the following summary comments (see also Table 6-2).

cluding Zone

3,

were deposited in a submerged situation.

The zones, in-

For this to have

occurred, the water level in the sinkhole had to remain within the range
of -9.5 to -13 m.

Combined information on each zone and its data

are

presented in the tabulation on the following page.

Zone 2 and
sea level rise.

analyses follow currently held beliefs in post-Wisconsin

As sea level gradually rose, water levels also increased

in the sinkhole during a corresponding period of time.

at Zone 1, some revealing interpretations can be drawn.

However, when one looks

Due to Zone l's

composition of algal sludge, Clausen, Brooks, and Wesolowsky (1975)

have

determined that the sinkhole must have changed from a stagnant-water pond
to a free-flowing spring, approximating its present-day condition.

This could

have occurred only if the water level in the spring had increased to a level

equivalent to the present level.

That, likewise, implies a rise in mean sea

level, which ultimately controls the potent iometric level of the spring.

283

As

CO

CO

U
CJ
c

PC

en

CO

PC

pc

CNl

PC

e
ccj

S
CJ
4-1

a)

CO

CD

>,

CD

M
T3

CJ

>-.

rH

rH

5-i

rH

5-1

cO

cO

W m

(D

CD

o
O
rH
O

rH

C
a)
a
o

5-1

rH

>,

cfl

PC

O
PC

U
CD

u
CD

m
rH

<

CD

JC

CO

4-1

>
CD
4-1

+J

CD

4-J

CO

CO

o
o
m

CD

CO

T)

Ph

CO

O
O

CM

n a

(C

CD

#N

4-1

t3 H
4-1

o
H

T3

5-1

C
O

o
o
LO

CD

00

00

o
o
o

o> Ph

PM

pq

PQ

4J

4-1

CO

o
o
o

r-l

o
o
m
*

cO
CD
>

CO
5-4

CO
CJ

>.

<4-l

(^

cO

rH X)

CO
CD

C
o
N

CX

CD
CD

CD

o
rH

PC rJ

rfl

4J

4-1

JC H

u
CD

-a

U-4

H-l

o
o

I
I

a)

CJ

4-1

rH

rH

oo

a,

to

xi

<D

e
o

o o
CO

OJ

a M

H
JC
to

C
o

Ctf

l-i

H
CD

H rC
4-1

4J

X M H
i

4-1

cx H

CO

a)

5-i

a a a
w
Q
m
4J
cd

CJ

CO

rH T3
CD
!-i

e
CC

s d M
o CJ o
hJ

CO

<4-l

CD

d
o

CO
rH

S
<4H

CM

0)
CD

rH
rQ

5-4

0)

PL)

ea

CO
CO

13

P< en

p>

CJ
cd

ih

CT rH
CD

oi

CD

3=

l-i

*.

CD

a
o
H
4J

5-1

3
oo

-a

<D

CM

CD

CD

4-1

CD

rH
Xi
cO

CD
CO

00
rH

rH

CJ
CO

<

284

rQ

ro
4-1

cO
CD

P4

the researchers have stated, "The simplest explanation for the transformation
...

is that post-Wisconsin sea level ... had reached a point closely approxi-

mating present MSL, and that the springs, responding to a potentiometric

water level similar to that existing at present, began to flow" (Clausen,
Brooks, and Wesolowsky, 1975:206).

The authors, however, do not rule out

other less likely causes for the water increase in the spring.

precipitation rate is one such possibility.

A higher

Work is continuing at this

important site under the direction of Wilburn A. Cockrell of the Florida

Department of Archives and History.

Due in part to its marine setting, the Fish Creek Site in Hillsborough

County (8 HI 105) is representative of eastern Gulf Paleo-Indian sites

(Karkins, 1970).

A large portion of Florida Paleo-Indian point finds,

especially Suwanee points, are associated with aquatic situations, and the Tampa
The Fish Creek

Bay area has been a frequent source for many such finds.

Site is also typical in that it is devoid of any stratigraphic reference

which could be used to determine occupational sequences; therefore, the

assignment of age must depend on the time spans assumed for the

projectile point.

Although habitation at the site persisted beyond the

Paleo-Indian Period into late pre-ceramic Archaic, Orange, and Transitional

Periods, its significance as a Paleo site is foremost in this review.

Lying in a mangrove swamp which borders Old Tampa Bay, the Fish Creek
Site is totally submerged at high tide.

The muck along the western edge

of the mangrove and fill beside the creek mark the most productive locales

for artifact occurrence.

Known surface collections over the years have yielded totals of 183 potsherds and 304 stone tools.
cupations.

The sherds are indicative of more recent oc-

The stone tools collected represent the earlier occupations

and are predominatly ovate and trianguloid knives (52%)

while points

and scrapers together compose 34% of the total assemblage.

285

The fre-

quencies of the point types represented include 12 Suwanees, one Bolen Plain,
10 Archaic Stemmed, one Florida Morrow Mountain,

two Newnans, eight Culbreaths,

three Lafayettes, one Westo, none Citruses, one Hernando two Bradfords, one
,

Pinellas, and two Fish Creek points.

The remainder of the lithic assemblages consists of plano-convex endscrapers, bifacially flaked end-scrapers and side-scrapers, drills, picks,

Shell artifacts

hammerstones, sandstone abraders and "horse's hoof" cores.

include perforated Melongena shell hammers, a possible conch-shell gouge, and

a columella pendant.

Paleo- Indian and Early Archaic Projectile points along with large chipped

bifacial tools and bones of extinct Pleistocene vertebrates have been reported
also from deeper submarine oyster shell deposits of Tampa Bay (Warren, 1964;

Goodyear and Warren, 1972).

The material has been collected from shell piles

that were dredged from the submerged deposits.

The shell deposits themselves are huge; there has been continuous dredging
for some 40 years.
thick.

The reefs are reported to be as much as 16 km long and 15 m

They are composed predominately of Crassostrea virginica (greater than

99 percent) with minor inclusions of conch (Busycon caricum )

mercenaria)
(

sea pen (Atrina rigida )

Strombus alatus )

and others.

olive (Oliva say ana )

clay

Venus

Florida conch

Pieces of water-worn wood, cypress knees,

concretions of sandstone or limestone, cobbles of cemented oyster shell

(sometimes smoothly polished) and clay lumps also occur in the dredged

material.
Shells showing a higher degree of mineralization have been dredged from

depths of -12 to -15 m below a limestone or marly cap of about 45 cm in

thickness (Goodyear and Warren, 1972:55).

Chipped artifacts predominate which are manufactured from local limestone

cherts and silicified coral.

Although a few pottery sherds have been found,

neither shell nor ground stone objects have been noted.

286

The most common types

are crude bifacial choppers, roughly chipped unifacial scrapers of small size,

large unifacial core planes (about the size and shape of a horse's hoof), a

spheroidal hammerstone, and projectile points.

Paleo-Indian and Early Archaic

points include Suwanee, Nucknolls Dalton, Greenbriar Dalton, and Bolen points
(or Big Sandy I variants).

from the dredged shell.

Nine Putnam and Newnan points have been recovered

These are believed to represent the Middle and Late

Archaic.

Pottery is present in small quantities though less common than the stone
tools.

Only five sherds were reported by Goodyear and Warren in 1972.

Pleistocene bones from depths of -12 to -15 m include Alligator

mississipiensis

and Odocoileus sp

Geochelone sp

Proboscidea, Equus sp

Lamine Camelid,

Goodyear and Warren present several lines of evidence which lead them to

believe that the oyster shell deposits are middens, at least in part.

If so,

they represent the oldest reported shell middens in the Northern Gulf area.

Another area of Florida which is noted for recurring surface finds of

Paleo-Indian projectile points is around Choctawhatchee Bay in the Florida

panhandle.

Mrs. Yulee Lazarus, Director of the Temple Mound Museum of Fort

Walton Beach, and her late husband, William Lazarus, started keeping records of
Paleo-Indian projectile point finds in the area over 15 years ago.
is the map shown in Figure 6-54.

The result

While interpretation of single surface

finds must be made with caution, the pattern of finds in the Choctawhatchee

Bay area is believed to be very significant.

With few exceptions, only projectile points have been reported from
most of the locales.

Three have later components:

OK 53, 8 WS 8 (Lazarus,

1965b) and 8 WL 29 (not identified in Figure 6-54; see Vol.

reported by Lazarus, 1965a).

Plate 5,

3,

The bayshore in the vicinity of

WL 31 has

produced three chipped-stone tools which also appear to be related to early

occupations.

As illustrated in Figure 6-55, A-C, these tools are steep

edge-chipped gravers of the Friesenhahn Cave and Salt Mine Valley type.
287

cd

G
O

en

> U
cd
N
c cd

XI

cd

hJ

^
cd

13
H

U
o

<u
0)

T-\

fr >

PQ
CU
<U

U
4J
cd

en
^-<
*-*

o
>s

43

en
0)

Is

4-J

id

M
1
o
U

o
o

a
T3

3
3 CU
o en
u 3
cd

en

x:

T3

g
3

3
H
14-1

*=

CU

C rH
CU
O B

CU

01

iH ,3
H o

cd
CU

0)

PQ

r-

O c
o
-i

r.

4-J

r-i

It

cd

T3

M3
;

4->

M
o
PM

O
a)

l-l

P-.

iw

<*

^D
a)

n
3

M
Pn

288

g
o

cd

dJ

13

cO

uh

T3
H

c
o

cu

u
3

4J

00

CO

O u

iH
fn

cO

rH Td
43 CD
4-1

0)

01

Ai

>

cd

00
rH
cfl
#.
H H
u M-l
43 Pn f^ 00
s

CO

&

rH
CO

4-1

l_J

^2

13
a.

jz H
o 43
cO

4=

CJ

0)

00
TD

S-i

0J

OJ

M-i

W
u

m
cfl

CO

1
1

cu

h
3

M
H
Pm

4=

oo

0)

to

s
OJ
4-J

H
in

^
0)
o>
>-i

> U
OJ

ex

CO

4-J

oj

oo
C/j

rH 43
CO

oo

en

-j-

m o

>x>

a>

o u
I

o>

H
H

CJ

as o

/^\
l_J

LO

u
o

<

>.

tO

6 M S
o CO n)

4->

OJ

a)
<D

Ph oo

289

cd

CO

CO

<!

43

o
M

-u

U-l

01

to
0)

The entire Paleo- Indian Interval is well represented by the projectile

points from the Choc tawhatchee Bay area. In addition, the steep edge-chipped

gravers may represent a pre-fluted point horizon.

In this area, we again find evidence of Paleo-Indian occupations in a

coastal setting.

The locales fall into four ecological settings

in the following

list:

BARRIER SPIT COMPLEX

TERRACES AND UPLANDS

Upland
Stream

Bay
Margin

Back
Barrier

Gulf ShoreDune Lake

5

Total

27

41

8
8

as shown

WS 8
JA "A"

8
8
8

8
8
8

OK
OK
OK
OK
OK
WL

26
67
20
53
58

8
8

8
8

8
8

OK
OK
OK
OK
OK
OK
WL
WL
WL

23

OK 57
OK 5
OK 74
OK 63

BY "C M

44
35
46
54
62
35
71
31

8
8

No data are available regarding the specific settings of the upland

stream locales.

Most of the bay-margin finds have been made in situations

where old terraces around the inner bay shores are eroding back.

The arti-

facts, winnowed out of eroding sites, are commonly found along the active

beach or in the shallow waters of the bay.

The site situation at back barrier sites is very similar to that along
the inner bay shores.

Artifacts are usually exposed by erosion and found along

the beach or in the bay shallows (Figure 6-56).

However, a distinction has

been made between the two types of settings because of differences in origin,
age, and fauna of the geomorphic features upon which the sites may have been

located.

Gulf shore-dune lake sites are also associated with barrier spit complexes.

Two of these locales presently occupy what might be considered bay

290

Figure 6-56.

Figure 6-57.

The Alligator Point Site (8 BY "C")

located
along the entrance to St. Andrews Bay. View
to the southwest.
Date: 8/12/75.
,

Freshwater pond in coastal dune field between

Choctawhatchee and St. Andrews Bays, Florida
View to
This is also the location of 8 WL 63.
Date: 8/15/75.
the west-northwest.

291

margin locales (8 OK 57 and

OK 5).

However, the barrier island (Santa Rosa

Island) that presently separates them from the open Gulf is believed to have

formed subsequent to the time when the points were made.

That is, Santa Rosa

Island is a product of the Late Holocene transgression and post-dates the

projectile points.

While these two point-find locales are related to modern

beach erosion, the sites from which they were presumably derived were either on
the open Gulf beaches or more likely in dune fields behind the beaches.
8

WL 74 and

Locales

WL 63 are associated with freshwater ponds or lakes in dune

fields on the barrier spit complex (see Figure 6-57)

The distribution of Paleo-Indian projectile points in this area implies

that the barrier spit complexes with which they are associated were active during
the Paleo-Indian Period (Subintervals H^ - H3)

The distribution also provides

an indication of the kinds of coastal geomorphic situations where Paleo-Indian

materials might be found.

From the standpoint of negative evidence, it is

significant that in spite of substantial evidence of Paleo-Indian use of the

area, concentrated accumulations of artifacts or other site indicators have

not been found.

Nor have Paleo-Indian shell middens been identified in this

Detailed studies of this key area would undoubtedly contribute substantially

area.

to our understanding of both Paleo-Indian culture and Late Quaternary geological

history.

Eastern Gulf Archaic

The Lake Kanapaha Site in Alachua County (8 AL 172) provides an excellent

example of a Florida Archaic, upland site because it contains mainly scattered

concentrations of lithic tools and tool-making remains (Hemmings and Kohler,

1974).

Located on the western shore of the 200-acre (800,000 m

2
)

Lake Kanapaha,

the site encompasses approximately 500 acres (2 million m2) of mostly pastureland and lies generally within the 80-foot (24.4 m) contour line, representing
an environmentally suitable location for habitation today, as it probably

did 7,000 years ago.

292

The site

lies at the junction of three geomorphic divisions of

the Alachua County highlands:

plateau zone to the north and east, a karst

plain to the west, and a lake and prairie zone to the south and southeast
(Figure 6-58).

These divisions indicate distinct landforms and drainage

patterns which reflect different underlying structural formations, chiefly

Hawthorne sediments and Ocala limestones.

The subsequent diversity in bi-

otic exploitation potential was strongly influential in the selection of

this locale for habitation and camping activities since Archaic times.

Ex-

posed Arrendondo limestone quarries within 1.6 km of the site present another

geologic unit that would have favored early settlement.

Excavations at this site took place at five separate areas and revealed
three distinct occupational zones, extending down 2.4 m through recent aeolian sand layers.

Below the aeolian sand lies a sandy-clay substrate with

relict chert boulders at its uppermost surface (Plio-Pleistocene?)

boulders not only provided raw material for making tools

These

but also

contributed to the selection of this site for Archaic occupation, indicating

the strong interrelationship of environment and settlement patterns.

A total of 142 square meters and 56 trenches were involved in the excavations, and the resulting three-fold stratigraphic division was based on
the occurrence of diagnostic ceramics and projectile points.

Areas 1 and

proved to be the most productive archeologically and are the basis for gener-

alizations about the site.

The uppermost ceramic zone was usually about 35 cm thick and yielded
39 sherds, 10 stone-cutting and scraping tools, and a charred turtle cara-

pace, the only evidence of food remains found.

This level indicates a Deptford

Period occupation of small temporary campsites along the Lake Kanapaha

shoreline.
The middle, or pre-ceramic Levy Zone (named for a recurring projectile

point type found at the site)

extended from 30 cm to 71 cm and was well

293

Figure 6-58.

Location of the Lake Kanapaha Site (8 AL 172) and surrounding

lithic and ceramic sites.
Note also the location of the Arredondo Fossil locale (After Hemmings and Kohler, 1974).

294

represented in Areas 1 and

by 54 stone tools, including 14 Florida Archaic

Stemmed projectile points, seven scrapers, nine bifaces, six cores or planes,
six irregular cores, one hammerstone, and 8,000 flakes and fragments, usually

associated with

f lintknapping

activities.

The stone was of a local source

and often displayed heat treatment.

Throughout Areas 1 and 2, over 7,000 pieces of stone debitage were differ-

entially distributed.

Their distribution reflects the settlement and use

Areas with a large percentage

patterns for at least these areas of the site.

of debitage from cores,
f lintknapping

preforms, and hammerstones were presumably used as

locales, whereas tools such as projectile points and scraping

and cutting tools were found in areas of low flake density.

This suggests

an aboriginal areal division between living and working areas.

A third pre-Levy zone, located from 0.8

poorly represented by seven stone tools:

"horse's hoof" core or plane,
flakes.

to 2

m below the surface, was

three end or side scrapers, one

one small bifacial fragment, and two utilized

All bifaces, cores, and flakes were from local chert.

An absolute

date has not been assigned to this level.

Excavations in Area

did not exhibit the same stratigraphic divisions

as Areas 1 and 2, but a single Late Archaic component

Zone,

(equivalent to the Levy

in Areas 1 and 2) was indicated by the tools and debitage found in the

topmost 61 cm of sand.

The appearance of 26 stone tools and a multitude of

heat-modified chert flakes helped define Area

as a focal point for heat

treatment and related knapping activity.

The stratigraphy and artifacts at the Lake Kanapaha Site allow an in-

terpretation of the functional and temporal placement of this site.

Similar

flint tool kits and debitage and, frequently, similar environmental conditions characterize many of the upland Archaic sites in the eastern Gulf area.
The aeolian origin of the sand bed from which all artifacts were ex-

cavated is

of

more than passing interest.

295

Characteristically, it is a

tan, homogenous sand locally reaching 2.44 m in thickness, masking the cor-

rugated surface of the underlying Plio-Pleistocene (?) clay.

Hemmings and Kohler (1974:48) note that, "The Lake Kanapaha aeolian
sand is not an isolated phenomenon, as the artifact-bearing deposits at

Bolen Bluff

Johnson Lake

(8

AL 439) in Alachua County and Silver Springs

MR 63), and the Senator Edwards Site

(8

MR 92),

(8

MR 122) in Marion

County are all essentially identical (Bullen 1958; Neil

1958; Bullen and

(8

Dolan 1959; Hemmings 1973; Purdy 1973).

This sand can also be seen at the

surface in many roadcuts and borrow pits in the Central Highland area."
The archeological data indicate that the aeolian sand was deposited during
the interval from 7,000 to 4,000 years B.P.

(Intervals

and J).

Eastern Gulf Poverty Point

The cultural transformations marking the termination of the Archaic

Period in the eastern Gulf area are evidenced in sites of the Orange, Transitional, and Elliot's Point Complex culture periods.

In the

Choctawhatchee

Bay area of Florida, these changes are discernable in a cluster of sites which

comprises what is considered to be the temporal and art if actual equivalent

of the Poverty Point Period in the central Gulf area.

This Florida site

cluster was randomly represented by surface finds of fired-clay objects

(Figure 6-59) until excavations at the Elliot's Point site (8 OK 10)

(Lazarus,

1958) permitted the inference of stratigraphic sequences and resulted in a

definition of the "Elliot's Point complex."

There are presently 18 sites in

northwest Florida that exhibit Elliot's Point traits.

Elliot's Point is a bay-margin site on the margin of 3.7 m above sea
level in Fort Walton Beach, Florida.

The stratigraphy of the Elliot's Point

site as revealed in excavations of three 24-square foot (2.2-square meter)

sections presents an upper zone of coarse yellow sand, extending down to the

296

The only discontinuities in this two-

maximum excavation depth of 90 cm.

fold division are a brown sand pocket intruding into the yellow sand in

Section III and a layer of black sand occurring in the upper layers of

Section

I.

Stratigraphy and analysis of the 418 artifacts recovered during the

excavation yielded chronological data relevant to the Florida clay-object
complex.

The top

51 cm of the three sections produced a total of 362 sherds,

along with coral fragments, fire-blackened quartz pebbles, charcoal, and

shells of the eastern oyster, southern quahog, Florida arith, and snails.
The underlying yellow sand was the matrix for almost all of the baked clay
objects, which are, at 8 OK 10, typically peach-shaped spheroids with shallow

longitudinal grooves.

The clay objects were usually found with flint chips

and flakes below the pottery zones, and less frequently, with sherds from
the Deptford Period, which as an approximate time span of 500 B.C. - 200 A.D.

The repetitive association of baked clay objects with flint chips and tools

exposes this as a basically pre-ceramic clayball assemblage, which is so far

represented in Florida by the Choctawhatchee Bay area almost exclusively.

Most of the sites in the Choctawhatchee Bay area having an initial

occupation during the Elliot's Point Period (about 3,500 B.P.) also have
components from later culture periods

Many sites in this same area have also

yielded projectile points (mostly from undetermined contexts) which date

back 10,000 years (see Figure 6-60).

As previously discussed, these points

are sometimes found in sites assumed to be of Elliot's

explanations for their occurrence can be offered.

Point age.

Three

They may have been an-

tiques collected by Elliot's Point people; or the sites, most of which are
in environmentally favorable situations, may have been initially occupied

297

$jp

-ijjjp ij/L

*$W

Figure 6-59.

Elliot's Point Complex clay balls, similar to

Poverty Point objects found in Louisiana and
Mississippi.

Figure 6-60.

Paleo-Indian and Archaic projectile points from

Choctawhatchee Bay area. Sites 8 OK 35 and 8 OK 5
have also yielded artifacts of Elliot's Point age.
A,B, "Paleo point" bases; C, Westo; D, Suwanee; E,
Tallahassee; F, Late Archaic "limestone" point; G,
Alora; H, Dalton; I, Bolen (?)
J, Archaic "hematite" point.
,

298

during Paleo- Indian times and then reoccupied during Elliot's Point time;
or the points may have a long time-span dating from Paleo times through
to Elliot's Point times.

The second explanation seems the most favorable,

299

CHAPTER VII
AN ILLUSTRATION OF METHODOLOGY:
THE MISSISSIPPI DELTA AREA

Introduction
Ideally, this section of the report should

summarize chronolog-

ically the relationship between the prehistoric occupation sequence

and the relict landforms on the continental shelf and in the coastal
zone for each major area.

However, it was reluctantly concluded to be

an impossible task within the scope of the present study.

lems encountered in such a summary include:

1)

Major prob-

Archaic, Paleo-Indian,

and Early Man archeology is still very poorly known in most of the

study area;

2)

the relationship between human occupation and geologic

events is not Well understood, particularly in the time period of

concern;

3)

the Late Quaternary geology beyond the past 5,000

years is still very poorly understood;

4)

there are large un-

resolved discrepancies in published geologic interpretations within

areas and between areas; and

5)

there are large discrepancies in in-

terpretation of Late Quaternary dating between a purely geological

approach to the problem and an archeological-geological approach.

The fact that many of these problems remain unresolved does not

indicate lack of interest from the scientific community.

literature treating these topics is indeed voluminous.

The scientific
The difficulties

lie in the magnitude and complexities of the problems themselves.

In spite of these difficulties, it is felt that an attempt to

further illustrate the methodology is in order.

This will be done

through a case study of a single large area, the Mississippi Delta,

which combines the east and west Louisiana shelf areas and the related coastal zone (Volume III, Plate 2).
300

The Mississippi River Delta

The Louisiana coast is dominated by a 300-mile lowland consisting

of large tracts of marshes and swamps and innumerable lakes and bays.

This extensive near-sea-level area makes up the Late Quaternary deltaic

plain of the Mississippi River and has resulted from deposition of river
sediment (Figure 7-1).

The Mississippi River Delta, like all deltas,

is a zone of interactions

between fluvial and marine processes and

constitutes one of the most dynamic situations in nature.

The balance

between the rates of sediment deposition and the combined effects of

subsidence and erosion by the sea cause shorelines in deltas alter-

nately to advance and retreat.

Seaward growth occurs at the mouths

of active streams, whereas erosion results near the mouths of inactive

streams which no longer transport sufficient sediment to sustain their

positions or advance seaward.

Delta building can be thought of as a

contest between the river and sea.

If the river deposits sediment

faster than the sea is able to remove it, new land is added to the
shore and the delta builds seaward (Figure 7-2)

As the delta is ex-

tended, it gradually builds upward, or aggrades, by processes associated with lateral shifting of channels, by sediment deposited during

overbank flooding, and by accumulation of plant and animal remains

(Figure 7-3)

Deterioration of the delta occurs if all or part of it

is deprived of the supply of river-borne sediment necessary for its

continued growth.

This results in removal of the seaward edge by wave

attack and the settling or subsiding below sea level of the surface
(Coleman and Gagliano, 1964; Frazier, 1976; Gagliano and van Beek, 1970)
In a low coastal plain where plant and animal communities are

delicately adjusted to minor differences in elevation and salinity,

301

302

A. INITIAL PROGRADATION
NATURAL IEVE

DELTA-FRONT
SILTY SAND
SILTY CLAY

AND

PRODELTA
CLAY

SILTY

B.

ENLARGEMENT BY FURTHER PROGRADATION

DELTA. PLAIN

DELTA-PLAIN

INORGANIC
SILTY

CLAY
DELTA. PLAIN N A TU R A

CLAYEY
C.

SILT

DISTRIBUTARY

AND

L-

SILTY

LE

VE E

CLAY

ABANDONMENT AND TRANSGRESSION

MORIBUND DISTRIBUTARY>vcTns? *,P=25^^^^ N EE sV UL ? N G " BUTARY

FROM STREAM
l

TRANSGRESSIVE
DELTA-MARGINISLAND SAND

TRANSGRESSIVE
BAY DEPOSITS
D.

REPETITION OF CYCLE

REOCCUPATION OF OLD
DISTRIBUTARY COURSE

Figure 7-2.

Block diagrams illustrating progradation and

transgression in a delta with a bifurcating
branching habit, such as the Mississippi
(After Frazier, 1967).

303

LOWER DELTAIC PLAIN

UPPER DELTAIC PLAIN

COASTAL MARSH

LEVEE FLANK

NATURAL

DELTA. FRONT FACIES

PRODELTA

MAJOR
STREAM
COURSE FRESH- WATER INLAND SWAMP

T***

FACIES TRANSGRESSIVE^

BAY FACIES
DELTA. PLAIN

INORGANIC
FACIES

INTERDISTRIBUTARY
p EAI

PRODELTA

BLOCK A
NATURAL-LEVEE
DEPOSIT POINT BAR
DEPOSIT

COASTAL AREA
OVERLAPPING DELTA LOBES
ASSOCIATED WITH
DISTRIBUTARY

NETWORKS

BLOCK

"7~

DELTA-FRONT
DELTA- PLAIN

THICK INLAND.

INORGANIC

SWAMP pEAT

INLAND AREA
INITIAL

PROGRADATION

AND CONTINUING AGGRADATION

ASSOCIATED WITH
MAJOR STREAM COURSE

Figure 7-3.

Diagrammatic representation of the relationship between

major morphologic features and sedimentary facies in an
advanced stage of delta building (After Frazier, 1967).

shifting of streams and changing of shorelines are accompanied by

changes in ecology and environment.

Therefore, during the long evolu-

tion of the deltaic plain, not only has the morphology changed through
time, but at any given place a gradual environmental succession has also

occurred.

Deltaic areas have been important to man since earliest prehistoric times.

They usually abound in wildlife and edible plants, and

the many waterways provide natural routes of transportation.

From

the biological standpoint, marshlands which make up a major part of

the deltaic surface have the highest primary production of any conti-

nental habitat.

One measure of this productivity comes from modern

304

commercial fisheries records.

The deltaic coast of Louisiana con-

sistently accounts for approximately 25 percent of the nation's

fish harvest (exclusive of Alaska)

(Lindall

e_t

a_l

1971).

It is no

accident that early civilizations flourished in the deltaic lowlands

of the Nile and the Tigris-Euphrates.

Alluvial soils are usually

fertile, and irrigation and soil replenishment result from annual flooding.

Man's utilization of deltas, however, has not been an entirely

happy experience since these areas often are subject to floods, harbor
diseases, and the adverse effects of coastal storms.

As shown in Figure

7-4, the area of the Mississippi Delta was no exception.

More than 600

known archeological sites attest to the attractiveness of the deltaic

plain to prehistoric peoples whose economies were based on hunting and

gathering or primitive agriculture.

It should be noted that the known sites represent only a sample

of the total number of archeological sites.

Systematic survey of

even high-probability areas has never been completed in the entire

area.

Intensive ground survey has been done in only a few relatively

small areas.

Innumerable sites are believed to be buried or veneered

by natural sedimentary deposits.

A very large number of sites has been

lost to erosion and modern destruction. There is an important need to

undertake a carefully designed intensive survey in a portion of the

Louisiana coastal area to test the concepts of high-probability areas

and to determine what percentage of the total population of archeological

sites is represented by the known sample.

Archeologists in south Louisiana have emphasized the close relationship between prehistoric habitation and changing morphology and
environment.

Effects of the dynamics of delta building and change on

both the environment and man's activities have received particular attention.

This approach and viewpoint have resulted from the work of

305

s
H
CO

3
O

u
CO
CO

o
cj

ti

H
en
cu

u
H
en

o
H
00

o
O

tH
CD

C
o
H

X>
iH

M
4-1

CO

r^.

0)
S-i

3
to
H

306

Kniffen (1936), Mclntire (1958), Saucier (1963), Gagliano (1963), and

others.

South Louisiana can be considered as a distinctive area which

was subject to cultural influences from adjacent regions as well as to

developments which appear to be more or less indigenous.

Thus, in the deltaic setting there is a unique situation in that
the morphology and environment have been subjected to continuous and

rapid changes through time, and primitive peoples inhabiting the area

were forced to adjust to these changes, at the same time experiencing

gradual and meaningful evolution in culture and technology.

The un-

raveling of this complex story of interactions between primitive man

and his environment has resulted in a unique interdisciplinary approach
to archeology and the study of coastal environments and has produced
a method and theory that is particularly useful in the present study.

For this reason, natural processes and forms and related cultural as-

sociations are considered in some depth.

The geomorphic and environmental framework of the delta exerts

strong influence on such things as settlement pattern and economy.

In

the coastal portion, the distribution of habitation sites is dictated by

the location of Gulf and lake beach ridges and Mississippi River natural

levees (Figure 7-5)

These features provide virtually all of the firm,

relatively high places suitable for even periodic occupancy.

Deltaic Plain
The deltaic plain consists mainly of near-sea-level lakes, former

stream courses, marshlands, and swamps.

The main streams which pre-

sently flow through this area are the Mississippi River and its major

distributary, the Atchafalaya River.

The Mississippi flows across the

eastern side of the plain and discharges its sediments through several
active distributaries, forming the present birdfoot delta.

307

Evidence of

CU

si

c
o
CU
4-1

CO

O
o
CU
cu

CO
CD

p
H
CO

C
co

MC

CO
0)

>-,

C
CO

00
T3
-H

u
cu
CU

>

a-

cu

Cd

rH

CO

.H

o
cd

T3

C
co

u-i

O
cu
CU

T3

> 5
3
O

CU

--H

.h

oo

CO

U
3

4=
00
4J -H
CO J^

(3

u-i

H
4J
CU
CO
1

CO

CU

>
-H
-U

CO

cO

rH

>-(

CU

m
cu

oo
fa

308

>^
-H

abandoned stream courses comes from numerous minor drainage lines and

topographically high natural levees of former streams which vein the

coastal marshes.

Marginal Plain
Almost one-third of the Louisiana coastal area lies west of the
deltaic plain proper and is termed the marginal deltaic plain, or
chenier plain.

This marginal plain, which developed during the Holocene,

owes its origin primarily to sedimentation accompanying westwardthe

flowing littoral currents when the river mouth was located toward

western Pleistocene boundary.

When the river shifted eastward,

sedimentation ceased, and concurrently with coastal retreat, beaches

were formed.

The relict beach ridges, or cheniers, mark progressive

positions of the Gulf shoreline during minor retreats in a general pro-

gradation that has taken place during the past 3,500 to 4,000 years
since sea level reached its present stand.

The beach ridges have pro-

vided habitation places in the otherwise marshy coastal environments.

It has been demonstrated that initial human occupation on a ridge ap-

proximates the time when it was an active Gulf beach.

Thus, the oldest

sites in the marginal deltaic plain are found on relict beaches farthest removed from the present shoreline.

In another section of this

report (Chaper VI), the Copell Site (16 VM 102), which is located on
one of the oldest beach-ridge complexes in the marginal plain, is discussed.
It is also important

to note that the marginal plain is the surface

of a prism of Holocene deposits which overlaps a Late Pleistocene del-

taic surface.

This older deltaic surface has been tilted so that it

stands as a terrace north of the marginal plain and is submerged on

the continental shelf to the south.

309

Marginal Basin
In reference to the dominant longshore currents, the marginal

basin lies updrift of the Holocene deltaic plain of the Mississippi

River.

For this reason, it has received less sediment from the river

system than either the deltaic plain or the marginal plain and has

developed into an extensive basin area as the deltaic plain prograded

seaward.

The basin is dominated by three large, shallow fresh-to-

brackish lakes (Lakes Maurepas, Pontchartrain, and Borgne)

These

lakes, along with bordering swamps and marshes, lie between the natural

levee ridges of the modern Mississippi River on the south and on the
The Late Quaternary history of the basin

Prairie Terrace to the north.

is complex.

Included in the area are Pleistocene and Early Holocene

barrier-island complexes, old delta lobes of the Mississippi River, and

complex estuarine areas of smaller streams draining parts of the coastal

plain east of the Mississippi (i.e., the Amite, Tickfaw, Tangipahoa,

Tchefuncte, and Pearl Rivers).

Alluvial Valley
The boundary of the Late Holocene deltaic plain, which merges

inland with the alluvial plain of the lower Mississippi Valley, is ar-

bitrarily placed where the valley begins to widen toward the Gulf of
Mexico.

Inland in Louisiana, the Late Holocene alluvial valley is in-

cised into Pleistocene terraces and Tertiary material.

These older sur-

faces heighten from near sea level at the contact between the Prairie

Terrace and Late Holocene coastal wetlands to several hundred feet in

the northern part of Louisiana.
The valley is occupied by active and relict meander belts of the

Mississippi and Red River.

At times, the Red has been tributary to the

310

Mississippi; at other times, it has had an independent course to the

Gulf.

There are also extensive freshwater swamp and lake areas, some

of which have persisted for thousands of years.

Terraces and Uplands West of the Alluvial Valley

Included in this category is the Prairie Terrace, extending from
the west wall of the Late Holocene alluvial valley of the Mississippi

River to the Sabine River.

Relict features on the terrace surface in-

clude remnants of a major meander belt of the Mississippi River and a

large freshwater swamp laced with crevasse distributary channels and an

independent course of the Red River which terminated in lacustrine deltas.

Salt dome islands in the southeastern part of the area stood as

prominent topographic features when the delta was active.

Subsequent

events have tilted the old delta surface upward and resulted in its

drainage, so that it now stands as a terrace.

Geologists have long regarded the Prairie Terrace and relict

features on its surface as Pleistocene in age (Fisk, 1944, 1956; Kolb
and van Lopik, 1958; Bernard and LeBlanc, 1965, and others), with age

estimates of from 80,000 to more than 100,000 years.

is deeply entrenched in the geological literature,

While this concept

it is basically in-

compatible with archeological data and some newly developed geological

data.

An alternate hypothesis is presented here in which relict fea-

tures on the Prairie surface are considered to have formed during Early
and Middle Holocene times.

The Prairie Terrace is bordered on the northwest by older, more

deeply dissected and weathered Pleistocene terraces.

Distinctive es-

carpments mark the contact between the Prairie surface and these older
terraces.

In places, valleys of streams draining these uplands and ex-

tending onto the Prairie surface have alluvial terrace equivalents of

the Prairie extending for some distance into the uplands.

311

Terraces and Uplands East of the Alluvial Valley

The area from the east wall of the Late Holocene alluvial valley
and northern rim of the marginal basin of the Mississippi Delta
to Mobile Bay in Alabama is divisible into the Prairie Terrace unit

bounded on the north by dissected and weathered Pleistocene Terraces

(Uplands).

The Prairie Terrace in Louisiana is relatively flat, but

does exhibit a number of rather subtle relict features.

These include

delta lobes of coastal plain streams (Amite, Tangipahoa, and Pearl

Rivers), relict beaches, and dunes derived from point bar sands of
coastal plain streams.
The Prairie Terrace along the coast of Mississippi is dominated

by relict barrier islands and lagoons.

The Amite, Pearl, and Pasca-

goula Rivers exhibit a series of sub-Prairie alluvial terraces that have

been termed Deweyville and related to Early to Middle Holocene (Gagliano

and Thorn, 1967).

Prehistoric Land Use

The Indian chose the sanctuary of the natural levees, salt domes,

cheniers, and beaches of the delta as the locations for his campsites
and villages.

While this area had its disadvantages in the form of

ever-present insects and the constant threat of flooding, it balanced

these with several advantages.

Perhaps the most important was the

abundant food supply provided by the natural landscape.

rich in fish, molluscs, and edible plants.
tiful as well as fruits, berries, and nuts.

The waters were

Animals and fowl were plenThis plentiful food supply

attracted man to this deltaic area for thousand of years.

Physical evi-

dence found in the numerous middens, mounds, and quarries attests to the
long, continuous sequence of human occupation.

312

Of the more than 600 known sites in the area, more than half are

Middens are refuse heaps composed mainly

coastal shell middens.

of the shells of molluscs, which provided a basic element of the

diet, but include ash, bones, shell, soil, pottery fragments, and other

debris of daily living.

The midden is one of the most desirable places

for the collection of cultural remains because it provides the best cross-

section of the daily habits of prehistoric life styles.

materials represent

But the cultural

only a part of the resource value of these sites

Since the inhabitants gleaned food from neighboring environments, the

midden matrix contains a concentrated sample

sequence

usually in a stratified

of environmental indicators (ecof acts)

of a wide variety of animals, seeds and pollen.

such as shell, bones

Shells of the

brackish-water clam Rangia cuneata make up the greatest portion of the

midden material and attest to the importance of these animals in the

economy of the coastal inhabitants.

Crassostrea and Unio are also

often found in the midden and reflect different environmental situations.

While the vast majority of these coastal shell middens are less than
3,000 years old, and thus by our earlier definition

are too recent for

consideration in this study, they are mentioned here to emphasize the

attractiveness of

deltaic environment to early peoples and because

they provide essential models for prospecting for drowned sites on the

continental shelf.

Relationships between delta development and human habitation

have now been extended back some 12,000 years in the coastal

Louisiana area (Gagliano, Weinstein, and Burden, 1975).

Although

many of the details are still very sketchy, the overall pattern

313

is reasonably well established and provides an important conceptual

framework for students of archeology and Quaternary geology.

of nine delta complexes and lobes is shown in Figures 7-6,

A sequence

7-7, and 7-8,

spanning the time from about 12,000 years ago in historic times.

Ar-

cheological sites related to development of the first nine lobes are

An approximate chronology of the complexes and lobes is given

shown.

in Figure 7-9.
1)

Lafayette Complex

Paleo - Indian to Early Archaic

Based on

habitation pattern and fossil vertebrate associations, boring logs,

surface morphology, and radiocarbon datings, we now have a picture of
the Paleo-Indian deltaic landscape.

The most prominent feature of that

landscape was a well-developed meander course of the Mississippi River

trending through what is now the Avoyelles Parish area near Marksville,
Louisiana, and continuing south through Opelousas and Lafayette, Louisiana (Figure 7-10)

The complexity of meandering and cutoff suggests

that this course was active for 1,000 to 2,000 years.

South of the

present location of Lafayette, Louisiana, the trunk channel branched

into three separate deltaic lobes extending into the Gulf of Mexico.

One of these is identified as the Sabine Bank Delta Lobes, the second is

unnamed and the third is called the Tiger Shoal Lobe.

make up the Lafayette Delta Complex.

Collectively they

These have been down-dropped by

faulting and subsidence and now lie drowned on the continental shelf.
Landward of the deltaic lobes was a vast overflow swamp (Eunice
Swamp)

Lacing the swamp were crevasse distributaries from the Mis-

ssissippi as well as an inland delta of the Red River.

While many freshwater swamps are not as productive as marshlands,

overflow swamps receiving a large annual supply of river water and sediment do tend to be highly productive from the biological standpoint.

314

315

316

317

<->

*
...
.....

:*::

.......

0)

|h

i:

cr
(0

(0

Z
UJ

o
(0
>

u
2

LU
CC

LU

o
00

o
<
-1

0)

3
O

03

0)

(0

J3

>>

T3
CU

u
CU
U

4-)

-H

elec

ng

_l

CO i-t

***

00

o
cc
<
z

T3

r-l

ti

CO

ta.

i
CU

X)

a
***

isch

rial

rH

6
O
U

cu

co

,0

es

*-*

k.

CU

o
E

LU
CO

0)

Q.

****

**

(/)

o
O

cfl

4-1

rH

cO

>
CO
,fl

QJ

(0

h-

(/)

Z
3
o S
o
z : ;&:
< I5>
'

0)

;i!

O
4-1

00

d
-H

X)
rH
>H

!!!

....

ctj

CU

cc

LU
>-

'H

a)

CO

<

10

O rH
rH 4J

****

<

Q.

LU
cn

o
4-1

*"*

cc

CD
*-"

iiiiiii

X
o

CU

-i

LU

T4-

cfl

H ,C

Q.

00

O
LL

O
4.)

u
O

ji

;. ;.;.;.;

0)

Q.

CO

CNJ

LU
CO
LU

w-i

Xi

h-

CO

3
O
****
'

4-1

LU
J.

o
LU

(X
H

?5j?

-i

CU

>

CO

CU

CO

>H

4J

O
C

bel

Missi

was

is

Q
z
<

nd

O U

1 :#: x
g
:::

CO

u
=
u

(/>

O
'*

<D

th

x:
00

3
O
rJ

hron

i*::
......

)-i

rH

:::

>-

cfl

cu

^
>
00 -H

or

B.P.
anyon

nary

<fr

;;;

s
u u o

-J

yea

TRO

Quate

submarine

17,000

SIPPi
Late

::: .;:

(0

SS:
:::

:
:

!i!i j:

c/>
</)

2
5

ON
1
1

CU

u
3
oC

GO

318

H
fe

e
E

o"

\
_J

(0
-J

CO
CO

to
to

o
LL

o
_

LU
t-

CO

(0

<

H
O
<
_
H

ffl

LU

H
CC

LU

<

>

Q
Z
<

X
_

-J
co

O
O
CC
LU
cc
cc

<
ffl

LU
CC

LU

O
O
H
_J

10

_J

3
O

CO
a.
cc

<

<

CO

to

LU

-I

H
Q
-1
LU
CQ
>

LU

DC
LU

cc-

>l-

LiJ

to
Ml OC

CO

COH
Q LU Z
z (TO
< OCT
LU
LU
i DC
li.

319

<

cr
cc
LU

H
_l

tf

<
COh

<
-

2 _

CO

< LU
CCQ
H

o
o

<

The annual overflow replenishes nutrients and creates ideal habitat conditions for the red swamp crawfish

Prochambarus clarki )

a very important

food-chain organism in alluvial swamps in the northern Gulf coast area.

The modern Atchafalaya Basin in Louisiana provides an excellent example
of such an overflow swamp system.

(See Gagliano and van Beek, 1975.)

The old swamp bed now stands elevated and drained on the terrace surface,

but its flat topography constitutes the rich rice-growing area of south-

western Louisiana.
East of the major delta and meander belt was an embayment (Vermilion

Embayment) along the eastern shore of which was a chain of uplifted salt
dome islands.

The important Salt Mine Valley site at Avery Island (16

IB 23) was either on or very near to the shore of this embayment at this

time.

The cut-and-fill sequence of deposits related to local streams

on the surface of Avery Island (discussed in Chapter 4) provided one of

the keys leading to the present interpretation.

Shells of the brackish-water clam Rangia cuneata in the lower levels

of the Salt Mine Valley site in a horizon dated at about 12,000 years
B.P.

(Figure 7-10, Radiocarbon Date 2) were probably obtained from the

nearby embayment.

The size of the Salt Mine Valley site suggests that

it may have been an important seasonal gathering place for Paleo-Indian

peoples.

Other data come from sites intimately associated with alluvial

features of the old river system.
and Grand Prairie (16 SL "A")

For example, at Tate Cove (16 EV "A")

a series of sites occurs in associa-

tion with old Mississippi meander scars (Figure 7-11)

Over one hundred

Late Paleo-Indian projectile points from these sites have been examined.

Although not systematically worked, the sites have also produced tools
and debitage.

The Bayou Grand Louis site (16 EV 4) has yielded Late

320

92*00
30*45

<$

PALEO SITE

(If

FOSSIL LOCALE

NATURAL LEVEE

CREVASSE CHANNEL -LEVEE

ELEVATIONS

Figure 7-11.

IN

FEET

Reconstructed paleogeography of a part of the Lafayette

meander belt (circa 8,500 years B.P.) showing related
Paleo-Indian sites and vertebrate fossil locales. Late
Paleo-Indian projectile points and artifacts have been
recovered from the Tate Cove (16 EV "A"), Grand Prairie
Mas(16 SL "A"), and Bayou Grand Louis (16 EV 4) sites.
todon remains have been found in backswamp deposits at
the Bayou Mallet locale and in point-bar deposits at
Bayou Callahan.

321

Paleo-Indian fluted and piano points as well as tools and debitage.

This site and the Vatican site (Lafayette Parish) are located on natural
levee ridges of old crevasse distributaries that emptied into the swamp.

Bayou Grand Louis also has an Archaic component believed to be related

to some later ecological changes that will be discussed in another

section.
The Bayou Blue site (Allen Parish) is a stratified Paleo-Indian

through Archaic site associated with an upland stream that entered

the Eunice Swamp area.

The Strohe site (16 JD 10) is also strati-

fied with Paleo-Indian, Archaic, and Marksville through Plaquemine

components.

The Strohe

site appears to be on the distal end of an

old crevasse distributary natural levee ridge.

Mastodon bones have been recovered at the Bayou Mallet locale

(St.

Landry Parish) from a backswamp facies, while at Bayou Callahan

(St.

Landry Parish), mastodon remains have been found in point-bar

deposits.

To the south at the Trappey location (Lafayette Parish),

mastodon remains have been recovered from a soil horizon below natural
levee deposits.

These bones have been dated at about 12,000 B.P.

(Figure 7-10, Radiocarbon Date 3).

Lying east of the meander belt was an abandoned delta complex

that is known only from borings.

Radiocarbon datings suggest

that

the lobe of this delta, in the Ship Shoal area, was active about

13,000 years ago and at a lower sea level.

it was abandoned by the river,

By Paleo-Indian times,

and by 12,000 B.P., it was partially

submerged and in a condition of transgression or retreat.

In what is now southeastern Louisiana, there appears to have been

another large embayment (Pontchartrain Embayment)

Scattered Paleo-

Indian finds have been made on remnants of beach ridges that were
once active at the north shore of this embayment.
322

While the dating is still tentative, streams draining the uplands

such as the Sabine, Amite, Tangipahoa, and Pearl Rivers appear to have

been building deltas during this interval.

Late Paleo-Indian sites and

locales in the general vicinity of the present-day city of Baton Rouge

are associated with areas marginal to the Amite Delta.

At Jones Creek

(16 EBR 13) mastodon remains and late Paleo-Indian artifacts have been

found and the Denham Springs locale has produced mastodon bones.
Two isolated finds of Paleo-Indian projectile points, a Clovis
and a Quad, have been made within the old delta area of the Tangipahoa

River, but specific geologic associations remain to be established.

Several other bits of geological data contribute to an understanding of the overall picture.

At Tunica Bayou, for example, organic

deposits associated with mastodon remains at the base of an alluvial

terrace on an upland stream have been dated at 11,000 to 12,000 years
B.P.

(Figure 7-10, Radiocarbon Date 1).

The cut-and-fill sequence at

Tunica Bayou and along many of the upland streams of the surrounding
area matches the sequence of events at the Salt Mine Valley site

on

Avery Island (see Chapter VI, Figure 6-12).

Two important bits of data come from core samples.

In his impor-

tant 1974 paper, David E. Frazier lists a date from a wood and brackish-

marsh peat immediately beneath transgressive deposits obtained at

a depth of 35 meters

Date 4).

below present sea level (Figure 7-10, Radiocarbon

The sample yielded a date of 10,525

215 years B.P.

The

sample may be associated with the distal end of one of the delta lobes
of the Lafayette Delta Complex.

Another important radiocarbon date published by Frazier in the

same paper (1974) comes from the abandoned Ship Shoal Delta Lobe

323

(Figure 7-10, Radiocarbon Date 5).

Here, brackish marsh peat was

obtained at a depth of-43 meters and dated at 10,700 + 150 years B.P.
This date probably relates to a transgressive phase of the Ship Shoal

Delta Lobe.
The reconstructed paleogeography leads to the conclusion that sea

level was relatively stable during the interval from about 12,000 to

8,500 years B.P.

Furthermore, the level of the sea may not have been

as low as many workers have reported (20 to 45 meters).

In fact, the

extensive freshwater swamp surface suggests a level near that of the

present

stillstand.

The level subsequent to the time of the Layfayette

Delta Lobe and associated features is not easily recognizable, however,

since the surface has been tilted and displaced by faulting.

Part of

the surface now stands as a well-drained terrace (Prairie Terrace)

while the delta lobes have been subsided and are submerged on the continental shelf.
2)

Maringouin Complex

Early to Middle Archaic

About 8,500

years ago sea level started to fall from the high stand which occurred

during Paleo-Indian times.

resumed.

Following a short reversal a slow rise

By 7,000 years ago, it was approximately 10 to 15 meters be-

low its present level.

During this reversal, a series of stepped

terraces east of Lafayette was formed and meanders of the Mississippi

cut into the now-elevated and abandoned Lafayette meander belt

pronounced erosional scarp was formed and became the western wall of
the Late Holocene or modern alluvial valley.

A delta lobe developed

in the vicinity of present day Marsh Island, Vermilion, and Atchafalaya

Bays (Figure 7-12).

Shoals in the Gulf remain as remnants of the

324

WIW.

ifliii
111

V)
_l

V)
LU
i-

CO
(0

<
Q
Z

o
u.

o
03
CC

oO

(0

CO

<

u.

o
<
cc

H
<
t<r

<

\-

<
LL
1CL

<

o
z
<
CO

<
\_l
LLI

CO
UJ

OC

o
z

_l
_l

(0

3
o

<

co

LL

_l

UJ

(0

OL
QC

<

_l

<

(0

co

Illffl

wis

H
mm
fc

325

now-submerged delta lobe.

A number of Early to Middle Archaic sites

are associated with this development.

Of particular interest is the fact that the Eunice Swamp, the

vast Paleo-Indian interval overflow swamp, was gradually drained and

converted to a prairie and gallery forest setting.

Some lakes may have

Archaic peoples utilized the margins

persisted in poorly drained areas.

of these lakes.

The geological record suggests that very significant changes

occurred, and were apparently associated with the sea level reversal
that resulted in the initiation of the Maringouin Delta Complex.

tensive areas of coastal wetlands diminished in size.

Ex-

The radius of

curvature on meanders of relict stream scars along coastal plain streams,

such as the Pearl, Sabine, and Trinity, became larger during this interval, indicating higher flood discharges.

regime was more erratic.

This suggests that the flow

It was also during this interval that the

Pleistocene megafauna became extinct

This interval is one of the least

understood, and, possibly because of climatic implications which in turn

affected settlement pattern and absolute population size, one of the most
significant during Late Quaternary times from the standpoint of continental
shelf resource management.

While the present scope of work will not permit an in-depth treatment, it should be mentioned that a combined archeo logical-geological

approach could contribute greatly to our understanding of events that

occurred during this rather obseure

Sale-Cypremort Lobe of the Teche Complex

3)

Archaic

time period.

Middle to Late

The most recent or Late Holocene rise of sea level, begin-

ning about 6,000 years ago, drowned the Maringouin Delta Complex and
led into the modern succession of lobes.

326

During the interval between

approximately 6,000 and 4,000 years B. P. delta building continued in

f

the general area of the old Maringouin Complex. However, geological

data suggest

a distinctive new cycle or pulse of sedimentation,

probably related to a major delta lobe.

Archeological associations

with this lobe are poorly defined, but probably include the Mound
Point (16 IB 14) andRabbit Island sites (16 SMY 8).
A number of other important features was

associated with this

Among these was the formation of beach-ridge complexes re-

rise.

worked from channel and natural levee sands.

Fan-shaped complexes

at Pecan Island (Vermilion Parish) and Little Chenier (Vermilion

Parish) are particularly prominent.

Archaic projectile points from

the Pecan Island area are related to occupation of these features

when they were active.

Metairie Lobe of the St

4)

Tchef uncte Period

Bernard Complex:

Poverty Point and

As sea level approached and reached its present

stand (4,000 to 3,000 years ago), a major delta lobe began to develop
in what is presently the eastern part of the deltaic plain (Figure

7-13).

By Poverty Point times, the growing deltaic land mass had

created a marginal deltaic basin in the areas now occupied by Lakes

Maurepas and Pontchartrain.

Poverty Point peoples began utilizing

the highly productive environments associated with the growing delta

lobe.

Four Poverty Point sites have been found that were associated

with the lobe, two of which were located on natural levees of active
distributaries.

The delta lobe sites were all specialized fishing-

hunting-gathering camps and villages.

A major village or ceremonial center

was located on the eastern side of the Pearl River estuary (Claiborne
22 HC 35,

Hancock County, Mississippi).

During this interval,

Poverty Point peoples continued to use environments associated with

abandoned and deteriorating lobes of earlier delta developments.
327

CD

rC
4-1

M-l

o
>^
42
a.
cd

u
60
O

CO
<u

<D

6
o H
60

4J

a)

H
cO

CD

Pm

4-

u
d
3

PL.

4-1

CD

PQ 42

O
H
O
on T3

CM

rt

4J

4-1

O
o
o
o Pm
^
<f
v

4-1

M
>
X O
CI)

a)

P4

rH
#*

a.

6
o

CJ

H
cO

43
co
4-)

rH

O
u

<

0)

<d

4J

TJ

CO

S-i

rJ

CO

d
ft
0J

PQ

60

d
M
3

TJ

4-1

C/>

CO
Q)

O M

X4J

CO
cfl

4-1

4-1

rH

CD

CD

42

rH
0)

H
r4

H
0(X
H
CO
CO

H H
ad
4-1

CO
CO

ti
ro

CD

d
60
H

328

Summary and Conclusions

Application of an archeological-geological approach to the Mississippi

Delta area of the central Gulf coast has resulted in the interpretation
of the prehistoric landscape and settlement patterns presented in this
chapter.

While to some extent the hypothesis presented varies from

previous interpretations, suggesting that many features of the area are

considerably younger than generally believed, its value to the conti-

nental shelf problem is largely as a model.

Clearly, many continental

shelf features in this central Gulf area are of deltaic origin and the

known relationships which can be worked out in the onshore parts of the
coastal zone can be projected offshore.
There is clearly a continuity of features and presumably of
related archeological sites between the coastal zone and the cont-

nental shelf.

To a great extent, interpretation of the coastal zone

is the key to interpretation of the continental shelf.

The importance

of additional research in key coastal areas is self-evident.

Major vertical movements have differentially displaced relict

features of the central Gulf area during at least the past 12,000
years.

Surfaces deposited at or near the level of the sea have been

tilted, warped, and faulted subsequent to their formation.

Differen-

tial vertical movements of 50 meters or more may have influenced the

surface of the Lafayette Alluvial Valley and Deltaic Plain since its

abandonment approximately 8,500 years ago.

During the period from approximately 12,000 to 8,500 years B.P.
(Intervals HI to H3) there was a very slow rise or relative stillstand
of the sea.

Major coastal progradation occurred with associated

development of extensive coastal wetlands (swamps and marshes)

Pleistocene megafauna flourished and there is abundant evidence of

man throughout the area, particularly during the closing interval.
329

The climatic implications of the hypothesis for the period from

approximately 8,500 to 6,500 years B.P.

and I) are

(Intervals H
4

particularly significant.

Major geomorphic and ecological changes

occurred, including a reversal of the long-term trend of sea level

rise and stability.

This reversal caused a ria cycle and a general

displacement of the shore zone.

The extent of coastal wetlands was

greatly diminished and the Pleistocene megafauna became extinct.

Great fluctuations in runoff conditions and stream discharge occurred.

There were effects on settlement pattern and population size.

The

impacts of these changes are recorded in the archeological and geological

record throughout the study area.

During the past 6,000 years (Interval J to K)

slowly and attained a relative stillstand.

sea level has risen

Conditions have been rela-

tively stable and not greatly different from those of the present.

There is a continuity of the geological and archeological record during

this time.

330

CHAPTER VIII
CONCLUSIONS AND RECOMMENDATIONS FOR FURTHER STUDY
Summary

Cultural resources are not evenly or randomly distributed across the

earth's surface, but are clustered or concentrated in areas of human activThis is true not only for terrestrial areas, but also for the conti-

ity.

nental shelf.

We have tried to demonstrate that there is a relationship

between prehistoric human activity and landforms.

While it is not possible

to predict specific site locales, certain landforms were clearly favored for

habitation and other activities.

These preferred locations in respect to

landform geometry can be defined as areas of high probability of site occurrence.

We have also tried to demonstrate that relict landforms can be identified on the Gulf bottom.

The distribution of these relict landforms is re-

lated to former natural systems and has an orderly and predictable arrangement.

A relict landform can be shown to be not submerged during

particular period,

and favored for habitation by prehistoric peoples flourishing in nearby areas

during that period.

But although this gives us an area of high probability

for the occurrence of a site, it may be a rather large area that contains a

few rather small sites.

Discovery of sites on the continental shelf therefore becomes a matter of

prospecting.

Two general approaches to prospecting for sites are available.

The first involves underwater search by divers or archeologists in underwater

vehicles of some type.

While direct underwater search techniques may be

preferable in some cases, there are strict limitations imposed by water depth,
visibility, costs, and time factors.

The second general category involves

searching with remote-sensing tools from surface vessels.

For site discovery

and verification, remote-sensing techniques seem far more efficient and

promising.

Once high-probability areas have been defined, the next obvious

question is "what do we look for?"

331

Cultural Signatures
Our survey of the literature of the northern Gulf area, as presented
in Chapters 5 and 6, indicates that there are eleven site situations and

types of sites that occur frequently enough within the time intervals of

concern that they may be anticipated in the OCS area.

Table 8-1 presents

these eleven types and the probable periods of occurrence of each.

A summary

of the basic characteristics of each of the eleven site types follows.

These

basic characteristics are called "signatures" in analogy to the use of the phrase

"electronic signatures" in computer interpretation.

The totality of a site

can not be explored when it is on the ocean floor, and site determination must

be made on the basis of a few isolated and always limited indicators.

As is

true of electronic signatures, these cultural signatures can be positive or

negative.

They are discrete site indicators, capable of being detected by pre-

sently available instrumentation.

A few signatures of each type site situation

are offered below.

Quarry Sites
1)

Outcrops of chert, quartzite, or other rocks that are suitable for

chipping or grinding.
2)

Scattered debitage and chipped tools.

areas from approximately 1,000 to 200,000 m

3)

Most material occurs as surface lag.

Concentrations occur over

(1/4 acre to 50 acres).

Some stratification of chip-

ping floors may occur, particularly in the vicinity of rock outcrops.

Stra-

tification may also occur if there are ponds near the rock outcrops.
4)

Perishable materials are generally absent unless a pond or depres-

sion is associated with the rock outcrop.

Salt Dome Sites

1)

Around depressions (solution ponds) on dome surface or associated

with small streams on dome, on margin of dome adjacent to stream, lake, or

bay.

332

Table 8-1. Occurrence of site types by culture period or stage

4-1

CULTURE PERIOD OR STAGE

C
o

C
o

CO

rt

x:

Pm

x:

cd

a
u

<

J-l

rt

X)

MC

o
>-)

<:

<:
(D

Pu

cO

H
T3

CD

iH

u
CD

CO

,C

13
XI
H

a;

TYPE OF SITE

>
o
Pi

Quarry Sites

Salt Dome Sites

Spring (Sinkholes) Sites

Valley Margin Sites

Natural Levee Sites

Point Bar Sites

Bay Margin Sites

Coastal Dune Lake Sites

Shell Middens

Conical Earth Mounds

Crescentic and Circular

Villages

X - Known occurrence
P - Probable occurrence

333

CO

cO

iH

CD

co

S-i

PM

2)

Scattered debitage and stone tools, particularly if gravel or rock

is available on dome.
9

Artifacts may be locally concentrated in areas of

?

z
(10 acres).
(100 square feet) to 40,000 m

3)

Stratified deposits, including bone beds and organic deposits, may

occur in sinkholes or other depressions on the dome surface.

4)

Perishable material may be preserved if sinkholes or depressions are

present.

Spring Sites
1)

Around the margins and within sinkholes in karst areas.

2)

Scattered debitage and stone tools may occur in the vicinity of the

sinkhole rim.
3)

Artifacts and bone may be incorporated in deposits in the bottom of

the sinkhole.

Mineralized bone may be anticipated.

All material was dropped

in from the surface and is randomly distributed.

4)

Stratified deposits may be found on ledges around the inside walls

of the sinkhole.

These ledges may have served as shelters or burial places.

Preservation of bone and vegetable matter may be excellent due to the high

mineral content of the spring waters.

In addition to stone tools, artifacts

and ecofacts of bone, shell, wood, fiber and other perishables may be found.

Valley Margin Sites

1)

Found near the escarpment marking the boundary of a floodplain

stream valley.
deposits.

or

Distinguished by the escarpment and/or change in the nature of

Stream valleys may often be identified by bathymetry, sub-bottom

profiles, and/or cores.

Sites may occur in close proximity (within 100

meters) of the valley wall scarp.

2)

Types of sites may be variable, including small campsites, gravel

quarry sites, crescentic villages, shell middens, or earth mounds.

Most

typically, the sites may be marked by stone artifacts and debitage scattered
334

over the surface.

Most artifacts are manufactured from stream gravel.

Where

gravel was readily available in nearby streams, the site may take on char-

acteristics of a quarry site.

3)

Stratification is poorly developed and poorly preserved.

4)

Preservation is usually poor.

Only materials highly resistant to

chemical weathering can be expected to survive.

Natural Levee Sites

1)

Located near the crests of natural levee ridges.

Preferred loca-

tions include the cut bank side of meander loop cutoffs, junctions of

tributaries and distributaries, and/or crevasse distributaries.

Natural

levees are distinguishable by bathymetry and/or sub-bottom profiles.

2)

Sites may include camps, shell middens, conical earth mounds, and

crescentic villages.

Distributary levees are prime locations for shell

middens.
3)

Sites may be stratified and interbedded with overbank sediment

deposits.
4)

Preservation may be poor to moderate.

Some bone and shell may

survive, but usually only chemical-resistant materials.

Point Bar Sites

1)

Habitation of actively accreting point bars of a meandering stream

results in archeological materials being incorporated in the point bar deposits along sloping depositional surfaces.

Materials may remain exposed

on the crests of accretion ridges in a point bar sequence.

2)

If the stream was transporting gravel during habitation,

the site

may have the characteristics of a quarry, with abundant debitage and artifacts produced for gravel.
3)

Hearths may be present in the point bar.

Lateral stratification is related primarily to active sedimentary

processes of point bar accretion.

335

4)

Preservation of perishable materials will probably be very poor.

Usually only stone will be preserved.

Bay Margin Sites

1)

The margins of bays and estuaries may be typically marked by low

escarpments defined by bathymetry (if well preserved)

It can be anticipated

that sites will frequently be located near such escarpments, particularly

along sheltered parts of the bay.

Sites may range from small camps or

shell middens to very large mounds, shell middens, or crescentic villages.

2)

Shell middens are likely to be well represented around bay margins.

3)

Preservation is likely to be poor to moderate.

Preservation of

bone, shell, and seeds may be good in shell middens.

Coastal Dune Lake Sites

1)

Sites may be located around the margins of lakes and ponds in dune

fields.

Artifacts and ecofacts may be incorporated in the sedimentary de-

posits within such shallow water bodies.

In some instances, these depres-

sions may be identified from bathymetry, sub-bottom profiles, or cores.

and calcified organic sands often form in the freshwater ponds and

Humid

lakes of coastal dune fields.

Such deposits may be more resistant to ero-

sion than surrounding, unconsolidated dune fields and may produce inverted

topographic features on the sea bottom.

extent.
2)

Deposits should rarely be more than

meters thick.

Scattered stone tools and hearth stones may occur, but will rarely

be abundant.
3)

Site area may be a few acres in

Small shell middens may be associated.

Stone beds and plant materials may be associated with the humic

deposits or cemented pond deposits.

In such instances,

perishable materials may be very good.

336

preservation of

Shell Middens
1)

Sites may be associated with a variety of coastal landforms,

especially around margins of estuaries, lakes, bays, flood plains, on

natural levee crests, on relict beach ridges, salt domes, etc.

Shell mid-

dens may be identified on bathymetric charts and sub-bottom profiler

records and with coring or dredging samplers.

Side-scan sonar may have

limited effectiveness as well.

2)

Although scattered artifacts are possible, they are probably not

abundant unless midden material has been redistributed along the bottom.
Rangia cuneata and Crassostrea virginica are the most common shells,
but often other molluscs

are present in lesser numbers.

Midden

geometry may range from large, crescentic or circular villages covering

over 40,000 m
height.

(10 acres)

to small piles only

m in diameter and 30 cm in

The most common geometry is likely to be linear, cigar-shaped

ridges
3)

Stratigraphy is probably good if the midden is not scattered.

4)

Preservation is probably good if the site is in situ

terial may be locally or entirely cemented.

Midden ma-

Calcium carbonate forms the

cementing agent, and shells, bone, and artifacts may be contained in the

aggregate.

Conical Earth Mounds

1)

Sites may be located in any of the areas described for shell mid-

dens (see above).

If mounds should be preserved on the Gulf bottom,

they

may be located through bathymetry, sub-bottom profiles, and side-scan sonar,

2)

Mounds generally occur in groups of two.

50 cm or less to over 5 meters.

Height may vary between

Some mounds may be over 30 m in diameter.

Artifacts will probably be very scarce on an in situ mound, but may be

located among the remains of eroded or scattered mounds.

337

Typically early

mounds of Poverty Point, Late Archaic, and Middle Archaic times exhibit a

Possible village or shell midden areas

paucity of accompanying artifacts.

may be anticipated nearby.
3)

Stratigraphy is most likely excellent in a well-preserved mound.

4)

Preservation should be fair with usually only stone, baked clay,

and poorly preserved burials found.

Crescenticand Circular Villages

1)

Located along terrace margins overlooking a flood plain or marine

estuary, along the bay side of barrier islands, and on natural levee crests
of relict streams.

Terrace margins can usually be located by bathymetry

and/or sub-bottom profiles.

2)

Crescentic sites are located in the coastal Gulf area, while cir-

cular villages are found along the Atlantic coast, with the possibility that
Sites are composed of oyster and/or

they can occur in the eastern Gulf.

Rangia shells or black midden areas.

diameter and covering
as 0.03-0.06 km 2

Claiborne)

km

They range in size from 1Q00

m in

(Poverty Point Site) to as small

(500 acres)

(8-15 acres) with diameters of about 75-90 m (Cedarland,

The central area of the village is void of shell or midden

accumulation, with a shell "ring" from

m to

m high.

Commonly, there

are more than one or two villages close together (i.e., Cedarland and

Claiborne, and Sapelo Island shell rings).

3)

There is a good possibility that village areas also have an out-

lying conical earth mound or mounds associated with them.

Artifacts exhi-

bit evidence of wide trade network and mostly appear as chipped stone points,

steatite bowls, ground-stone plummets, exotic lithics, such as quartz crystals, and fiber-tempered pottery.

usually associated.

Baked clay, Poverty Point objects are

Artifacts are likely to be abundant.

338

4)

Stratification will probably be fair to good.

Lithif ication of mid-

den material may occur with bone, shell, and artifacts in the matrix.
5)

Preservation of organic matter is probably fair to good in shell

middens, poor in earth middens.

Remote Sensing Techniques

The characteristics presented above are, in effect, signatures which
must be identified by remote-sensing or surface-testing techniques in order
to discover and confirm a site.

At this point, we really do not know how

effective presently available remote-sensing techniques are in identifying

these signatures.

However, based on a working knowledge of the remote-

sensing techniques and the site characteristics, we can evaluate the techniques from the standpoint of relative effectiveness.
is presented in Table 8-2.

Such an evaluation

As indicated in the table, good bathymetry and

sub-bottom profiles are basic tools for identifying characteristic

of landforms and large or conspicuous sites.

geometry

Side-scan sonar may be of

some value in defining the geometry of a conspicuous site or conspicuous

landform.

Various types of sampling devices, specifically grab samples,

cores, and box samplers, may assist in verifying the interpretation of the

form by providing material from which the landform is composed, or the

matrix material of the site.

The recovery of artifacts or ecofacts can be

accomplished by direct sampling approaches.

These include drag samples,

bucket samples, cores, and box samples.

Only a very few sites in the continental shelf areas of the United States

have been found and investigated by direct diver inspection.

However, if we

review the reports on artifacts, sites, and ecofacts that have been found, several

interesting things appear.

sonal communication).

Most notable is the work of Ruppe in west Florida (per-

One shell bank, possible a midden, has been reported in deeper

339

Table 8-2. Effectiveness of remote sensing survey and testing tools

u
CD

CD

60

REMOTE SENSING
TECHNIQUE

O
u

4-1

CD

I
42

TYPE OF SITE

CO

&
o

t-l

CD

Pm

CD

CO
!-i

CD
4-1

4->

e
CO
o

CO

CD

s
O
4->

CD
CD

43

u
CD
u
o

b0
T3

M
CD

O
CJ

4-1

00

4-1

CD

CD

>-i

^i

>-i

G
o

Q
O
3

pq

4-1

CO

4-1

tO

CO

CO

PM

PQ

PQ

CJ

0-1

0-1

Point Bar Sites

0-1

Bay Margin Sites

0-1

Coastal Dune Lake Sites

Shell Middens

Conical Earth Mounds

Crescentic and Circular

Villages

3
CO

H
CO

Quarry Sites

Salt Dome Sites

Spring (Sinkhole) Sites

Valley Margin Sites

Natural Levee Sites

EFFECTIVENESS:
1 -

Virtually useless
Occasionally effective

- Highly effective

Limited effectiveness

340

sa

CO

PQ

U
CD
.-H

water off the Atlantic coast by Emery (1969)

A number of artifacts and fossil

bones has been recovered in fishing nets and geological bottom-sampling devices,

usually drag samplers.

This leads one to conclude that drag and bulk sam-

pling devices may be the most effective way of sampling and testing drowned
or submerged sites.

Given the present state of the art of OCS cultural

resource methodology and underwater remote-sensing and diving technology,

the most effective sequential approach to prehistoric site discovery seems
to be as follows:
1)

Interpretation of relict landforms and relict systems from large-

scale bathymetry, identification of high-probability areas from these bathy-

metric maps, interpretation of the age of the features and associated high-

probability areas, development of models for site types most likely to be

associated with the high-probability areas.
2)

Sampling and testing with remote-sensing techniques.

This can be

done either through a specific research project in a specific area or through

the present OCS survey requirement.

ing should be employed.

A hierachical or step approach to test-

The first array of instruments at Step

clude small-scale bathymetry (fathometer)

should in-

sub-bottom profile to 30 feet or

to define the upper 30 feet, and a drag sample or grab sample.

step survey indicates an anomaly or probable site, Step

If this first-

testing might in-

clude side-scan sonar, bottom core or cores, additional grab samples and

drag samples.

Step 3 inspection would include underwater photography or

television viewing, box core sampling, and/or direct inspection by divers.

Zone Map

We have attempted to apply the findings of this study to the problem

of survey requirements for the northern Gulf OCS area.

Based on the distri-

bution of relict landforms and our interpretation of the age of these landforms, the northern Gulf OCS area has been zoned.

341

The zone boundaries were

combined with those developed for shipwreck occurrence and evaluation (see

Volume II).

Plate 11 of Volume III presents the combined zonation showing

probable cultural resource occurrence for prehistoric sites and historic shipwrecks.

Zone 1 on Plate 11 is predicted to have high productivity for pre-

historic cultural remains dating back to and including Paleo-Indian, Early

Archaic, Middle Archaic, and Late Archaic periods 12,000 - 3,000 B.P.

In

the Chandeleur Sound area of southeastern Louisiana, sites as recent as 800

years B.P. may be submerged.

The seaward limit of Zone

of Paleo-Indian habitation sites.

indicates the extent

The seaward limit of Zone 3a approximates

the maximum low stand of sea level during the Wisconsin glacial stage.

The

zone contains relict shore features ranging in age from approximately 19,000
to 12,000 years B.P.

anticipated.

Habitation sites within the same time range can be

Zone 3b is a zone of banks which were probably exposed as

islands during the maximum Wisconsin low stand.

Prehistoric sites within

the age range of 19,000 to 17,000 years B.P. may occur.

Zone 4 exhibits

apparent shoreline features in water depths of 90 to 200

ra.

of the nature or age of these features at present.

Little is known

They may be related to

Illinoisian glacial stage.

While these zonations and recommendations were not arrived at lightly, it

should be emphasized that the map was developed through qualitative, graphic

approaches and should be characterized as a "best judgment" map.

Clearly,

the methodology developed in this volume is untested in the OCS area.

do believe, bowever,

We

that the rationale is sound and can be demonstrated in

the coastal plain of the northern Gulf, and that the methodology is amenable
to testing.

Recommendations For Further Study

The methodology should be tested in pilot-study areas both onshore
and offshore.

Major objectives of the onshore work should be to:

342

1)

better

understand Late Quaternary geologic events and landforms; 2) test the "high-

probability area" methodology and compare with random sampling, transects,

and other sampling designs, 3) better define early cultures which may be

represented in the OCS area; 4) better understand site-landf orm geometry

relationships, and 5) better define site signatures.

These objectives can

best be achieved in a terrestrial situation.

The OCS pilot study that is needed now is simply a more detailed application of the methodology developed in this paper.
1)

The objectives would be:

study the bathymetry and available geological data at a scale and level

of detail that was not possible in the present study; 2) map relict features

and paleogeography and identify high-probability areas of site occurrences;

3)

using remote-sensing techniques, systematically survey high-probability

areas; 4) using the step approach discussed above, attempt to verify and
test possible sites; 5) if feasible,

divers.

inspect and test sites using

Such a program should result in a significant improvement in defining

the survey requirements from the standpoints of both efficiency and effective-

ness.

In addition,

the fascinating study of the prehistoric archeology of

the outer continental shelf would be advanced.

343

REFERENCES

Adam, D. P.
1975a. The Tropical Cyclone as a Global Stabilizing Mechanism.
Geology, Volume 3, pp. 625-626.
1975b.

Ice Ages and the Thermal Equilibrium of the Earth, III.

Quar ternary Research, Volume 5, pp. 161-171.

Alt, D. and H. K. Brooks

Age of Florida Marine Terraces.
1965.
Volume 73, pp. 406-411.

Journal of Geology,

Ambler, R. J.
Three Prehistoric Sites near Cedar Bayou, Galveston Bay
1967.
Area, Texas.
State Building Commission Archeology Program, Report No. 8.
The Lost River Phase of the Lower Trinity River Delta.
Manuscript submitted to the National Park Service, Southwestern Region, by Texas Archeological Survey. Austin:
University of Texas.

1973.

Aten, L. E.
Excavations at the Jamison Site (41 LB 2), Liberty County,
1967.
Texas.
Houston: Houston Archeological Society Report
No.

1.

Archeological Excavations at the Dow-Cleaver Site, Brazoria County Texas.

Texas Archeological Salvage Project,
Bulletin No. 1.

1971.

Aten, L. E. and C. N. Bollich

1969.
A Preliminary Report on the Development of a Ceramic
Chronology for the Sabine Lake Area of Texas and Louisiana.
Texas Archeological Society Bulletin, Volume 40,
241-258.
pp.
Bader, R.
1957.

S.

Two Pleistocene Mammalian Faunas from Alachua County,

Florida.
Florida State Museum Bulletin, Volume 2, pp.
52-75.

Bergantino, R. N.
Submarine Regional Geomorphology of the Gulf of Mexico.
1971.
Bulletin of the Geological Society of America, Volume 82,
pp. 741-752.
Bernard, H. A.
Quaternary Geology of Southeast Texas. Unpublished Ph.D.
1950.
dissertation on file at Louisiana State University at
Baton Rouge.
165 pp.

344

Bernard, H. A. and R. J. Leblanc

Resume of the Quaternary Geology of the Northwestern Gulf
1965.
In, H. G. Wright, Jr., and D. G. Frey (eds.) The
of Mexico.
Quaternary History of the United States. Princeton: Princeton University Press,
pp. 137-185.

Bernard, H. A., C. F. Major, Jr., B. S. Parrott, and R. J. LeBlanc Sr.

Recent Sediments of Southeast Texas: A Field Guide to the
1970.
Brazos Alluvial and Deltaic Plains and the Galveston Barrier Island Complex.
University of Texas, Bureau of Economics Geology Guidebook II.
,

Bird, E. C. F.
Coasts.
1973.

Cambridge:

M.I.T. Press.

Bloom, A. L.
1967,

Pleistocene Shorelines: A New Test of Isostasy.

Bulletin
of the Geological Society of America, Vol.78, pp. 1477-1494.

Brooks, H. K.
The Plio-Pleistocene of Florida in Late Cenozoic Strati1968.
graphy of Southern Florida: A Reappraisal.
In, R. D.
Perkins (ed.), Miami Geological Society, pp. 4-42.
1973.

Geological Oceanography.
In, A Summary of Knowledge of
the Eastern Gulf of Mexico, 1973.
Coordinated by the State
University System of Florida, Institute of Oceanography.

1974.

General and Geotechnical Characteristics of the Structural Deposits on the West Florida Shelf.
Proceedings of
Marine Implications of Offshore Drilling, Eastern Gulf of
Mexico.
Published by the University of Southern Florida
and the Florida Department of Natural Resources, pp. 301308.

Bryant, W. R.
A. A. Meyerhof f , N. K. Brown, Jr., M. A. Furrer, T. E.
Pyle, and J. W. Antoine
1969.
Escarpments, Reef Trends, and Diapiric Structures, Eastern
Gulf of Mexico. Bull. American Association Petroleum Geologists
Volume 53:12, pp. 2506-2542.
Bullen, R. P.
1958.
Six Sites near the Chattahoochee River in the Jim Woodruff
Reservoir Area, Florida.
Bureau of American Ethnology,
River Basin Survey Papers No. 14, Bulletin No. 169.
,

Bullen, R. P. and E. M. Dolan

1959.
The Johnson Lake Site, Marion County, Florida.
Anthropologist, Volume 12, No. 4.

Campbell, T.
1947.

Florida

N.

The Johnson Site:

Type Site of the Aransas Focus of the
Texas Coast. Abilene:
Texas Archaeological and Paleontological Society Bulletin, Volume 18, pp. 40-76.

345

Clausen, C
1972.

Underwater Excavations at Little Salt Spring and Warm

Mineral Springs, Florida. Paper presented to the Society
of American Archaeology, Bal Harbour, Florida.

Clausen, C. J., H. K. Brooks, and A. B. Wesolowsky

The Early Man Site at Warm Mineral Springs, Florida. Boston:
1975.
Journal of Field Archaeology, Volume 2.

Coleman, J. M.
Recent Coastal Sedimentation:
Central Louisiana Coast.
1966.
Coastal Studies Institute Report No. 17.
Baton Rouge:
Louisiana State University.
Coleman, J. M. and S. M. Gagliano
Cyclic Sedimentation in the Mississippi River Deltaic Plain.
1964.
Transactions of the Gulf Coast Association of Geological
Societies, Volume 14, pp. 67-80.
1965.

Sedimentary Structures: Mississippi River Deltaic Plain.

In, Primary Sedimentary Structures and Their Hydronamic
Interpretation.
Society of Economic Paleontologists and
Mineralogists, Special Publication No. 12, pp. 133-148.

Coleman, J. M. and W. G. Smith

1964.
Late Recent Rise of Sea Level. Bulletin of the Geological
Society of America, Volume 75, pp. 833-840.

Collins, H. B., Jr.

1927.
Archaeological Work in Louisiana and Mississippi. Explorations and Fieldwork of The Smithsonian Institution
in 1926.
Washington, D.C:
Smithsonian Miscellaneous
Collections, Volume 78, No. 7.
1941.

Relationships of an Early Cranial Series from Louisiana.

Journal of the Washington Academy of Science, Volume 31,
pp. 145-155.

Cooke, C. W. and S. Mossom

1929.
Geology of Florida.
Florida Geological Survey, Twentieth
Annual Report, State Geological Survey, pp. 29-288.

Corpus Christi Geological Society

1958.
Sedimentology of South Texas.
Trip.
May 15-17, 1958.

Guidebook, Annual Field

Crook, W. W. , Jr. and R. K. Harris

A Pleistocene Campsite near Lewisville, Texas.
1958.
Antiquity, Volume 23, No. 3, pp. 233-246.

American

Curray, J. R.
1960.
Sediments and History of Holocene Transgression, Continental
Shelf, Northwest Gulf of Mexico.
In, F. P. Shepard, E. B.
Phleger, and T. H. van Andel (eds.), Recent Sediments, Northwest Gulf of Mexico. Tulsa: American Association of Petroleum Geographers.

346

1961.

Bulletin of the
Late Quaternary Sea Level: A Discussion.
Geological Society of America, Volume 17, pp. 1707-1712.

1965.

Late Quaternary History, Continental Shelves of the United

In, H. E. Wright and D. G. Frey (eds.), The
States.
Quaternary History of the United States. Princeton:
Princeton University Press, pp. 723-736.
11 figures.

Dickerson, M. W.
Report at Meeting of October 6, 1846. Philadelphia:
1846.
Proceedings of the Academy of Natural Sciences of Philadelphia,
Volume 3, No. 5, pp. 106-107.
Donaldson, A. C., R. H. Martin, and W. H. Kanes
Holocene Guadalupe Delta of Texas Gulf Coast.
1970.
In, J. P.
Morgan and R. H. Shaver (eds.), Deltaic Sedimentation,
Modern and Ancient.
Society of Economic Paleontologists
and Mineralogists Publication No. 15, pp. 107-137.
Edwards, G. Serpell
1971.
The Geology of West Flower Garden Bank.
TAMU Sea Grant Program, Publication 71-215, 199 pp., 47 figures, 6 tables.

Fairbridge, R. W.
1961.
Eustatic Changes in Sea Level.
In, L. H. Ahrens e_t al. (eds),
Physics and Chemistry of the Earth, Volume 4. New York:
Pergamon Press, Inc., pp. 99-185.
1974.

The Holocene Sea Level Record in South Florida.

In, P. J.
Gleason (ed.), Environments of South Florida: Past and
Present.
Miami Geological Society Memoir No. 2, pp. 223231.

Fisher, W. L., L. F. Brown, Jr., J. H. McGowen, and C. G. Groat

1973.
Environmental Geologic Atlas of the Texas Coastal Zone:
Beaumont - Port Arthur Area. Austin:
Bureau of Economic
Geology, University of Texas.
Fisher, W. L.
J. H. McGowen, L. F. Brown, Jr., and C. G. Groat
1972.
Environmental Geologic Atlas of the Texas Coastal Zone:
Galveston - Houston Area. Austin: Bureau of Economic
Geology, University of Texas.
,

Fisk, H. N.
1939.
Depositional Terrace Slopes in Louisiana,
morphology, Volume 2, pp. 181-199.

Journal of Geo-

1944.

Geological Investigation of the Alluvial Valley of the Lower

Mississippi River. Vicksburg: U. S. Army Corps of Engineers Mississippi River Commission.

1955.

Rome:
Sand Facies of Recent Mississippi Delta Deposits.
Proceedings of the Fourth World Petroleum Congress, Section
1,

1956.

pp.

377-398.

Nearsurface Sediments of the Continental Shelf off Louisiana.

Austin:
Proceedings of the Eighth Texas Conference on Soil
Mechanics and Foundation Engineering, September 14 and 15.
347

Padre Island and the Laguna Madre Flats, Coastal South Texas.
Baton Rouge:
Second Coastal Geography Conference, Coastal
Studies Institute, University State University, pp. 103-151.

1959.

Fisk, H. N. and E. McFarlan, Jr.

Late Quaternary Deltaic Deposits of the Mississippi River.
1955.
Special paper of the Geological Society of America, Paper
No. 62, pp. 279-302.
Ford, J. A. and G. I. Quimby
The Tchefuncte Culture, An Early Occupation of the Lower
1945.
Mississippi Valley. Menasha: Memoir of the Society for
American Archaeology No. 2.

Foxworth, R. D., R. R. Priddy, W. B. Johnson, and W. S. Moore

1962.
Heavy Minerals of Sand from Recent Beaches of the Gulf Coast
Mississippi State
of Mississippi and Associated Islands.
Geological Survey Bulletin No. 93.
92 p.
Frazier, D. E.
1967.
Recent Deltaic Deposits of the Mississippi River: Their
Development and Chronology. Transactions of the Gulf Coast
Association of Geological Societies, Volume 17, pp. 287-315.

Depositional Episodes: Their Relationship to the Quaternary

Stratigraphic Framework in the Northwestern Portion of the
Gulf Basin.
Bureau of Economic Geology, Geology Circular,
Volume 74, No. 1.
28 p.

1974.

Gagliano,

S.

M.

1963.

A Survey of Preceramic Occupations in Portions of South

Louisiana and South Mississippi. Florida Anthropologist,
Volume 16, No. 4, pp. 105-132.

1967.

Occupation Sequence at Avery Island. Coastal Studies Institute Report No. 22.
Baton Rouge: Louisiana State University,

1969.

Evidence of Early Man in the Central Gulf Coast. Paper

presented at the annual meeting of the Society of American
Archaeology, May 1, 1969. Milwaukee.

1970.

Archaeological and Geological Studies at Avery Island, 19681970.

Baton Rouge: Coastal Studies Institute, Louisiana
State University.

Gagliano,

S.

1965.

Gagliano,
1963.

S.

M. and H. Gregory
Paleo-Indian Projectile Points in Louisiana.
Studies Service, Volume 4, pp. 62-67.
M. and Roger T. Saucier
Poverty Point Sites in Southeastern Louisiana.
Antiquity, Volume 28, pp. 320-327.

348

Louisiana

American

Gagliano,
1967.

S.

M. and B. G. Thorn
Deweyville Terrace, Gulf and Atlantic Coasts. Coastal
Studies Institute, Bulletin No. 1.
Louisiana
Baton Rouge:
State University, pp. 23-41.

Gagliano,

S.

M. and J. L. van Beek

Geologic and Geomorphic Aspects of Deltaic Processes, Mississippi Delta System.
In, Hydrologic and Geologic Studies
of Coastal Louisiana, Report No. 1.
Baton Rouge:
Center
for Wetland Resources, Louisiana State University.
140 p.

1970.

An Approach to Multi-Use Management in the Mississippi

Delta System.
In, M. L. Broussard (ed.), Deltas, Models
for Exploration.
Houston:
Houston Geological Society,
pp. 223-228.

1975.

Gagliano,

S.

1970.

M. and C. H. Webb
Archaic-Poverty Point Transition at the Pearl River Mouth.
Paper presented at the 26th annual meeting of the Southeastern Archaeological Conference, November 15, 1969.

Macon, Georgia.

Gagliano,
1975.

S.

M.
R. A. Weinstein, and E. K. Burden
Archeological Investigations along the Gulf Intracoastal
Waterway: Coastal Louisiana Area.
Report submitted to the
New Orleans District, U. S. Army Corps of Engineers.
,

Gater, W. L.
Science.
1976.

(In press.)

Ginsburg, R. N,
In, R. N. Ginsburg
1964.
Introduction to Recent Sedimentation.
Coral Gables:
(ed.), South Florida Carbonate Sediments.
Institute of Marine Science, University of Florida, pp. 4-11,
Gould, H. R.
1970.
The Mississippi Delta Complex.
In, J. P. Morgan and R. H.
Shaver (eds.), Deltaic Sedimentation, Modern and Ancient.
Society of Economic Paleontologists and Mineralogists,
Special Publication No. 15, pp. 3-30.
Gould, H. R. and J. P. Morgan
1962.
Field Trip No. 9, Coastal Louisiana Swamps and Marshlands.
In, E. H. Rainwater and R. P. Zingula (eds.), Geology of
the Gulf Coast and Central Texas and Guidebook Excursions.
Houston: Houston Geological Society, pp. 287-341.
Gould, H. R. and R. H. Stewart, Jr.
1955.
Continental Terrace Sediments in the Northeastern Gulf of
Mexico.
In, J. L. Hough and H. W. Menard (eds.), Finding
Ancient Shorlines.
Society of Economic Paleontologists and
Mineralogists, Special Publication No. 3, pp. 2-20.

349

Grabau, A. W.
Principles of Stratigraphy.
1960.
pp. 384-466.

New York:

Dover Publications,

Graf, C. H.
The Late Pleistocene Ingleside Barrier Trend, Texas and
1966.
Louisiana. Unpublished thesis on file at Rice University.

Griffin, G. M.
Regional Clay-Mineral Facies - Products of Weathering In1962.
tensity and Current Distribution in the Northeastern Gulf
of Mexico.
Bulletin of the Geological Society of America,
Volume 73, No. 2, pp. 737-768.
Haag, W. G.
The Archaic of the Lower Mississippi Valley.
1961.
Antiquity, Volume 26, No. 3, pp. 317-323.

Hammond, A. L.
Paleoclimate: Ice Age Earth Was Cool and Dry.
1976.
p.

American

Science,

455.

Harvey, E. J. and J. L. Nichols

Stratigraphy of the Quaternary and Upper Tertiary of the
1960.
In, Guidebook, Cenozoic
Pascagoula Valley, Mississippi.
Field Trips.
Biloxi:
Gulf Coast Association of Geological
Societies, pp. 9-22.
Haynes, C. V.
Jr.
1964.
Fluted Projectile Points: Their Age and Dispersion.
Volume 145, (September 25).
,

Hemmings

1973.

Science,

Recent Excavations at Silver Springs, Marion County, Florida.

Paper presented at the 25th annual meeting of the Florida
Anthropological Society. Jacksonville.

Hemmings, E. T. and Kohler

Tallahassee:
The Lake Kanapaha Site in North Central Florida.
1974.
Bureau of Historic Sites and Properties, Bulletin No. 4, pp.
45-64.
Division of Archives, History, and Records Management,

Hoffmeister, J. E., J. I. Jones, J. D. Milliman, D. R. Moore, and H. G.

Multer
Living and Fossil Reef Types of South Florida. Coral Gables:
1964.
Institute of Marine Science, University of Miami.
Hoffmeister, J. E. and H. G, Multer
Pleistocene Limestones of the Florida Keys.
In, R. N. Gins1964.
burg (ed.), South Florida Carbonate Sediments.
Coral Gables
Institute of Marine Science, University of Miami.

350

K. W. Stockmein, and H. G. Multer

Hoffmeister, J. E.
Miami Limestone of Florida and Its Recent Bahamian Counter1967.
Bulletin of the Geological Society of America, Volume
part.
72, pp. 1051-1058.
,

Holmes, Charles W.
Distribution of Selected Elements in Surficial Marine Sediments
1973.
of the Northern Gulf of Mexico Continental Shelf and Slope.
Geological Survey Professional Paper 814, Washington, 7 p.
4 plates,
,

Honea, K. H.
The Bipolar Flaking Technique in Texas and New Mexico.
1965.
Bulletin of the Texas Archeological Society, Volume 36, pp.
259-267.

Hunter, R. E. and K. A. Dickenson

Map Showing Landforms and Sedimentary Deposits of the Padre
1970.
Island Portion of the South Bird Island, 7-1/2-Minute Quadrangle, Texas.
U. S. Geological Survey, Miscellaneous Inventory Map 1-659.
Hyne, N. J. and H. G. Goodell
1967.
Submarine Geomorphology of the Inner Continental Shelf off
Choctawhatchee Bay, Florida. Marine Geology, Volume 5, pp.
299-313.

Jervey, M.

Transportation and Dispersal of Biogenic Material in the

Ph.D. dissertation on file
Near shore Marine Environment.
at Louisiana State University at Baton Rouge.
205 p.

1974.

Jon, J. W.
1972.

Jr., et al.

Environmental Geology and Hydrology, Tallahassee Area,

Special Publication No. 16, State of Florida,
Florida.
Department of Natural Resources.

Jones, P. H.
Hydrology of Neogene Deposits in the Northern Gulf of Mexico
1969.
Louisiana Water Resources Research Institute BulleBasin.
Baton Rouge: Louisiana State University.
105 p.
tin GT-2.
Jones, P. H., A. N. Turcan, and H. E. Skibitzke
1954.
Geology and Groundwater Resources of Southwestern Louisiana.
Louisiana Geological Survey Bulletin No. 30. Baton Rouge:
Department of Conservation.
Joor, J. F., M.D.
Notes on a Collection of Archaeological and Geological Spec1895.
imens Collected on a Trip to Avery Island (Petit Anse)
February 1, 1890. American^ Naturalist , Volume 29, pp. 394398.

Jordan, G. F.
Bulle1951.
Continental Slope off Apalachicola River, Florida.
tin of the American Association of Petroleum Geologists,
Volume 35, No. 9, pp. 1979-1993.
1952.

Reef Formation in the Gulf of Mexico off Apalachicola Bay,

Florida.
Bulletin of the Geological Society of America,
Volume 63, No. 7 pp 741-744.
,

351

R. J. Malloy, and J. W. Kofoed

Jordan, G. F.
In,
Bathymetry and Geology of Pour tales Terrace, Florida.
1964.
R. N. Ginsburg (ed.), South Florida Carbonate Marine Sediments. Coral Gables:
Institute of Marine Science, University of Miami, pp. 68-69.
,

Jordan, G. F. and H. B. Steward, Jr.

Continental Slope off Southwest Florida. Bulletin of the
1959.
American Association of Petroleum Geologists, Volume 35,
No. 9, pp. 1978-1993.
Kanes, W. H.
Facies and Development of the Colorado River Delta in
1970.
Texas.
In, J. P. Morgan and R. H. Shaver (eds.), Deltaic
Sedimentation, Modern and Ancient.
Society of Economic
Paleontologists and Mineralogists, Special Publication No.
15, pp. 78-106.

Karkins, K.
1970.

The Fish Creek Site, Hillsboro County, Florida.

Anthropologist, Volume 23, No. 2, pp. 62-80.

Florida

Kniffen, F. B.
1935.

A Preliminary Report on the Mounds and Middens of Plaquemines and St. Bernard Parishes, Lower Mississippi River
Delta. Louisiana Geological Survey Bulletin No. 8. Baton
Rouge:
Department of Conservation, pp. 407-422.

Kolb, C. R., F. L. Smith, and R. C. Silva

1975.
Pleistocene Sediments of the New Orleans-Lake Pontchartrain Area.
U. S. Army Waterways Experiment Station,
Technical Report No. S-75-6.
7 p., 49 pi.

Kolb, C. R. and J. R. van Lopik

Geology of the Mississippi River Deltaic Plain, South1958.
eastern Louisiana. U. S. Army Waterways Experiment Station, Technical Reports 3-483 and 3-484.
2 Volumes.

Krieger, A. D.
Early Man in the New World.
1964.
In, J. D. Jennings and E.
Norbeck (eds.), Prehistoric Man in the New World. Chicago:
University of Chicago Press.

Kwon

H J
1969.
.

Barrier Islands of the Northern Gulf of Mexico Coast: Sediment Source and Development.
Coastal Studies Institute
Technical Report No. 75.
Baton Rouge: Louisiana State
University.

Lazarus, W. C.
1958.
A Poverty Point Complex in Florida.
gist, Volume 11, No. 1, pp. 23-32.

Florida Anthropolo-

1965a. Alligator Lake: A Ceramic Horizon Site on the Northwest

Florida Anthropologist, Volume 18, No.
Coast of Florida.
2, pp. 83-124.
352

1965b.

LeBlanc,

Significance of Dimensions of Big Sandy I-Like Projectile

Points in Northwest Florida.
Florida Anthropologist, Volume
18, No. 3, pp. 187-199.

R.

1972.

Leidy, J.
1866.

Geometry of Sandstone Reservoir Bodies in Underground Waste

Management and Environmental Implications. American Association of Petroleum Geologists Memoir No. 18, pp. 133-190.

Remarks. Philadelphia:
Proceedings of the Academy of Natural Sciences of Philadelphia, p. 109.

1899a. Notice of Some Fossil Human Bones.

Transactions of the
Wagner Free Institute of Science of Philadelphia, Volume
2, pp. 9-12.
1899b. Notice of Some Mammalian Remains from the Salt Mines of
Petit Anse, Louisiana.
Transactions of the Wagner Free Institute of Science of Philadelphia, Volume 2, pp. 33-40.

Lindall, W. L. et al.
Louisiana Coastal Zone: Analysis of Resources and Resource
1971.
Management in Connection with Estuarine Ecology. Florida
National Marine Fisheries Service.
323 p.
Lindsay, J. F.
T. H. Shipley, and J. L. Worzel
1975.
Role of Canyons in the Growth of the Campeche Escarpment.
Geology, Volume 3, pp. 533-36.
,

Logan, B. W. e_t al
1969.
Carbonate Sediments and Reefs, Yucatan Shelf, Mexico. American Association of Petroleum Geologists Memoir No. 11,
pp. 1-198.
.

Ludwick, J. C.
In, R. L. Miller
1964.
Sediments of Northeastern Gulf of Mexico.
(ed.), Papers in Marine Geology, Shepard Commemorative Volume.
New York: Macmillan Co. pp 204-238.
,

Ludwick, J. C. and W. R. Walton

Shelf Edge Calcareous Prominences in Northeastern Gulf of
1957.
Bulletin of the American Association of Petroleum
Mexico.
Geologists, Volume 41, No. 9, pp. 2054-2101.

Lundelius, E. L.
Jr.
1972.
Fossil Vertebrates from the Late Pleistocene Ingleside Fauna
Bureau of Economic
Austin:
San Patricio County, Texas.
Geology, Report of Investigation No. 77, University of Texas,
,

McFarlan, E.
1961.

Jr.

Radiocarbon Dating of Late Quaternary Deposits, South Louisiana.

Bulletin of the American Association of Petroleum
Geologists, Volume 72, No. 1, pp. 129-158.
9 figures.

353

Mclntire, W. G.
Prehistoric Indian Settlements of the Changing Mississippi
1958.
River Delta.
Coastal Studies Series Report No. 1.
Baton
Rouge:
Louisiana State University.

Methods of Correlating Cultural Remains with Stages of

In, R. J. Russell (ed.), Second Coastal
Coastal Development.
Geography Conference, pp. 341-359.

1959.

Mclntyre, A. and Members of the Climat Project

1976. Surface of the Ice Age Earth, Science 191, pp. 1131-37.
MacDonald, G. F.
1968.
Debert: A Paleo-Indian Site in Central Nova Scotia.
207 p.
thropology Papers No. 16, Museums of Canada.
MacNeil,

An-

S.

Pleistocene Shorelines in Florida and Georgia. U. S. Geological Survey Professional Paper 221-F, pp. 95-107.

1950.

MacNeish, R,
1976.

Macintyre,
1975.

S.

Early Man in the New World.

No. 3 (May and June).

American Scientist, Volume 64,

B. W. Blackwelder
L. S. Land, and R. Stuckenrath
G.
North Carolina Shelf Edge Sandstone: Age, Environment of
Origin, and Relationship to Pre-Existing Sea Levels.
Bulletin of the Geological Society of America, Volume 86,
pp. 1073-1078.

I.

Martens, J. H. C.
1936.
Beaches of Florida. Florida State Geological Survey, 21st
and 22nd annual reports, pp. 67-119.
Martin, P. S., G. I. Quimby, and D. Collier
Indians Before Columbus: Twenty Thousand Years of North
1947.
American History Revealed by Archaeology. Chicago: University of Chicago Press.
Mason, R, J.
The Paleo-Indian Tradition in Eastern North America.
1962.
Anthropology, Volume 3, No. 3, pp. 227-246.

Current

Mattison, G. C.
Bottom Configuration in the Gulf of Mexico. Journal of
1948.
Coastal and Geodetic Survey, Volume 1, pp. 78-82.
Moore, C, B.
Journal of
Some Aboriginal Sites in Louisiana and Arkansas.
1913.
Natural Sciences of Philadelphia, Volume 16, pp. 7-99.

Mickelson, D. M. and E. B. Evenson

Pre-Two Creekan Age of the Type Valders Till, Wisconsin.
1975.
ology, Volume 3, pp. 587-590.

354

Ge-

Milliman, J. D. and K. 0. Emery

Sea Levels during the Past 35,000 Years.
1968.
162, pp. 1121-1123.

Science, Volume

Moran, J. M.
Return of the Ice Age and Drought in Peninsular Florida.
1975.
Geology, Volume 3, pp. 695-696.

Morgan, J. P., J. M. Coleman, and S. M. Gagliano

1968a. Mudlumps at the Mouth of South Pass, Mississippi River:
Sedimentology, Origin, and Relation to Deltaic Processes.
Coastal Studies Series No. 10. Baton Rouge: Louisiana
State University.
190 p.
1968b. Mudlumps: Diapiric Structures in Mississippi Delta SediAmerican Association of Petroleum Geologists Memoir
ments.
No.

8.

Muller-Beck, H.
1966.
Paleohunters in America: Origins and Diffusion.
Volume 152 (May 27).

Science,

Neil, W. T.
A Stratified Early Site at Silver Springs, Florida.
1958.
Anthropologist, Volume 16, No. 4, pp. 99-104.

Florida

The Association of Suwanee Points and Extinct Animals in

Florida.
Florida Anthropologist, Volume 17, No. 1, p. 17.

1964.

Nelson, H. T.
and E. T. Bray
Stratigraphy and History of the Holocene Sediments in the
1970.
Sabine-High Island Area, Gulf of Mexico.
In, J. P. Morgan
and R. H. Shaver (eds.), Deltaic Sedimentation, Modern and
Ancient.
Society of Economic Paleontologists and Mineralogists, Special Publication No. 15, pp. 48-77.
13 figures,
,

pi.

Neumann, A.
The Configuration and Sediments of Stetson Bank, Northwestern
Texas
Gulf of Mexico.
Project 24, Section VII, Res. Rept.
A & M University, Department of Oceanography and Meteorology.
125 p. , 18 figs.
5 tbs.

1958.

Neumann, A. C. and W. S. Moore

1975.
Sea Level Events and Pleistocene Coral Ages in the Northern
Bahamas.
Quaternary Research, Volume 5, pp. 215-224.

Newman

1968.

Coastal Processes.
In, R. W. Fairbridge (ed.), The Encyclopedia of Geomorphology. New York: Reinhold Book Corp., pp,
150-155.

Osmond, J. K.
J. P. May, and W. F. Tanner
1970.
Age of the Cape Kennedy Barrier-and-Lagoon Complex,
of Geophysical Research, Volume 75, pp. 469-479.
,

355

Journal

Otvos, E. G.
1972a. Pre-Sangamon Beach Ridges along the Northeastern Gulf Coast Fact or Fiction? Transactions of the Gulf Coast Association
of Geologic Societies, Volume 22, pp. 223-228.
1972b. Mississippi Gulf Coast Pleistocene Beach Barriers and the Age
Problem of the Atlantic-Gulf Coast "Pamlico"-"Ingleside" Beach
Ridge System.
Southeastern Geology, Volume 14, No. 4, pp.
241-250.
1973.

Geology of the Mississippi-Alabama Coastal Area and Nearshore Zone. New Orleans Geological Society Field Trip
Guidebook.
67 p.

1975a. Ingleside Strandplain Sand of Central Texas Coast: Discussion.

Bulletin of the American Association of Petroleum Geologists,
Volume 59, No. 12 (December).
1975b. Southern Limits of Pleistocene Loess, Mississippi Valley.
Southeastern Geology, Volume 17, No. 1 (September).

Owen

R
1863.

TransacOn a Quar ternary Rock Salt Deposit in Louisiana.

tions of the St. Louis Academy of Sciences, Volume 2, pp.
250-252.

Pampe, W. R.
A New Pleistocene Marine Fossil Locality in Chambers
1971.
Bulletin of the Gulf Coast Association of
County, Texas.
Geological Societies, Volume 21, pp. 395-410.

Parker,

R.

1960.

H.

Ecology and Distributional Patterns of Marine Macroinvertebrates, Northern Gulf of Mexico.

In, E. P. Shepard e_t al
Bulle(eds.), Recent Sediments, Northwest Gulf of Mexico.
tin of the American Association of Petroleum Geologists,
pp. 302-337.
.

Parker, R. H. and J. R. Cur ray

Fauna and Bathymetry of Banks on Continental Shelf, North1956.
west Gulf of Mexico. Bulletin of the American Association
of Petroleum Geologists, Volume 40, No. 10, pp. 2428-2439.
6 figs.
1 pi.
,

Pearson, F. J., E. M. Davis, M. A. Famere, and R. W. Johnstone

University of Texas Radiocarbon Dates, III. Radiocarbon
1965.
Volume 7, pp. 296-314.

Phillips, Philip
Archaeological Survey in the Lower Yazoo Basin, Mississippi,
1970.
Papers of the Peabody Museum of Archaeology and
1949-1955.
Cambridge:
Harvard University Press.
Ethnology, Volume 60.
2 Volumes.

356

Poag, C. W.
1972.

Shelf-Edge Submarine Banks in the Gulf of Mexico: Paleoecology and Biostratigraphy

Transactions of the Gulf Coast
Association of Geological Societies, Volume 22, pp. 267-287.
.

19 figs.

pis.

Late Quaternary Sea Levels in the Gulf of Mexico.

Transactions of the Gulf Coast Association of Geological Societies,
Volume 23, pp. 394-400.

1973.

Poag, W. C. and W. E. Sweet

1972.
Geology of Claypile Bank, Texas Continental Shelf.
In, V.
J. Henry and R. Rezak (eds. ), Contributions on the Geological
and Geophysical Oceanography of the Gulf of Mexico.
Texas
A & M University Oceanographic Studies, Volume 3, pp. 223-261,
20 figs.

Price, W. A.
The Role of Diastrophism in the Topography of the Corpus
1933.
Christi Area, South Texas. Bulletin of the American Association of Petroleum Geologists, Volume 17, pp. 907-962.
1954.

Dynamic Environments: Reconnaissance Mapping, Geologic and

TransGeomorphic, of Continental Shelf of Gulf of Mexico.
actions of the Gulf Coast Association of Geological Societies, Volume 4, pp. 75-107.

1958.

Sedimentology and Quaternary Geomorphology of South Texas.

Transactions of the Gulf Coast Association of Geological Societies, Volume 8, pp. 41-75.

1968.

Barriers - Beaches and Islands.

In, R. W. Fairbridge (ed.),
The Encyclopedia of Geomorphology, III.
New York: Reinhold
Book Corp., pp. 51-53.

Purdy, B. A.
1973.

The Senator Edwards Chipped Stone Workshop Site (Mr 122),

Marion County, Florida: A Preliminary Report of Investigations.
Paper presented at the 38th annual meeting of the SoSan Francisco.
ciety for American Archaeology.

Pyle, T. E., W. R. Bryant, and J. W. Antoine

1974.
Structural Framework of the West Florida Continental Shelf
Proceedings of
and Recommendations for Further Research.
Marine Implications of Offshore Drilling, Eastern Gulf of
Mexico: 1974.
University of Southern Florida and Florida
Department of Natural Resources, pp. 295-300.

Quimby, G.
1956.

I.

Jr.

The Locus of the Natchez Pelvis Find.

Volume 22, No. 1, pp. 77-79.

357

American Antiquity,

Rehkemper, L. J.
Sedimentology of Holocene Estuarine Deposits, Galveston Bay.
1969.
In, R. R. Lankford and J. J. W. Rofers (eds.), Holocene Geology of the Galveston Bay Area.
Houston:
Houston Geological
Society, pp. 12-52.
Rezak, R. and W. R. Bryant
1973.
West Flower Gardens Bank.
Transactions of the Gulf Coast
Association of Geological Societies, Volume 23, pp. 377-382.

Rolingson, M. A. and D. W. Schwartz

Late Paleo-Indian and Early Archaic Manifestations in Western
1966.
Kentucky.
Lexington: University of Kentucky Press.
Royal, W. and E. Clark
1960.
Natural Preservation of Human Brain, Warm Mineral Springs,
Florida. American Antiquity, Volume 26, No. 2, pp. 285287.

Ruppe, Reynold.
Personal communication.

Rusnak, G. A.
Sediments of Laguna Madre, Texas.
1960.
In Recent Sediments,
Northwest Gulf of Mexico, 1951-1958. American Association
of Petroleum Geologists, pp. 153-196.

Russell, R. J.
Plains and Sea Coasts.
1967.
Berkeley and Los Angeles:
of California Press.
173 p.

University

Saucier, Roger T.
1963.
Recent Geomorphic History of the Lake Pontchartrain Basin,
Baton
Louisiana.
Coastal Studies Institute Series, No. 9.
Rouge:
Louisiana State University.
1967.

Geological Investigations of the Boeuf-Tensas Basin, Lower

Mississippi Valley. U. S. Army Waterways Experiment Station,
Technical Report No. 3-757.

1974.

Quaternary Geology of the Lower Mississippi Valley.

sas Archeoiogical Survey, Research Series No. 6.

1977.

The Northern Gulf Coast during the Farmdalian Substage: A

Search for Evidence.
Environmental Effects Laboratory, U. S.
Army Engineer Waterways Experiment Station, Miscellaneous
Paper No. Y-77-1.
39 p.

Schnable, J.
1968.

Arkan-

E. and H. G. Goodell
Pleistocene Recent Stratigraphy, Evolution, and Development
of the Apalachicola Coast, Florida.
Geological Societies of
America Special Paper No. 112.
72 p.

Scholl, D. W.
Recent Sedimentary Record in Mangrove Swamps and Rise in Sea
1964.
Level Over the Southwestern Coast of Florida. Marine Geology,
Volume 1, pp. 334-466; Volume 2, pp. 343-364.
In two parts.

358

Scholl, D. W. and M. Stuiver

1967.
Recent Submergence of Southern Florida: A Comparison with
Adjacent Coasts and Other Eustatic Data. Bulletin of the
Geologic Society of America, Volume 78, pp. 437-454.

Scholl, D. W., F. C. Craighead, and M. Stuiver.

Florida Submergence Curve Revised: Its Relation to Coastal
1969.
Sedimentation Rates.
Science, Volume 163, pp. 562-564.

Sellards,
1916.

E.

H.

Human Remains and Associated Fossils from the Pleistocene

of Florida.
Florida Geological Survey, Eighth Annual Report, pp. 121-160.

1917.

On the Association of Human Remains and Extinct Vertebrates

at Vero, Florida.
Journal of Geology, Volume 25, pp. 4-24.

1919.

The Geology and Mineral Resources of Bexar County.

University of Texas Bulletin, 1932.

1940.

Pleistocene Artifacts and Associated Fossils from Bee County,

Texas.
Bulletin of the Geological Society of America, Volume
51, pp. 1627-1658.

1952.

Early Man in America: A Study in Prehistory,

versity of Texas Press.

Austin:

Austin:

Uni-

Shepard, F. P. and D. G. Moore

1955.
Central Texas Coast Sedimentation: Characteristics of Sedimentary Environment, Recent History, and Deagenesis. Bulletin of the American Association of Petroleum Geologists,
Volume 39, Part 2, No. 8, pp. 1463-1493.
Shepard, F. P. and H. R. Wanless
1971.
Our Changing Coastalines,

New York:

MacGraw-Hill.

563

p.

Shier, D. E.
1969.
Vermetid reefs and Coastal Development in the Ten Thousand
Islands, Southwest Florida.
Geological Society of America
485-508.
Bulletin, Volume 80, p.

Shumway, G.
G. B. Dowling, G. Slasman, and R. H. Payne
1962.
Submerged Forest of Mid-Wisconsin Age on the Continental
Geological Society of AmerShelf off Panama City, Florida.
ica, Special Paper No. 68, p. 271.
,

Smith, F. G. W.
1971.
Atlantic Coral Reefs.

Coral Gables

University of Miami,

Smith, W. G.
1968.
Sedimentary Environments and Environmental Change in the
Peat-Forming Area of South Florida. Ph.D. dissertation on
254 p.
file at Pennsylvania State University.
Snow, C. E.
1945.

The Skeletal Remains.

In, The Tchefuncte Culture, An Early
Occupation of the Lower Mississippi Valley. Memoir of the
Society for American Archaeology, No. 2.
359

Snowden, J. 0. and R. R. Priddy

Loess Investigations in Mississippi. Mississippi Geologic,
1968.
Economic and Topographic Survey, Bulletin III.

Spackman, W.
D. W. Scholl, and W. H. Taft
Field Guidebook to Environments of Coal Formation in Southern
1964.
Florida.
Preconvention field trip guide of the Geological
Society of America.
,

Stewart, T. D.
Antiquity of Man in America Demonstrated by the Fluorine
1951.
Science, Volume 113, pp. 391-392.
Test.
Suhm, D. A. and E. B. Jelks
Handbook of Texas Archeology: Type Descriptions. Texas
1962.
Archeological Society, Special Publication No. 1, and
Texas Memorial Museum Bulletin No. 4.

Swift, D. J. P., D. J. Stanley, and J. R. Curray

Relict Sediments on Continental Shelves A Reconsidera1971.
tion.
Journal of Geology, Volume 79, pp. 322-346.

Tanner, W. F.
Florida Coastal Classification. Transactions of the Gulf
1960.
Coast Association of Geological Societies, Volume 10, pp.
259-266.
Thorn,

B.

G.

1971.

The Dilemma of High Interstadial Sea Levels during the Last

Glaciation. Progress in Geography, Volume 5, pp. 170-246.

Trickey, E. B. and N. H, Holmes, Jr.

1971.
A Chronological Framework for the Mobile Bay Region, Revised, 1970.
Journal of Alabama Archeology, Volume 17,
No.

2.

Tunnell, J. W. and B. D. Causey

Vertebrate Pleistocene Fossils from the Continental Shelf,
1969.
Northwestern Gulf of Mexico. Texas A & M University
Studies, Volume 2, pp. 193-203.

van Lopik,
1955.

J.

R.

Recent Geology and Geomorphic History of Central Coastal

Louisiana.
Coastal Studies Institute, Technical Report
No. 7.
Baton Rouge: Louisiana State University.
89 p.

Veatch, A. C.
The Five Islands.
1899.
In, G. D. Harris and A. C. Veatch (eds.),
Special
A Preliminary Report on the Geology of Louisiana.
Report No. 3, pp. 213-262.

Vergnaud Grazzini, C.
18o Changes in Foraminifera Carbonates during the Last 10
1975.
Science, Volume 190, pp.
Years in the Mediterranean Sea.
272-274.

360

Watts, W. A.
A Late Quaternary Record of Vegetation from Lake Annie,
1975.
South-Central Florida. Geology, Volume 3, pp. 344-347.
Webb, C. H.
Evidence of Pre-Pottery Cultures in Louisiana.
1948.
Antiquity, Volume 9, pp. 386-394.

American

Weigel, R. D.
Fossil Vertebrates of Vero, Florida. Tallahassee:
1962.
Geological Survey, Special Publication No. 10.

Florida

Willey, G. R.
Archeology of the Florida Gulf Coast.
1949.
Smithsonian Miscellaneous Collections, Volume 113. 599 p., 60 pis.

An Introduction to American Archeology, Volume I - North

and Middle America.
Englewood Cliffs: Prentice Hall.
526 p.

1966.

Williams,
1965.

S.

and J. B. Stoltman
An Outline of Southeastern United States Prehistory, with
In, H. E.
Particular Emphasis on the Paleo-Indian Era.
Wright and D. G. Frey (eds.), The Quaternary of the United
States.
Princeton: Princeton University Press, pp. 669684.

Wilkinson, J. H. McGowan, and C. N. Lewis

1975.
Geologic Notes.
In, Bulletin of the Geological Society of
America, Volume 59, No. 2 (February), pp. 349-350.
Wilson, M. T.
1895.
On the Presence of Fluorine as a Test for the Fossilization
of Animal Bones.
American Naturalist, Volume 29, pp. 301719-725.
439-456,
and
317,

Winchester, P. D.
Geology of the Freeport Rocks, Offshore Texas. Transactions
1971.
of the Gulf Coast Association of Geological Societies, Volume
21, pp. 211-222.
Wolman, M. G. and L. B. Leopold
1957.
River Flood Plains:
Some Observations on Their Formation.
U. S, Geological Survey, Professional Paper No. 282-C.

Wormington, H. M.
Ancient Man in North America. Denver Museum of Natural
1957.
History, Popular Series No. 4.

361